The rise of inverter-interfaced generation along with the displacement of conventional generation leads to decreasing synchronous inertia. Assuming that the size of contingencies remains the same, lower values of inertia lead to higher absolute values of Rate of Change of Frequency (RoCoF) after a disturbance. In case that the absolute value of RoCoF is too high, the frequency containment reserves might not be adequate to arrest frequency before it reaches the Under Frequency Load Shedding (UFLS) threshold. Therefore, frequency response assessment becomes a necessary task for the System Operators (SOs). The most important part of this assessment is the estimation of the maximum instantaneous frequency deviation or frequency nadir, since this will define if the UFLS threshold will be violated.
Several studies have examined the development of dynamic equivalent models, that reproduce the actual system's response after a disturbance. These studies show that simplified equivalent models are sufficient to represent accurately the frequency response of an actual power system.
This paper contributes with a method that identifies the parameters of a simplified frequency response model and assesses the frequency response of a power system. Therefore, the method is divided into two parts. The first part regards the parameter identification. First, the available data should be processed. Possible noise/oscillations are filtered out, the times of the onsets of the available events are identified, and the selected region for the identification is specified. Then, the dynamic model that constitutes the frequency response model for this study is selected. In the last step of the parameter identification, the optimization algorithm is applied. This requires the definition of the initial values and the limits of the parameters, as well as the selection of the stopping criteria of the algorithm. The identification is achieved by minimizing the error residuals between the simulation model's and the actual system's measured frequency. The employed frequency measurements are obtained from historic underfrequency events from the Nordic power system. However, the method can be applied to any power system.
The second part of the method includes the frequency response assessment. After the parameters are identified, the selected frequency response model can be simulated to estimate the frequency nadir of events that have not been included in the identification. Since the parameters are identified based on actual events, the simulation results are considered to constitute a realistic representation of the frequency response of the respective power system.
Therefore, the paper contributes with a complete solution, regarding the identification of frequency response model parameters and the frequency response assessment. The results show that the proposed method provides accurate parameter identification, as well as accurate frequency response assessment.
The content in this paper is part of the European Union’s Horizon 2020 Research and Innovation Programme, TotalControl Deliverable no. D4.1, This paper is concerned with grid frequency response by changing the active power output of the wind plant, and aims to provide a simulation framework which can be used to investigate and test wind plant control algorithms designed to help grid stability. This is achieved by coupling together a wind power plant simulator capable of time-domain simulations of wind farm performance including the control systems (LongSim), to a grid simulator (KERMIT).
LongSim is an engineering tool developed by DNV GL for wind farm analysis, both steady-state and dynamic, aimed at designing and testing of wind turbine and wind farm controllers. It is designed to run rapidly, to allow fast design iterations and long simulations. Turbine wakes can be represented by a range of engineering models of wake deficits and added turbulence. The wakes are embedded into a realistic ambient flowfield. The model was described in detail in TotalControl deliverable D1.9.
KERMIT (KEMA Equilibrium Renewable Integration Tool) is a proprietary DNV GL tool that is used to simulate the frequency behaviour of a (future) power system and its interconnections, inertia and the operation of primary and secondary regulation. KERMIT allows analysis of dynamic grid performance in future scenarios or during events such as generator trips, sudden load rejection, and variable renewable resource (wind, solar) ramping events. The software runs on the MATLAB/Simulink platform and incorporates inertial, governor and regulation response, balancing market logic and control of new technologies such as energy storage.
Combining LongSim and KERMIT results in a closed-loop simulation environment, where the wind plant responds to grid frequency by changing its power output, and the grid responds to this power output resulting in changes to the frequency which are detected by the wind plant. In the closed-loop simulation environment fast frequency response (FFR) strategies are implemented in the wind turbine controllers and tuned to achieve the best performance in terms of helping to stabilise the grid frequency, while remaining within the operational constraints of the wind turbines in the wind conditions available at the time.
In this paper a case study based on historical data for the Irish grid system is performed. Ireland is chosen as it already has a high wind power penetration, and suitable data is publicly available. The work has demonstrated that the combined model is a useful tool for tuning and testing FFR strategies integrated into the wind turbine controller, and example closed-loop simulations have shown that with appropriate tuning of these control strategies, the wind farms can have a significant beneficial effect on the frequency stability of the grid. The strategies have been tested against three real frequency dip events recorded on the Irish system.
The growing penetration of wind energy into the power system has led to the elaboration of grid codes to maintain system security and reliability. The process to assess each requirement is not only a license obligation for transmission operators but is vital for wind farm developers to determine the wind farm electrical design. As wind farms are now required to provide voltage support to the grid with the utilization of converter based technologies, the design procedure is becoming more complex.
One of the major issues is the design of reactive power compensation systems using flexible AC transmission system (FACTS) such as STATCOM, SVC and passive filters. The capacity of the FACTS device must be determined to fulfill the reactive power requirement on wind farms, while the harmonic filter may be installed together to meet the harmonic requirement. However, the harmonics from the FACTS device influence the filter parameters, while the reactive power from the filter also influence the FACTS device capacity. The interaction between harmonics and reactive power must be considered adequately to tune the parameters of the FACTS device and filter.
In this paper, an efficient procedure to determine the reactive power capacity for STATCOM and SVC along with the filter parameters to meet grid code requirements is described. Harmonic requirement, reactive power requirement and voltage regulation at the point of common coupling are considered. In addition, voltage regulation for the protection of wind turbine generators (WTGs) is taken into account, to set the reactive power limit of the generators.
The process is determined from the causal relation of each parameter, depending on the control principle of the FACTS device. For STATCOM, the harmonic current depends on the switching frequency of pulse width modulation control, thus the harmonic analysis to determine the filter parameters should be carried out first. On the other hand, the harmonics for SVC depend on the reactive power output of the thyristor controlled reactor (TCR). Therefore, assessment of reactive power must be carried out prior to the harmonic analysis. Moreover, the LC filter which composes the SVC must be tuned to suppress the characteristic harmonics of the TCR and the harmonic resonance within the network. For both STATCOM and SVC, the harmonic current from the WTGs is also considered and the shifting of the resonant frequency due to the filter installation is discussed.
Finally, the developed procedure is verified for a model wind farm of 32 wind turbine generators. Results of the FACTS device capacity and filter parameters are presented, validating the effectiveness of the proposed procedure in meeting multiple requirements.
Covid-19 has changed the way of working in Japan. More than the half of people working in Tokyo are doing remote working using digital technology such as video communication tools. If people life is back to before Covid-19 level, the second and third wave of Covid-19 will occur and Covid-19 will become a protracted case. In addition, energy consumption and carbon emission will rebound to before Covid-19 level and make climate change solutions difficult. To avoid 3Cs, closed spaces with poor ventilation, crowded places with many people nearby and closed-contact setting such as close-range conversations, we need further change of working way and in a future, people will live in a place which they want to live in and work remotely thanks to the development of digital technology.
In this research, we have developed a detailed technology and GIS hybrid energy model and done quantitative analysis on the benefit to promote the changes of working way further. To identify the impacts of avoid 3Cs on carbon abatement, we have done two simulations. One is with Covid-19, which assume further changes of working way and another is without Covid-19 which assume that working way is back to before Covid-19.
To reduce 80% carbon mitigation target by 2050, electricity price of with Covid-19 scenario is 23.2% cheaper than without Civid-19 scenario and with Covid-19 scenario needs less development of further energy infrastructure such as electricity grid and grid storage.
One of big challenges of Japan to meet 80% carbon mitigation target in 2050 is that huge geographical un-matching between energy consumption and renewable energy potential. Further changing way of work will make the improvement of the geographical un-matching and lower the cost of carbon mitigation at more acceptable one.
Grid-connected converters are known to experience small-signal stability issues when the grid impedance is high. Examples of sustained or undamped oscillations have been reported in traction systems, HVDC-connected systems and wind power plants (WPPs). Small-signal stability is an important topic in offshore WPPs due to the increasing distance of the wind turbines from the point of connection to the transmission system. Modeling and validating the small-signal behavior of grid-connected converters has been an area of focus for both academia and industry.
A number of analytical methods have been used to model the small-signal response of grid-connected converters. These methods provide insight into how control structures, parameter tunings and grid components influence stability and harmonic resonance. The limitations of analytical approaches are the high modeling complexity and the assumptions embedded in the modeling process. Other methods use simulation models and impedance scans to measure the small-signal impedance of a converter. These methods, whilst not providing detailed insight into the converter control structure, are user-friendly and can incorporate nonlinear control and switching elements used by the actual converter.
A Feed Forward controller approach is developed in order to inject perturbation voltage signal superimpose with nominal voltage for a medium voltage grid simulator to evaluate the wind turbine response and scan the wind turbine system impedance. This opens the possibility for validating high-power converter control and power hardware against simulation and/or analytical models.
In this paper, the results for impedance scans extracted experimentally from a SG DD-167 wind turbine manufactured by Siemens Gamesa Renewable Energy (SGRE) connected at Fraunhofer-IWES test facility are presented. The measured impedance is validated against the impedance obtained from a corresponding PSCAD model of a same system. The two impedance scans closely align over the measured frequency range, highlighting the efficacy of experimental method.
The Commission Regulation (EU) 2016/631 of the 14th of April 2016 establishing a network code on requirements for grid connection of generators introduces a framework for building a structural harmonization of technical stipulations across Europe. Network operators must implement this regulation in national grid codes or federal legislation.
This regulation requires new power generating modules (PGM) connected to the system to satisfy technical requirements throughout its life. Therefore, provisions on compliance testing and simulation are introduced during the PGM commissioning stage.
This paper describes the PGM certification structure related to this regulation in Spain in the form of technical standards (NTS) for the conformity assessment. NTS is a guideline for submitting the corresponding PGM conformity certificates and the final version is planned by the end of 2020.
RMS dynamic simulation is the primary tool for stability studies carried out for a large power system. To perform planning and operation stability studies, commercial positive-sequence RMS dynamic programs are used. The increasing penetration of wind generation has created new challenges in modeling and simulation. An RMS model of a large power system with many wind power plants (WPPs) should reflect the dynamics of individual WPPs, the mutual impacts of the WPPs and other facilities in the system and the dynamics of the power system as a whole.
Unlike synchronous generators of a conventional power plant, all wind turbine generators (WTGs) of a large WPP cannot be modeled explicitly and at least some need to be aggregated. The multiplicity of WTGs is another specific feature of wind generation. Despite the complexity of WTG/WPP dynamics, at present there exists a trend towards simplification of WTG/WPP models: more detailed representations of the collector are displaced by a single WTG equivalent and more accurate vendor-specific WTG dynamic models are gradually supplanted by their simplified versions and generic models. The rationale behind this trend is to limit the growth of system-wide models and reduce the computational time required by batch contingency simulations.
Given the trend towards simplification, it is difficult to expect that the same system-wide model can be equally suited for stability studies of different types and for all contingency scenarios assumed by a particular study. Presumably, the accuracy of a system-wide model with respect to different areas and even portions of the system may differ significantly. As the level of wind penetration grows, the determination of whether and how a system model needs to be modified becomes more complicated, which defines many simulation challenges.
The intent of this paper is to outline interdependent simulation challenges directly or indirectly caused by the simplification of WPP representation in both power flow and dynamics. Among these challenges, there are: the choice of level of collector aggregation and loading for WPP clusters and for remote WPPs; defining operating conditions of system components electrically close to WPPs; and customization of system-wide models to simulation of contingency scenarios assumed by a particular study.
The paper emphasizes the need for holistic approaches to model development and contingency simulations. An approach where the collector aggregation and loading of WPPs in power flow and the models of WPPs and nearby power system components used in dynamics are consistently coordinated with contingency scenarios would help ensure the representativeness of simulation results. As more experience is gained with this approach, industry requirements for both model development and actual studies may need to be modified accordingly.
As the wind penetration in power systems increases, more wind power plants (WPPs) and HVDC transmissions with line-commutated converters (LCCs) are located electrically close and may interact. Commutation failures, to which LCC-HVDC transmissions are prone, affect the dynamics of nearby WPPs as well. However, these mutual impacts are not reflected in RMS simulations of power systems.
To account for these impacts, dynamic models of the LCC-HVDC transmissions need to reflect the commutations of the thyristor valves of the LCC converters. A three-phase six-pulse bridge converter (the Graetz bridge) is a core element of LCC-HVDC main circuits. Usually, waveform models of the bridge are thought to be too computationally demanding and therefore not applicable to RMS power system simulations.
This paper discusses simplified waveform models of the Graetz bridge that reflect the valve commutations and are suitable for multirate simulations with commercial positive-sequence RMS dynamic programs. The main intent is to propose compact and computationally efficient mathematical models of the bridge and substantiate the choice of the one most appropriate for multirate simulations. The paper considers continuous-time (CT) and discrete-time (DT) models of the Graetz bridge with emphasis on their applicability to different LCC-HVDC topologies, interface with the main RMS program and possible numerical issues.
The CT models are three-phase instant-value tools based on differential equations. Each thyristor valve is represented by a circuit with parameters that are varied depending on the valve’s state. Such models reproduce the waveforms of DC/AC voltages and currents for any combination of conducting valves. The size of the internal time step is assumed to be about 50 microseconds. The DT models are single-phase discrete-value tools based on difference equations of the bridge. The paper discusses a model based on the equation that defines a recurrence relation between the initial values of the DC-side current for two consecutive repetition intervals. For the steady-state operation, the internal time step is one sixth of the fundamental frequency cycle. In the case of normal commutations, the calculated values of the DC current match those calculated by CT models at the corresponding time instances.
A CT model where the valve is represented by a variable resistor in parallel with a resistive-capacitive branch with constant parameters is found most advantageous. With this model, there is no need to solve any nonlinear algebraic system at each numerical integration time step by iteration. The bridge equations are written in a compact form that makes the model a competitive multirate simulation tool. Assuming its further elaboration, especially in terms of the interface with the main simulator, this model can become a building block for LCC-HVDC transmission models intended for multirate RMS simulations of power systems with WPPs.
Currently, wind farms are operated with the primary control objective to maximise power generation. As the penetration of wind energy rises, this strategy may no longer be feasible, especially if government subsidies are reduced. In such a scenario, wind farm flow control (WFFC) can offer wind farm operators additional flexibility to maximise profits. For example, if electricity prices fall operators could curtail power generation in favour of load reduction strategies. To date, the lack of convincing evidence for the economic case for WFFC has prevented its widespread adoption. This paper addresses this by introducing a set of showcases that researchers within the WFFC community can use to assess the positive impact of their control strategies. These showcases are based on the TotalControl Reference Wind Power Plant with weather simulation data and estimated electricity prices for both 2020 and 2030, provided courtesy of the DTU Balancing Tool Chain. With this data, researchers can evaluate the performance of their control algorithms using an internationally recognised tool.
Wind and solar generation resources do no provide synchronous inertia and, furthermore, being near zero marginal cost resource, this generation displaces more expensive conventional synchronous generation in generating unit commitment. The expectation is that eventually this will result in decline of overall system inertia, especially during periods of low load and high wind and/or solar production. Seeing rapid increase in installed and planned capacity from wind generation resources, ERCOT in 2013 started offline analysis wind generation impacts on total system inertia. In 2015 online monitoring tool was put in place in the control room to monitor system inertia in real time and forecast expected inertia for the upcoming hours. Inertia monitoring is based on telemetered status (on/off) and corresponding inertia contributions from each synchronous generator on the system. It should be pointed out that vast majority of synchronous generation as well as all wind generation in ERCOT is transmission connected and ERCOT receives telemetry of various data points from each transmission connected generator every 4 seconds.
ERCOT also determined critical inertia level, i.e. inertia level below which, after the largest generation contingency, existing frequency response measures are not sufficiently fast to arrest system frequency before it reaches first step of underfrequency load shedding. ERCOT is monitoring system inertia against critical inertia value and developed a process by which additional synchronous generation can be brought online if system inertia starts approaching the critical level.
This paper will present analysis of 7+ years of inertia data. Intuitive expectation is for inertia to decline in proportion to added wind or solar generation however this trend was not confirmed in ERCOT. There is a number of factors affecting inertia apart from the addition of wind and solar generation. In ERCOT system these are: rapid load growth, lower natural gas prices and coal generation retirements resulting in more combined cycle generation commitments, significant inertia contribution from private use synchronous generation at the industrial sites, overall windiness of each year under analysis. The paper will analyze and present impact from each of these factors and present the conclusions.
Energinet, the Danish Transmission System Operator (TSO), has been planning for the past decades the integration of renewable energy into power systems to meet CO2 reduction requirements in Europe. As a part of strategic goals in Energinet, the penetration of renewable energy will be increased to more than 70% by 2030 and the long-term goal is to achieve 100% renewable energy supply by 2050. Therefore, a smooth transition of the energy system from conventional power plants to inverter-based systems is a very important topic in order to prepare for the future power system. One of the most important challenges, which should be highlighted, is to investigate how short circuit contribution will be in the future when the systems are mostly changed to inverter-based systems. Correct representation of short circuit current contribution from inverter-based systems would have significant influence on systems in the future. Therefore, based on upcoming high penetration of inverter-based systems, it is highly required to update the model database used in Energinet for the short circuit report, which is regularly issued for power system design and operation purposes e.g. protection relay settings and requirements of high voltage components. Furthermore, the short circuit levels play an important role in the design of renewable energy systems such as wind farms and photovoltaic power stations. Conventional power plants are able to provide a large amount of fault current to the grid under faults. In contrast, converter based systems are able to provide only max. 2 [p.u] of their current rating as fault current, which are highly relying on their hardware capabilities. To meet grid code requirements from TSO, a variety of control and protection strategies are developed by manufacturer and used to Voltage source converters (VSC) for different operating points and grid configurations. This enables VSCs to provide diverse characteristic of fault current. Therefore, it is of great importance to take correct representation of inverter-based systems into account for system level short circuit calculations instead of using simple representations of converters via conventional methods like IEC 60909. In this paper, Type 4 wind turbines and VSC-HVDCs are considered for characterization of short circuit currents because these types are main source of contributions from converters. Selected EMT vendor models are therefore tested in EMT simulation in order to characterize test objects. The characteristics measured via EMT models are used to parameterize settings of the static short circuit calculation tools in static models of DigSilent PowerFactory, which is the main tool used for system level short circuit calculations in Energinet. Furthermore, characteristics measured by both tools are compared to discuss limitations of tools and models, and recommended settings for system level short circuit current calculation in PowerFactory will also be discussed.
For effective climate change mitigation, the share of renewable energy is significantly increasing. Their intermittent nature creates undesired curtailment need to accommodate them on grid for periods withproduction exceeding the demand.
Consequently, instead of directly being curtailed, some kind of energy storage is needed to balance power supply and demand. Promising storage methodologies has compressed hydrogen through electrolysis a clearly uprising path. This work tackles the on-shore wind farms, due to the frequent potentialhigher wind speeds range need to be dealt with. Also, for low-emission carbon neutral aviation, syntheticfuels from renewable energy sources are assumed to be increasingly used in the future.
The X-to-Liquid (XtL) routes proved to offer promising solutions in the chemical energy storage area, and this work considers the Power-Biomass-to-Liquid (PBtL) process as an energy storage solution. Yet, publications concerning the PBtL process discuss only the steady-state mode operation mode. However, since renewable energy is of a fluctuating nature, the dynamic process operation is of a particular interest.
This thesis compares the steady-state and dynamic operation mode for a PBtL process. The plant sizeis scaled to an 24.8 MW electrolyzer power input. These process operation modes simulated on AspenPlus and Aspen Dynamics. Moreover, they are economically assessed with the DLR in-house program Techno-Economic Process Evaluation Tool (TEPET).
The dynamic mode results in a net production cost (NPC) of 5.06 €/l, while the steady-state resultsin 3.84 €/l. The fixed capital investments (FCI) of the dynamic mode amounted to 2.45 times that of thesteady-state. The Fischer-Tropsch (FT) products per annum in the dynamic mode is 92% of that of thesteady-state mode.
In further investigations of the dynamic mode; the most economic setting for the process was achievedby utilizing compressed hydrogen storage tanks with the maximum capacity it can offer and discharging down to a minimum pressure inside the tanks of 5 bars, with a second compressor after the tanks.
To ensure the frequency stability of the Continental European power systems, the first real time lever currently implemented is the ‘Frequency Containment Reserve’ (FCR). So-called dispatchable power plants mainly provide this service so far. In some European countries, wind farms already contribute to the frequency stability by providing ‘Frequency Restoration Reserve’, either automatic (aFRR) or manual (mFRR), as well as ‘Replacement Reserve’ (RR), especially when asymmetrical (downward) reserve provision is allowed in the ancillary services market. To our knowledge, FCR is not yet provided by wind turbines in practice, and pilot projects and/or related literature are lacking.
This paper addresses how FCR can be procured from wind generation, by explaining the approach applied for reserve allocation on a wind farm and by sharing some results of the experimentations being conducted in France. The analyses reveal that the capability of real-time estimation of the wind maximum available active power (AAP) is essential to allow reserve allocation and FCR provision. Furthermore, the knowledge of the estimated AAP makes it possible to monitor the performance of wind-FCR response through the assessment of the actual FCR gain. The experimental results obtained show that wind farms can technically contribute to FCR when being equipped with a dedicated AAP estimator, and that the actual wind FCR performance is globally compliant with the current TSO’s requirements in terms of dynamics and gain.
The work presented in this paper is part of the French demonstration of the ‘EU-SysFlex’ European project.
Keywords - Frequency Containment Reserve (FCR), Wind Power Plants (WPP), Available Active Power (AAP), wind reserves, FCR performance control, FCR gain, balancing by wind
The introduction of inverters with grid forming capabilities poses new challenges for modelling active distribution grids for power system stability studies. An aggregated equivalent distribution grid model connected to a detailed transmission grid model allows comprehensive studies on the stability of the power system. Previous work shows that grey-box models provide accurate aggregated representations of distribution grids dominated by inverter-based generation in grid feeding control. The grey-box approach is capable of differentiate between voltage level, generation technology and control strategy of the generators. However, all components of one voltage level are connected to the same busbar resulting in a negligence of different voltages at the point of common coupling.
This paper shows the challenges and limitations of the grey-box approach when grid forming inverters are considered. A short-circuit fault is simulated on a test grid dominated by inverter-based generation, while inverters are modelled with grid feeding as well as grid forming controls. The active and reactive power flow from the transmission grid to the distribution grid is compared between the detailed grid and the equivalent grid. For grid feeding inverters the equivalent grid is capable of capturing the dynamic behaviour of the inverters in comparison to the detailed grid. The implementation of grid forming inverters, however, leads to differences between detailed and equivalent grid, since the response is highly dependent on the effective impedance between the point of common coupling of the inverters and the fault location. In the equivalent grid, the inverters of one voltage level are connected to the same busbar resulting in significantly different power flow compared with the detailed grid.
The transmission grid of Denmark-West is part of the Continental European synchronous area with HVAC connection to Germany and multiple HVDC interconnectors to Norway, Sweden, Denmark-East (part of the Nordic system), The Netherlands and, in few years, to Great Britain. Denmark-West has become a hub where the meshed HVAC grid includes numerous converter stations of the HVDC interconnectors, large offshore wind and photovoltaic power plants. The converter-interfaced transmission and production are necessary steps in the green transition of the energy system and require expansion of the meshed 400 kV grid to support the transition. With strong public opposition to new overhead lines (OHL), there is high demand for the application of underground cables in the Danish transmission grid. Underground cables give rise to a range of technical challenges, one of which is the amplification of harmonic distortion known from HVAC connections to offshore wind power plants and, more recently,experience in the meshed 400 kV grid in Denmark. The first challenging experience has arrived due to the commissioning of the Vejle-Ådal 400 kV cable system, which is the two parallel approx. 7 km long cables, in combination with an approx. 40 km OHL. After the commissioning, a significant amplification of the 11th and 13th harmonic voltage has been measured in the other two, distant from the Vejle-Ådal, substations of the 400 kV grid. The dampening of the harmonic amplification is achieved by the filters located in a third (different) substation. The harmonic amplification has been linked to the HVDC converter stations. The strange (with regard to locations) propagation and dampening of the harmonic voltages relates to the factum that Denmark-West is a meshed grid with multiple harmonic emission sources. This work presents the method and model development of the transmission grid of Denmark-West for the harmonic assessment. The model range is up to the 50th harmonic order. This work demonstrates that both harmonic impedance (passive-part), harmonic emission sources (active-part) and feedback are relevant for accurate and suffice harmonic assessment of the meshed transmission grid with multiple harmonic emission sources. The harmonic emission sources are modelled by the vendor-specific, non-linear, harmonic emission magnitudes and, as a novice of the method, by the empirically defined phase angles. This method is applied in replacement of standardized superposition rules which failed working in the meshed grid with multiple harmonic emission sources. The presented method and model are confirmed by the validation cases by the measured harmonic voltage distortion.
The last two decades have been witness of a transformation on how electrical energy is supplied. Although it is imperative to supply environmentally friendly electricity, existing techno-economical obstacles must be overcome. This work presents an overview of the NSON II project investigating the required advancements to enable an effective deployment of further 30 GW of offshore wind to be installed in the German North Sea exclusive economic zone (EEZ) according to current grid development plans by 2035. To reach this target, HVDC transmission systems are the most suitable solution for the connection of offshore wind farms (OWF).
Goal of the works at hand is boosting the proposal of new alternatives for the connection of OWF. Voltage source converters (VSC) are currently the standardized technology for the connection of OWF using HVDC links. Nevertheless, with the objective of reducing the required investment costs in the short term, alternative technologies such as line-commutated converter and diode-rectifier units in parallel to VSC-based links have been proposed. Besides that, the parallel connection and operation of existing and new OWF seems to be an effective way to optimize investments and increase operational flexibility.
The offshore generation, connection and transmission capacities open up the possibility of creating an interconnected HVDC grid. This HVDC grid structure introduces new opportunities for electricity trade, whereas questions on the stability of the interconnected AC grid have to be answered. To turn the plan of an HVDC-interconnected European grid into a reality, the reduction of the current costs, the development and deployment of new technologies as well as an optimized planning and operation of OWF connection systems is mandatory.
However, the control of the wind farms, as well as the control of such HVDC topologies are still a topic of research. Aside from the parallel connection, with an interconnected HVDC grid emerging in the North Sea, also the power flow controls of offshore AC connections and grids dominated by multiple HVDC converters are crucial. Therefore, it is imperative to correctly plan short and medium-term goals, so that long term goals can be technically feasible, while at the same time investments are optimized.
The following fields of research have been identified as next steps to be conducted on the path to an economical connection and international integration of offshore wind energy in the German EEZ:
The above-mentioned steps will be detailed in the proposed paper.
This paper shows the potential of wind energy
plants to deliver second step inertia and primary control
reserve. Different control approaches for droop-control are
compared. A complete small-signal analysis of the combined
control of the machine-side and the grid-side converter proves
the stability of the suggested option.
In course of the German energy transition, congestion management measures like redispatch, feed-in management and the dispatch of reserve power plants have increased both in frequency and in volume. In this paper, an empirical cause analysis is conducted to shed some light on the correlation between renewable energy generation, the transnational electricity markets and these measures. For this purpose, a methodology is developed to synthesize a single time series of curtailed energy from heterogeneous data published by German DSOs. The analysis shows a strong correlation between both wind generation in Germany as well as the scheduled trade with Austria. Furthermore, four distinctive systems states were identified using a k-means++ cluster algorithm and described in detail, allowing further insights into those coherences and their future development.
As electricity feed-in is shifting from transmission grids towards distribution grids, future grid management strategies will have to include greater contributions to system stability coming from distributed reactive power sources. In this study, grid data from a segment of the German transmission network and an underlying distribution grid is examined. By identifying the reactive power sources, the potential of each technology for an extended usage of their reactive power capabilities for grid management is assessed. Moreover, the dynamic behavior of different reactive power sources is investigated. For this purpose, measurement values of set-point changes of various reactive power sources, guideline regulations and theoretical considerations are compared. The results allow for a comprehensive comparison of the dynamic behavior of these reactive power sources and can contribute to a better utilization of their potential.As electricity feed-in is shifting from transmission grids towards distribution grids, future grid management strategies will have to include greater contributions to system stability coming from distributed reactive power sources. In this study, grid data from a segment of the German transmission network and an underlying distribution grid is examined. By identifying the reactive power sources, the potential of each technology for an extended usage of their reactive power capabilities for grid management is assessed. Moreover, the dynamic behavior of different reactive power sources is investigated. For this purpose, measurement values of set-point changes of various reactive power sources, guideline regulations and theoretical considerations are compared. The results allow for a comprehensive comparison of the dynamic behavior of these reactive power sources and can contribute to a better utilization of their potential.
The multi-frequency (harmonic) interaction among
converters is an increasing challenge in converter-dominated
power grids. To study steady-state harmonics and transient harmonic
stability, it is crucial to develop accurate multi-frequency
models for converters. However, the frequency couplings have not
been addressed in the state-of-the-art analytical models as well
as model validation tests for converters. This paper overviews
the impedance measurement test methods for converters and
identifies the couplings in frequency-domain based on time-domain
simulations using typical Type 4 and Type 3 wind turbine
models. The analysis shows significant frequency and sequence
couplings which are not considered in the simplified analytical
A requirement of the grid operator, e.g. in Germany, is to receive a generic model, which represent the dynamic electrical behavior of the wind turbine and the wind power plant. Such generic models are usually open models built according to international standards. The dynamic behavior of these models can only be modified via parameterization, while the internal structure cannot be changed. Instead, the structure is defined in committees like the International Electrotechnical Commission (IEC) Technical Committee (TC) 88 Working Group (WG) 27 or the Western Electricity Coordinating Council (WECC) Modeling and Validation Work Group (MVWG).
Two separate standards define generic wind turbine models. Models according to the IEC standard 61400-27 are used to represent the dynamic electrical behavior at the wind turbine’s terminals and at the point of common coupling (PCC) of the wind power plant. Models defined by the WECC are designed to conduct bulk system studies. The WECC models are less accurate than the IEC 61400-27 models in terms of steady state and transient behavior. Therefore, the IEC 61400-27 models are used in this paper.
Formerly, the IEC model has mainly been parameterized according to, and compared with onshore turbines in a power range of 2-5 MW. In this paper, the (Low voltage ride through) LVRT validation results of the IEC 61400-27-1 Edition 2 model parameterized according to the Adwen AD8 8MW wind turbine will be presented. The validation procedure according to IEC 61400-27-2 Edition 1 will be used in order to compare the field measurements with the simulation results.
According to the Danish transmission system operator (TSO) - Energinet, 50% of total energy consumption in Denmark in the year 2019 came from renewable energy of which 47% accounted for wind power plants and the balance from solar systems. Solar power’s global share in power generation has reached an all-time high of 2.6% in the year 2019. In Denmark, the introduction of the feed-in-tariff program in the year 2015 saw a surge in the installations of small wind turbines. Germany and the UK had introduced similar feed-in-tariff programs to promote the uptake of small-scale renewable electricity generation in the years 2000 and 2010 respectively. By the end of 2018, 49% of the EU's cumulative photovoltaic (PV) capacity came from rooftop solar (residential 19%, commercial 30%). Furthermore, electricity generation via onshore wind is the cheapest form of new electricity in Denmark.
Power distribution systems with a high share of variable renewable energy sources (VRE) face problems due to the variability, uncertainty, and non-dispatchable nature of VRE. A substantially increased amount of VRE in future distribution networks transforms them into highly weather dependent networks. As a result, distribution system operators (DSO) are more challenged and have to deal with under- and overvoltages, overloading of transformers and feeders, rise in line losses, and increased stress on existing assets (tap-changers, voltage regulators, switching capacitors, etc.). An increased share of weather-dependent generation at a lower voltage level has an impact on the transmission system as well. The current lack of infrastructure for observing, monitoring, and control of rooftop PV and small-scale wind installations is an additional challenge for the DSOs.
This paper provides a review of the relevant aspects of wind and solar power in future VRE dominated distribution system. It expounds upon the problems of future distribution systems confronted with the imminent increase of controllable weather-dependent generation sources but with limited observability. The paper presents a synopsis of past academic research and case studies, concerning a large share of VRE in the distribution system. The overview provides a perspective on the relevant challenges endured by DSOs so far while integrating VRE in the system.
Following from grid-forming (but grid-connected) wind park field tests reported in 2019, a logical extension of the testing is to prove the ability of a 69 MW grid-forming wind park to operate in an islanded condition, and to provide system restoration and black start services. A test programme carried out during August-October 2020 demonstrated, in principle, the ability of the wind turbines and power park to provide such services. Islanded mode operation, and resynchronisation to grid-connected mode are described. Also, it is verified that a small number of grid-forming turbines can support a larger number of grid-following turbines in islanded mode. Finally, black start and system restoration tests are described, during which the park back-energises a “dead” MV and HV network up to 132 or 275 kV, that can subsequently be resynchronised at those voltage levels, without a transient reactive infeed.
The historical success of large scale electrification with AC power systems is closely linked to synchronous generators (SG), as their physical properties are essential for power system stability. In particular, SG intrinsically contribute to system inertia and provide voltage source behavior. Voltage source converters (VSC) that are equipped with a grid-forming control concept can mimic these essential properties, enabling system operation without any SG. Unlike state-of-the-art grid-following VSC control concepts with fast current control, grid-forming converters (GFC) generate a three-phase voltage system so that the resulting currents are free to react to grid voltage changes and, thus, meet the needs of the power system.
However, due to physical restrictions, GFC cannot be considered as fully equivalent to SG. In particular, any VSC can only provide currents up to or, at most, slightly exceeding nominal values. It is particularly important to protect VSC from overcurrents that would damage the power electronics. While this is straightforward for grid-following VSC, current limited GFC lose their voltage source property. Under such circumstances, the dynamic behavior of GFC is no longer equivalent to SG, which can provide, for a limited period of time, a multiple of the respective nominal current.
Large disturbances in the transmission system can cause significant changes of the grid voltage. Such changes may lead to large current amplitudes and can push GFC to their technical limit. Due to the loss of their voltage source property, power system dynamics are affected and instability phenomena may occur. Hence, there are unresolved issues with regard to the role of GFC in interconnected power systems. Immediately after a disturbance, the GFC closest to the disturbance may become overloaded and currents may have to be limited. In the time period afterwards, GFC must be able to leave current limiting mode in order to achieve load sharing with other feed-ins.
Previous studies have already shown that load sharing considering current limitation of GFC is not always guaranteed to be successful and may result in unstable system behavior. This contribution presents further simulative investigations in the EMT domain on the interplay of current-limitation and load-sharing of GFC in interconnected power systems. In the introduction, we present results of fundamental investigations in order to analyze the dynamic behavior of GFC after significant changes of amplitude and phase angle of the grid voltage. Then, the impact of this behavior on possible instability phenomena in the context of large disturbances is presented by means of transmission system simulations. The results of our contribution provide valuable insights into the requirements of transmission systems equipped with high shares of GFC. These insights can be helpful for the development and evaluation of GFC control concepts with regard to large-scale use in interconnected power systems.
In this work, a two-area four-machine system with wind power generation is developed in parallel in both
RMS and EMT environments. The fundamental frequency positive sequence (RMS) model is developed in DIgSILENT
PowerFactory and the electromagnetic transient (EMT) model, in PSCAD. Specifically, this paper focuses on the systematic
development of the comparable converter model in both environments. To model the wind farm, a PSCAD EMT
voltage source converter model is reconfigured and simplified to replicate the control of the default PowerFactory converter
model that is used. The performance of the two models is assessed within a simple system, before integrating the wind
farm into the two-area system. The selection of the control parameters is chosen in both models to provide a comparable
response for a step change to the reference power. The default fault ride through (FRT) control is enabled in PowerFactory
and a FRT control strategy is developed in PSCAD to achieve a realistic response. A three-phase short circuit fault on a parallel
transmission line segment and a 100 MW load step change are independently applied in order to compare the dynamic
response of both models. The specific limitations of converter modelling in RMS simulations are discussed.
The German power supply network offers a very high availability in terms of security of supply. This means that the provision of emergency generators, even for many critical consumers does not pay off and therefore these are not implemented. In addition, German grid operators have only very limited mobile emergency generators and these can continue to supply only a very small part of the power grid in case of disturbances in the e.g. transmission network.
In the course of the transition into green energy and the resulting elimination of conventional power plants available at any time, the risk of major disruptions in Germany may increase. To be more prepared for such events, alternatives for the emergency power supply in the event of a long-term major disruption or even a blackout must be considered. For grid operators, the questions arise: What can state of the art decentralized power plants contribute to increase the range of emergency power supply? Is it possible to form a temporarily islanded microgrid (TIM) easily and quickly with support of decentralized power plants, so that a grid operator is able to react on major disruptions on short notice and without implementing new resources or technologies.
Within this paper the approach and field test results to extend the range and electrical power of the emergency power supply with the help of a wind turbine via the public medium voltage network and thus in particular to be able to continue to supply critical consumers in an emergency case, are discussed.
The results demonstrate that a stable 20 kV TIM powered by a diesel generator and extended by a Nordex 3 MW-class wind turbine as well as PV infeed on low-voltage side can be established and operated safely and stable without major changes in the turbine´s configuration.
A part of a distribution management system (DMS) is presented on a conceptual level, focusing on the graphical user interface (GUI) to operate an aggregation system in an active distribution network. Power plant functionalities with numerous of small distributed generation (DG) shall be provided to the network operator in order to assist during system restoration. Two main challenges arise. On the one hand, large amounts of available information from DG must be condensed to be handleable by a human operator. On the other hand, an interface should be designed to provide effective control to the combined capabilities of DG. Literature review, as well as expert workshops involving the personnel of network operator’s control center, identified several key features for the planning and execution of build-together and top-down restoration approaches, exploiting the potential of task-specific forecasts in an improved manner by e.g. enabling an easy aggregation and visualization of these. Especially the identification of geograhical areas with reliable and flexible residual loads are of interest to a distribution system operator (DSO) as well as the actual control over DG to influence the residual load and to provide frequency maintaining ancillary services. The proposed GUI is specifically designed to be a crucial part of a future DMS used by DSOs in the case of critical grid situations or power system restoration with high shares of DG.
The pan-European power grid is experiencing an increasing penetration of Variable Renewable Energy (VRE). The fluctuating and non-dispatchable nature of VRE hinders them in providing the Ancillary Service (AS) needed for the reliability and stability of the grid. Today’s grid is reliant on synchronous generators. In case of sudden frequency deviations, the inertia of their rotating masses contributes significantly to the stabilisation of the system. However, as the modern power grid is gravitating towards an inverter-dominated system, these must also be able to replicate this characteristic. Therefore, Energy Storage Systems (ESS) are needed along the VRE. Among the different ESS, Pumped Hydro Storage (PHS) can be identified as particularly convenient, given its cost-effective implementation and considerable lifespan, in comparison to other technologies. PHS is reliant on difference in altitudes, which makes this technology only available if suitable topographic conditions exist. The ALPHEUS project will introduce a low-head PHS for a relatively flat topography. In this paper, a grid-forming controlled inverter coupled with low-head PHS that can contribute to the grid stability is introduced, emphasising its ability to provide different AS, especially frequency control, through the provision of synthetic system inertia, as well as fast Frequency Containment Reserves (fFCR).
According to the Renewable Energy Law (EEG) in Germany from 2017 the site quality of a wind energy turbine has to be determined after commissioning and verified at least every five years which has to be seen in the context of tendering process for renewable energies. This evaluation directly has an influence on the previous remuneration of the electrical power that has been injected into the public grid. The wind energy association of „Fördergesellschaft Windenergie und andere Erneuerbare Energien (FGW e.V.)“ has published a completely new corresponding technical guideline part 10 revision 0 (TR 10) in 2018 in order to define new methodologies for calculating given and assessing missing input data.
Within this technical guideline the legal background, technical properties of wind turbines as well as the regression of operational wind turbine data are explained. Moreover, the data saving respectively provision as well as the proceeding of evaluation are described. Until today, already several round robin tests between different working group members have been performed. As in every release of new technical guideline, even the procedures of TG 10 are still facing a lot of challenges.
However, it is not only the working group which is on pressure, but even wind farm operators as well as operating managers and manufacturers of wind turbines are forced to prepare their internal workflows and data management. The major challenge is to provide the necessary data in a correct and suitable manner – a tough work while the TR 10 is still under construction.
This presentation will give a look insight the status quo of the latest working group results and the main challenges which wind farm operators and manufacturers have to face in Germany – and perhaps in the future even on the international level.
The operation of a system with 100% converter-interfaced generation poses several questions and challenges regarding various aspects of the dynamic characteristics of the system. The effects related to the power-frequency control are among the most relevant issues which need to be investigated. The paper takes into examination the case of an existing isolated power network of a geographical island assuming a completely converter-interfaced generation. The factors playing a fundamental role in the assessment of which sources should be designated to be grid-forming are identified with a small-signal model formulation of a generic multiple oscillators system. Simulations and analysis performed on the case study of the island’s power network performed with the software NEPLAN are used to show the impact of different aspects which can play a role in the design and the operation of the system: these aspects are the operating mode of power converters control, location and distribution of the generation sources with grid-forming capabilities, as well as number and ratings of these sources. The impact of relevant control parameters such as virtual inertia constant and virtual impedance is also investigated.
The different measures for voltage and frequency maintenance as well as black start capability, which are mainly provided by conventional power plants today, are summarized. The ongoing energy transition, together with a politically decided phase out of nuclear and coal fired power plants in Germany, leads to a lack of availability of ancillary services which has to be filled with distributed generators (DG).
Within the research project “SysAnDUk”, methods for creating power plant characteristics to provide ancillary services for the network and system operation with a large number of DG shall be developed. The focus of the project lays in critical network situations and during system restoration. To do so, the two ideas how to aggregate numerous Wind Power Plants (WPP) connected to the high voltage grid and how to realise a technical virtual power plant (TVPP) that can coordinate a huge number of small PV-units connected to the low voltage and medium voltage grid are introduced. Appropriate probabilistic forecasts are implemented in both concepts to display available active power with different safety margins.
It is could be shown by dynamic simulation, that a WPP can provide voltage maintenance by providing a coordinated reactive power feed-in and a voltage control during early stage of system restoration starting in the high voltage grid. This provides new option for necessary compensation of unloaded lines.
The electric grid is characterised by dynamic and fluctuating changes of power consumption and generation, especially with high penetration of renewable energies. In order to maintain system security, grid operators have to know how much energy is fed into their grid and perform grid calculations based on this information. In addition, they must also know in advance the future feed-in and the resulting grid state, in particular vertical power flows. Especially renewables, which are strongly influenced by regional weather situations, make forecasts on a local scale necessary.
In this paper, we present an approach to predict the main components of the vertical power flow at transformers connecting high voltage and extra high voltage levels. In this work, the main components of the vertical power flow at transformers connecting high voltage and extra high voltage levels are predicted. The vertical power flow can be understood as the sum of all generated and consumed power processed by a transformer. The main components of power generation and consumption that are considered in this paper are:
While the vertical power flow is a measured quantity and can therefore be forecasted using a direct machine learning approach, its decomposition into these aforementioned components is generally unknown. Forecasts of the latent, underlying components at each transformer could give grid operators new insights into upcoming grid congestions by revealing the local sources of possible power peaks and troughs.
We propose a three-step approach to predict the vertical power flow as well as all of its components' power timeseries at each transformer. First, at each transformer the assigned capacity of each power component is calculated in a linear regression using vertical power flow measurements and local approximations of component signals. Additionally, master data of power plants and consumers are used as physical plausibility constraints. Second, for each transformer each component signal is predicted in a machine learning approach using an Extreme Learning Machine (ELM) and scaled by the assigned capacities. Third, a vertical power flow forecast is calculated as the sum of all component forecasts.
The approach is tested in an experiment including 176 transformers connecting distribution grids and the transmission grid in Germany.
Results show that our method is able to calculate day-ahead forecasts of the vertical power flow at most of the tested transformers with a good accuracy and no remaining bias, indicating that all major components are detected and predicted accurately. Additionally, we compare our method to a machine learning model trained directly on the vertical power flow and discuss the possibility of coupling both approaches, obtaining a root mean square error (rmse) optimised vertical power flow forecast and component forecasts at the same time.
To achieve future climate targets and compensate for diminishing fossil fuel resources, an increasing amount of clean, renewable energy is needed as an alternative. Hence, during the past decade, there has been a rapid increase in the integration of such renewable resources, mainly solar and wind, to the current electrical grid. However, the energy from such sources is highly intermittent and at times can be completely unrelated to the demand in the electrical network. Hence, in an electrical grid powered mostly by renewables, i.e. via wind turbines and photovoltaics, there would be a significant difference between the instantaneous power generation and consumption. To bridge this gap, a feasible method of large-scale energy storage is required. The Storage Power Plant (SPP), which uses hydrogen as its primary fuel, is one such solution.
In this investigation, the proposed SPPs are part of an isolated network that contains conventional thermal and hydroelectric power plants along with a large share of wind farms. Two separate events are created, the first involving excess power generation from the wind farms and the second representing the shutdown of a coal-fired thermal power plant. The dynamic interaction of the SPPs with the other power plants and the roles of its internal components are analyzed in both cases. The results highlight that the principles of power system control which are satisfied today with thermal power plants running on coal can also be met by the SPPs in the future.
This paper proposes the analysis of the frequency support from a Type-IV wind turbine interfaced to the grid via a grid-forming converter. The frequency support including the inertial response and the fast primary frequency support is implemented by integrating a grid-forming control to the control a back-to-back converter. Knowing that the inertial response from a grid-forming converter can be tuned by varying the inertia constant of a grid-forming control, it is possible to extract different amounts of kinetic energy from the rotating masses of a wind turbine and positively impact the grid frequency dynamics during a grid frequency event. However, the increased inertial response may lead to an increased loading imposed on the mechanical components of a wind turbine. To elaborate and verify these statements, the analysis is aimed at two objectives. Firstly, it is estimated through time domain simulations whether the inertial response from a grid-forming control-based wind turbine has a significant active power contribution or can be neglected compared to the share of the primary frequency support. Secondly, based on the analyzed impact of the inertial response on the grid frequency and the mechanical loading, a value of the inertia required from a wind turbine with a grid-forming converter is recommended.
Continued cost reductions of solar PV and wind have led to large-scale deployment of renewable energy globally. However, the potential of domestic renewable energy resources in Japan is often overlooked. Despite the limited land area available for large-scale solar farms, a large potential is enabled by alternative types of PV deployment, including floating PV, building-integrated PV and agrivoltaics. Our GIS analysis identifies a total PV potential of around 2500 GW, which is 3 times that required to supply a 100% renewable electricity system in Japan. Japan also has enormous offshore wind resources. The potential of offshore wind in Japan is often overlooked because there is limited shallow water. However, rapid development of floating offshore wind in recent years allows access to much better wind resources in deeper water. Using GIS analysis, we assign a score to every 300m * 300m cell in Japan's exclusive economic zone based on various factors. An indicative wind farm with a score of 0.6 is 30km away from the coast and has an annual mean wind speed of 7.5m/s at 150m and a sea depth of 100m. The sites that score higher than 0.6/1 represents a total capacity of 2100 GW. The identified PV and wind potential combined can supply more than 10 times Japan's current electricity demand. In this study, a long-term hourly energy balance analysis is presented of Japan's electricity system under 100% renewable scenario. The electricity supply is dominated by solar PV and floating offshore wind, together with existing renewable hydro, geothermal, and bioenergy capacities. Off-river pumped hydro energy storage (located away from rivers and therefore has minimal environmental impacts) and a hypothetical HVDC backbone that connects the 10 service areas are deployed to balance supply and demand. The Global Pumped Hydro Atlas developed by the ANU found 2413 off-river pumped hydro sites in Japan with a combined storage potential of around 53000 GWh. Differential Evolution is utilized to determine the optimized configuration of generation, storage, and transmission with a list of energy, resource, transmission, contingency, and reserve constraints. Our preliminary results show that the LCOE of the 100% renewable electricity system is $90USD/MWh, and requires 500 GW of PV, 50 GW of offshore wind, and 175 GW of pumped hydro with 11h storage. If the contribution of solar PV is limited to 10% to better utilize the wind resources, then the LCOE of the system increases to $120USD/MWh, with 75 GW of PV, 195 GW of offshore wind and 100 GW of pumped hydro with 24h storage. The modelled LCOE is competitive when compared with the current wholesale electricity price in Japan ($90 USD/MWh) and much cheaper than the cost of electricity produced from Australian hydrogen (>$180 USD/MWh). This study demonstrates that Japan can be self-sufficient for 100% renewable electricity not only at a competitive cost but also with minimal environmental impact.
The paper provides an overview of some of the challenges related to operating inverter-based generator units in weak grids. Special focus is on illustrating in a simple manner the change in voltage sensitivity caused by operating the inverter with a high voltage angle relative to the main load center of the grid. A new measure of the network strength based on the voltage sensitivity is proposed. It is shown that central voltage control is necessary to ensure a high transmission capacity. On top of this, a solution for providing dynamic voltage control at turbine level has been presented. The role of grid forming inverter-based generators and the challenges are discussed, and it is concluded that a gradual implementation will be the most efficient way forward.
This paper summarizes the findings of Danish project - NSON-DK mainly regarding the balancing operation of the power system of North Sea countries with focus on the Danish power system and its regions DK1 and DK2. The study covers all the steps of the operation from the Day Ahead (DA) Market to the real time balancing of generation and demand. For that reason, DTU Balancing Tool Chain is used. The paper analyses a) the value of offshore grid on balancing of forecast errors, b) the impact of forecast errors on manual and automatic reserves and c) the quality of electrical frequency in the near future considering high VRE penetration d) value of including wake in the balancing studies with high share of renewables. Recommendations are provided in terms of balancing requirements for future power systems.
The high rated rotating synchronous machines that were dictating the operation of power systems are gradually replaced by renewable energy sources. The removal of synchronous generation (SG) makes power systems “weaker” in terms of inertia and short circuit level. This large-scale integration of grid-connected voltage source converters (VSCs) lead to the necessity these units to operate in a grid-forming way. Grid forming converters should be able to provide a natural voltage source behavior. They must operate sufficiently, even in weak systems with low short circuit levels. Furthermore, as the transition from SGs to grid forming converters cannot be immediate, they must be able to coexist with SGs and grid following converters. The controller of the proposed grid forming converter will be presented. There are several subcategories in grid forming “family”. The grid forming scheme presented in this paper falls in the subcategory of Virtual Synchronous Machine (VSM). VSM are designed to mimic the dynamic behavior of a SG. The proposed grid forming controller is used to examine the stability of a big power system, dominated by converters. The stability of this converter-based generation system is examined, after big disturbances. The simulations results on a more stable power system as the share of grid forming converter increases, that can reach up to 100%. The grid forming converters are tested when they coexist with SG or grid following converters.
Large-scale integration of Inverter-based Resources (IBRs) such as wind parks brings new challenges in modelling, analysis and simulation of power systems for protection engineers and system operators. IBRs have complex controller and electronic circuit structures but the use of faster solvers in protection studies will save valuable engineering time and alleviate transfer of black box models between different parties including Original Equipment Manufacturers (OEMs) and system operators. This paper aims to evaluate solver relaxation in an objective manner for the first time in the analysis of impact of IBRs on power swing protections. The foundation of this work is a detailed Electromagnetic Transients-type (EMT-type) model of a wind park built in PowerFactory® software. Then, an equivalent of this model compatible with RMS solvers is developed in the same software by mapping circuit and controller components to ensure objective comparisons. The two models are equivalent in terms of power, short circuit ratings and fault behavior within the typical range of simulation resolution of RMS solvers. By using these two equivalent models, the impact of wind generation on power swing protection elements based on a conventional approach are studied in EMT and RMS solvers. The results suggest consistency in the response of the two solvers for the PSRC WG-D6 test case under investigation.
As the penetration levels of RES increase together with climatic changes producing more extreme weather conditions, the uncertainty of the weather forecasts and power production forecasts can no longer be ignored for the grid operation.
In meteorology, uncertainty has been modelled by so-called Ensemble Prediction systems (EPS) since the late 1990ies. The adaptation of uncertainty forecasts also took many years and a number of experimental investigations on how to present and communicate uncertainty of critical or adverse weather conditions to the public and decision makers alike. Up to date, there are no no established best practices in how to communicate forecast uncertainty and success in communicating and applying weather forecasts with uncertainty information very much depends on the application, the users and the objectives.
The samevarying acceptance and application of uncertainty forecasts is also observed in the power industry, where uncertainty forecasts from ensemble prediction systems are already available since the early millennium years. Nevertheless, applications using uncertainty information are still rare and mostly used in areas, where the penetration levels from renewable energy are at times in the range 60-90% of the load.
In order to support the industry in adapting uncertainty forecasts into their business practices, the IEA Wind Task 36 has started an initiative in collaboration with the Max-Planck Institute of Human Development to investigate the barriers that prevent the industry to adopt uncertainty forecasts into their decision processes.
In the first part of the initiative, we designed a forecast game as a online demonstration of a typical decision-making task in the power industry. The game was introduced during an IEA Wind Task 36 workshop and thereafter released to the public. When we closed data collection, the game been played by 120 participants. We will present the game, discuss first results, and introduce some new features of an updated version of the game. In addition, we will introduce new experiments we currently work on to investigate critical questions that arise when working with uncertainty forecasts. Along with these experiments, our aim is to provide training tools to demonstrate the use and benefit of uncertainty forecasts by simulating decision scenarios with feedback and allowing people to learn from experience how to use such forecasts. The experiments will also help to stimulate discussions and develop new ideas from the feedback we receive.
The paper analyses the supply costs of hydrogen for Germany on the basis of the German electricity mix and for the production by renewable energies worldwide. It examines the effects that significantly lower investment costs would have on hydrogen supply costs. For the global provisioning costs, transport via ammonia and the associated costs were analysed. The results show that with domestic production, even at current investment costs and taking into account the stack service life in operating hours, the most favourable operation mode would already correspond to 6000 to 6500 full load hours. With decreasing investment costs and increasing efficiency and lifetime this value decreases.
For global hydrogen production, it has been found that, in addition to electricity production costs, full load hours have a major influence. This means that it mainly prefers locations for the production of wind energy that generate the most attractive production costs. Transport costs on the contrary are mainly driven by conversion costs. Meaning that wherever it is possible to transport hydrogen via pipeline without additional conversion, this is preferable. Especially when hydrogen production costs are falling sharply, the influence of transport costs can be significantly high.
The operational use of wind and solar power production forecasts has become widespread in the electric power industry and their benefits for the management of the variability of the generation associated with these renewable energy technologies have been documented in a number of studies. However, there is considerable evidence that the full potential value of the wind and solar forecasts in many applications is often not realized. This is typically related to three factors: (1) the specification of the wrong forecast performance objectives in the forecast solution selection process, (2) the use of poorly designed benchmarks or trials to select a forecast solution for the user’s application and (3) the use of non-optimal evaluation metrics to assess the performance of candidates or existing forecast solutions.
This issue was addressed by a group of experts within the scope of the International Wind Energy Agency (IEA) Wind Task 36. This group prepared an IEA Recommended Practice on Forecast Solution Selection (RP-FSS), which provides guidance on the selection of a new, alternative or additional forecast solution, the execution of a forecasting trial or benchmark, and the evaluation metrics and methods used to assess forecast quality. The RP is composed of three documents. The first part, “Forecast Solution Selection Process”, deals with the selection and background information necessary to collect and evaluate when developing or renewing a forecasting solution. The second part of the series, “Benchmarks and Trials”, offers recommendation on how to best conduct benchmarks and trials in order to evaluate the relative performance and the "fit-for-purpose" of forecasting solutions. The third part, “Forecast Evaluation”, provides information and guidelines for the effective evaluation of the performance of forecasts and forecast solutions. The RP documents are intended to provide guidance to stakeholders who are seeking to initiate or optimize a forecasting solution that will maximize the benefit for their specific applications.
Initial feedback from users of the RP-FSS document series indicated that there was a strong desire for examples that illustrated the key points of the RP-FSS, their practical implementation and their impact on the identification of an optimal forecast solution. A number of examples have been constructed now, some in collaboration with an emerging online and open source tool for forecast evaluation called the Solar Forecast Arbiter (SFA)
The presentation will include (1) insight into the reasons why the full potential value from existing forecast solutions is often not realized and (2) a brief overview of the contents of the three IEA RP documents and where to obtain them but focus on (3) practical examples of key points from the RP-FSS.
A major technical challenge for the future energy system is to ensure stable network operation and maintain high
voltage quality. However, problems with critical harmonic content in PV and wind parks can already be observed,
which are due to electrical resonance formation between inverters and the grid. With the established methods for
determining harmonic emissions, however, no statements can be made about possible resonances and the associated
consequences. There are different approaches to determine harmonic emissions by using the impedance characteristic
of the solar inverters , . The impedance curve can determined by measurement, analytically or by simulation
. In this paper, a new measuring technique called impedance spectroscopy will be presented.
II. IMPEDANCE SPECTROSCOPY
Impedance spectroscopy is a method of determining the frequency-dependent impedance of a device under test
(DUT) by means of a stepwise voltage excitation. The spectroscopy of inverters is performed during operation of
the DUT. For this purpose, the supply voltage of 50 Hz must be provided first, and the excitation voltage is then
superimposed on it. For further processing, the voltages and currents measured at the terminals are transformed
into the frequency domain. The excitation voltage is variable in frequency and is gradually increased during the
test. A V (fexc) and I(fexc) can be calculated for each frequency. Using those to calculate the VI -ratio will lead
to a kind of impedance curve. However, it stands out that this curve shows discontinuities at different frequencies.
Since these discontinuities cannot be described by the behavior of passive electrical components, they need to be
caused by the presence of internal harmonic sources at exactly these frequencies. Hence, instead of calling this
ratio „impedance“ it should be entitled as frequency-depending voltage current ratio .
By varying the phase position or the amplitude of the excitation voltage, several measuring points per frequency are
generated. From the calculated current and voltage responses, the inverter can then be described for each frequency
step as voltage sources with internal impedance (so-called Thévenin equivalent). The voltage sources determined
can then be used as a measure for the actual harmonic emissions of an inverter. To analyse resonances that may
occur, the determined frequency-dependent inverter impedance can be compared with the impedance curve of a
given grid connection point (so-called impedance-based stability criterion ).
III. COMPARISON DIFFERENT IMPEDANCE CURVES FOR SOLAR INVERTER
In this work the impedance curves for different solar inverter will be presented. The examined devices differ in
power class, voltage level, filter and control topology. The power spectrum ranges from 20kW to 2500kW units.
The curves are compared with each other and possible factors influencing the curve are discussed. In addition, the
test bench set up for this purpose is presented.
With the increasing complexity of the electrical transmission and distribution grids, the expectations for
the grid-friendly behaviour of grid-connected generators and energy storage systems are rising. Among
other things, the devices are expected to react to grid disturbances with a certain behaviour. In the
past, this has concentrated on under-voltage (Under-Voltage Ride-Through, UVRT), whereas recently
many connection guidelines have included over-voltage (Over-Voltage Ride-Through, OVRT). Real over
voltage events are still largely unknown, which can lead to a potential over sizing in terms of OVRT
capabilities and thus to higher costs. Different types of test benches are available for testing the OVRT
capabilities. These can vary greatly in terms of transient events, voltage ramps or impedance changes
during the simulated fault. Since especially power electronic devices like inverters can be sensitive to
transient voltage changes, it is important to analyse which test bench can simulate the most realistic over
II. DESCRIPTION OF THE ANALYSED TEST BENCHES
In this paper the test facilities listed below are compared with each other. All the test benches are able to
create temporary over voltages at the connections terminals of a device under test. They differ in various
electrical and non-electrical properties, which will be explained in this work.
A. Inverted inductive UVRT test bench
The inverted inductive UVRT test bench makes use of a common under-voltage ride through (UVRT)
B. LC-resonant tank
The LC-resonant tank adds a capacitor to the UVRT-test setup in order to form a series LC-resonant
C. Grid simulator
A grid simulator is commonly used as an AC-amplifier with a programmable output voltage.
D. Series impedance with reactive power source
In this test setup, a capacitive reactive power source is added in parallel to the DUT.
E. Auto-transformer switchover
For this test setup, an auto transformer with several tappings and minimum two circuit breakers is used.
F. Auto-transformer switch-in
For this test setup, an auto transformer with several tappings is used, in this case, only one circuit breaker
III. COMPARISON OF THE TRANSIENT BEHAVIOUR OF TEST FACILITIES
Due to the use of different technologies in the generation of the over voltage events, the transient behaviour
of the devices can vary greatly. Within the scope of this work, comparative measurements were carried out
on all facilities in order to determine the behaviour at the beginning and end of the fault. The following
factors are considered among others. The voltage change rate, the Synchronicity of the voltage change
on each phase and the impedance change during the fault.
The constant increase of renewable energy systems share is leading to a change of the dominating feed-in technology in the grids. Whereas conventional power plants are dominated by synchronous generators, renewable energy sources are mainly connected to the grid through inverters. This leads to new challenges regarding ancillary services as well as essential physical qualities which would usually be provided by the characteristic of a typical synchronous generator. In inverter-dominated grids, however, these have to be re-created through new inverter technologies in order for the grid to remain stable.One of the most important features of a synchronous generator is the inherent power reaction to sudden load variations, which is based on its voltage source behavior. This behavior is important within the very first milliseconds to cover load variations. Opposed to this stands the fundamental difference of renewable energy sources typically acting as a current source. In a scenario where a high penetration of inverter-based generation in grids occurs, the thereby reduced number of synchronous generators and hence the absence of their voltage sources behavior and balancing characteristics, can lead to instability of the grid.
A possible solution for the scenario above is to use inverters as voltage sources to regain an inherent power reaction. The field of application is usually a stand-alone supply in island grids. In order to guarantee a safe coupling with the mains it is necessary to have a superior control emulating the elementary behavior of synchronous generators. This control will provide synthetic inertia in order to reduce frequency deviations. Thus, instead of using conventional power plants to provide the services of inertia, grid forming inverters with virtual synchronous machines controls can be used to stabilize the grids through synthetic inertia. However, since the inverter has a technical maximum current due to its hardware, the current must be limited by control in situations with typically large currents such as short circuits.
This paper discusses possibilities of current limiting mechanisms for grid forming inverters. One method is integrated into an existing inverter control system, where correction values are added to the virtual synchronous machine control in amplitude and, if necessary, in phase angle. The functionality of current limiting is afterwards demonstrated in laboratory tests. At first, tests are carried out considering a reduction of the voltage amplitude for fault modelling. In further tests, a test network with a grid fault event is then simulated using a power hardware-in-the-loop system. The concept for the current limitation in grid fault situations shows that it is possible to limit over-currents while maintaining the dynamic behaviour of grid forming inverters.
In Germany, the combined installed power of Offshore and Onshore wind energy generation has surpassed 60 GW. More than 50% of this capacity is located in the control area of the German Transmission System Operator (TSO) TenneT TSO GmbH (TenneT). Wind energy plants are usually connected to the high-voltage grid and therefore primarily controlled by Distribution System Operators (DSOs) or direct marketers. As a result contingencies located in the extra-high voltage (EHV) grid or its connection to the lower voltage grid need to be resolved by cooperation between the grid operators involved. Currently, the coordination required for the implemented cascaded process takes several minutes. Especially in an emergency state, such a task could e.g. be accelerated by establishing a direct control mechanism for TSOs in addition to the coordination with the DSO. By that, contingencies can be dealt with on a shorter timescale. As part of the funding program 'Smart Energy Showcases (SINTEG)', TenneT initiated a collaboration in the showcase region of 'NEW4.0' in Northern Germany. Together with direct marketers ANE GmbH & Co. KG, Statkraft Markets GmbH, virtual power plant service providers ARGE Netz GmbH & Co. KG, Energy & Meteo Systems GmbH as well as Fraunhofer Institute for Energy Economics and Energy System Technology (Fraunhofer IEE), several wind parks, were connected to the TenneT control room in different subprojects using different interfaces in order to compare both approaches. As part of different field tests those wind parks were curtailed and the resulting effect on the power flow from high voltage to EHV grid was measured directly at the transformers in the vicinity. This allowed the estimation of the potential influence of a control mechanism. While positive results were achieved and the effect of single wind parks could be measured without further knowledge of the medium voltage grid model, a clear conclusion is yet to be made. A permanent implementation use-case heavily depends on the outcome of the redispatch process being remodeled in the light of new legislation in Germany ('Redispatch 2.0'), as the possible synergies with measures of the alert state using the upcoming interfaces are yet to be determined.
On the path towards an energy system powered entirely by Renewable Energy Sources (RES), power electronic converters will have to take over more and more functionalities from Synchronous Generators (SG) to ensure a stable and secure operation of the power grid. Moreover, it is widely recognized that the use of Grid Forming Converters (GFC) is necessary to fully meet these requirements. Over the few last years, different concepts have been developed to achieve grid forming characteristics of static power converters.
The next essential step is to agree on an exact definition and specification of GFC electrical behaviour as well as to define a suitable conformity assessment procedure. For this purpose, a standardized testing guidelines for GFC are needed to assess those functionalities, which are relevant for dynamic grid stability. As a British-German joint work of the two research projects Battery-VSM and VerbundnetzStabil, a first draft of such guidelines is being currently developed.
In this first part, we summarize the requirements from GFC based generation from system perspective, review the current requirements on power electronic converters and line out the specifications determining whether a converter is grid forming or not. We define the behavior of a GFC in the normal operational range such as voltage source behavior, power quality impact and provision of inertial response. Furthermore, specifications to evaluate the grid-stabilizing behavior of GFCs during grid-fault or overload situations, which exceed the converter’s current limits, are given.
On the path towards an energy system powered entirely by Renewable Energy Sources (RES), power electronic converters will have to take over more and more functionalities from synchronous generators to ensure a stable and secure operation of the power grid. Moreover, it is widely recognized that the use of Grid Forming Converters (GFC) is necessary to fully meet these requirements. Over the last years, different concepts have been developed to achieve grid forming characteristics of static power converters.
The next essential step is to agree on an exact definition and specification of GFC electrical behaviour as well as to define a suitable conformity assessment procedure. For this purpose, a standardized testing guideline for GFC is needed to assess those functionalities, which are relevant for dynamic grid stability.
As a British-German joint work of the two research projects Battery-VSM and VerbundnetzStabil, a first draft is currently under progress.
This work is the second of four papers on testing GFC and focuses on steady state operation. The GFC must demonstrate stable operation in grid-tie mode, load sharing with other sources as well as an appropriate response to unbalances and harmonic distortion.
On the path towards an energy system powered entirely by Renewable Energy Sources (RES), power electronic converters will have to take over more and more functionalities from synchronous generators to ensure a stable and secure operation of the power grid. Moreover, it is widely recognized that the use of Grid Forming Converters (GFC) is necessary to fully meet these requirements. Over the last years, different concepts have been developed to achieve grid forming characteristics of static power converters. The next essential step is to agree on an exact definition and specification of GFC electrical behaviour as well as to define a suitable conformity assessment procedure. For this purpose, a standardized testing guidelines for GFC are needed to assess those functionalities, which are relevant for dynamic grid stability. As a British-German joint work of the two research projects Battery-VSM and VerbundnetzStabil, a first draft of such guidelines is being currently developed.
One of the necessary characteristics which makes a converter a GFC is its ability to provide inertial response during the dynamic frequency changes in the system. This paper focuses on how to demonstrate and quantify an inertia-equivalent behavior of a GFC. The response to a transient system event is quantified in terms of a damping constant D as well as an equivalent inertia constant H. A few alternative methods are proposed for empirical estimation of those parameters accompanied by selected laboratory test results and practical considerations.
With the increasing penetration of Renewable Energy Sources (RES) in electricity generation, more and more energy is provided using power electronic converters. Moreover, we are moving towards a future where RES can potentially supply all or close to all of the system demand in many countries, replacing Synchronous Generators (SG) used by conventional plants. Therefore, power electronic converters will have to take over more and more functionalities to ensure a stable and secure operation of the power grid. A promising solution for these tasks is the use of Grid Forming Converters (GFC). Over the last years, different concepts have been developed to achieve grid forming characteristics. In this situation, the next essential step is to agree on an exact definition and specification of GFC. In order to verify these characteristics and to ensure reliability, it is necessary to develop a standardized testing procedure. For this purpose, a draft version of a testing guideline is currently under progress, focusing on those functionalities, which are relevant for dynamic grid stability. It defines detailed test setups, calculation of relevant system parameters as well as evaluation criteria. Due to the complexity of the topic and the numerous aspects that are to be considered, we propose a session entitled "Testing Characteristics of GFC" consisting of four contributions on different aspects.
A third technical requirement of GFC is the capability to contribute to system needs even in situation, in which the physical current and power limits of the converter are exceeded. Due to the current limited nature of power electronic converters, GFC have to operate in abnormal operation modes in case of overload and grid fault situations. Therefore, GFC have to be tested under different overload and grid fault conditions to ensure grid stabilizing behavior. This contribution outlines a testing procedure for different fault scenarios. It includes voltage driven overload scenarios like over and under voltage events. Moreover, phase driven overload scenarios are addressed. This includes phase jumps, RoCoF events and system split scenarios. For all scenarios, the GFC should retain grid-synchronicity and inject stabilizing current.
As part of the International Energy Agency’s (IEA) Task 36 on the improvement and usage of wind power forecasts, it is essential that we take a close look at existing data definition and data exchange standards in the wind energy industry, weather forecasting community, and electricity system and energy market operations. The objectives of the IEA Task 36 Work Package 2 includes producing recommendations and best practice documents that can be adopted internationally with the intention of enabling quality improvements and ease of forecast usage. These objectives will ultimately lead to cost savings both from a forecast provider and consumer perspective.
Currently, there are no international standards for wind power forecast requisite data definitions or methods of transmitting forecast data. There have been, however, standards developed around related industrial applications and for specific regional applications of renewable energy forecasts. There are strengths and weaknesses associated with these established processes as well as from experienced forecast service providers that may be leveraged upon in the future.
In this presentation, we discuss two levels of standardization for data and data transfer methods based on extensible and proven methods for streamlining the configuration and delivery of renewable energy forecasts. While not every forecasting system is identical with regards to inputs that can be used and their relative benefit to the quality of the forecasts, there are common required and optional inputs that established forecast systems need from the forecast consumer.
The two levels are separated by the type of forecast consumer: (1) lower level users such as small developers or third parties participating in an energy market, and (2) high level users that can programmatically implement a repeatable and scalable system. The data definitions, documentation and data types are identical for both users. However, the low-level users, in our definition, require options for data exchange to/from the forecast provider.
This two-level approach is deliberate to capture the varying needs of a large majority of renewable energy forecast users thus increasing the likelihood of wider international adoption of the recommended practice. In addition, the recommended practice aims to build upon IEC standards already in use by, for example, European TSOs. We will also discuss during this presentation extensibility with future focused “smart” technologies that account for more widespread and distributed deployment of renewables.
Since the wind energy is renewable and environmental natural resource, the utilization of wind power plant increased quickly. In the future, the development of wind power utilization will focus on large offshore wind farms (OWFs) .
Due to the distance between OWF and the onshore point of common coupling (PCC), voltage source converter (VSC) HVDC technology is widely applied for large OWF grid connection. Also, in order to reduce the investment of offshore HVDC converter platform, diode-rectifier unit (DRU) is proposed . Since the DRU is a passive rectifier without control possibility, in order to ensure the stable operation of OWF, the wind turbine controllers should modified (e.g. grid forming control). Moreover, during the start of OWF, there could be a period where the DRU HVDC and MVAC cable system (e.g. 33kV or 66kV) in parallel operation. Large inrush current could occur during the parallel operation.
For the stable operation and improvement of the dynamic behavior of OWF with diode DRU, a new method is proposed in this paper: application of a phase shifting transformer (as shown in Fig. 1.) on the MVAC cable for stable operation of the OWF with DRU .
Fig. 1: Application of PST for OWF with DRU
In this paper, the blackstart, static and dynamic operation of the OWF and DRU with the PST are analyzed in detail.
 U. Karaagac, J. Mahseredjian, H. Saad, S. Jensen and L.J. Cai, "Examination of Fault Ride-Through Methods for Off-Shore Wind Farms Connected to the Grid Through VSC-Based HVDC Transmission", Paper published in 11th International Workshop on Large-Scale Integration of Wind Power into Power Systems, November 13-15, 2012 in Lisbon, Portugal.
 T. Hammer, S. Seman, P. Menke, F. Hacker, B. Szangolies, J. Meth, J. Dorn, K. Loppach, R. Zurowski, "Diode-Rectifier HVDC link to onshore power systems: Dynamic performance of wind turbine generators and Reliability of liquid immersed HVDC Diode Rectifier Units", Paper published in CIGRE 2016.
 Patent P26336DE: Verbindung zwischen Offshore-Energiesystemen und Landstromnetzen sowie Schwarzstartverfahren für Offshore-Energiesysteme.
In the face of the increasing penetration of renewable energy sources, the synchronous machine-based power plants are slowly being phased out. This fact is potentially creating severe system issues in the transmission network. Key among the problems faced are the reduction of short circuit levels, inertia, as well as steady-state and dynamic reactive power support. To enhance the transmission system power transfer capability, complementary technological solutions, including synchronous machines, energy Storage Systems, and reactive power compensation devices can be combined to provide more services than a single device. Phoenix project, funded by Ofgem through Network Innovation Programme, is deploying a hybrid synchronous condenser (H-SC), which is a hybrid system of synchronous condenser (SC) and static synchronous compensator (STATCOM), and at the same time looking at the option of combining SC and battery energy storage system (BESS). In order to maximize the benefits of the hybridization of SC and BESS, there remains a need to assess and quantify the grid support functionalities provided by the SC and BESS individually, as well as a part of a hybrid solution. Furthermore, the control of the BESS converter solution to mimic the synchronous machine termed as grid-forming control is also drawing significant interest from the recent years owing to the potential of improved grid response characteristic compared to the grid supporting controls using phase-locked loop (PLL) and DQ current control. In this paper, we analyse the grid support functionalities provided by the SC, BESS with grid-forming and grid-following control, separately as well as a part of a hybrid system with synchronous condenser.
The electrical infrastructure is becoming more complex due to the introduction of long HVAC cables, HVDC connections, widespread penetration of renewable energy sources (e.g. PV plants, wind power plants) and offshore electrical network development. The number of power electronic converters (PV-systems, wind turbines, STATCOMs, HVDC transmission systems, etc.) in modern power systems is rapidly increasing. In the past, devices such as wind turbines or PV-systems were directly coupled to low-voltage grids and connected to medium and high voltage grids via dedicated step-up transformers. However, with the greater availability of modular multi-level VSCs, power electronic devices are increasingly directly coupled to the HV and EHV grids. This trend creates challenges such as operational coordination of grid-connected converters and small-signal stability both in the sub-synchronous and harmonic (super-synchronous) frequency regions. This is mainly due to such systems being characterized by relatively low damping and hence exhibiting resonance interactions.
The increased use of power electronic converters in electric grids has led to several incidents in the past related to control interactions. With the emergence of VSC technology, these incidents occur at both sub- and super-synchronous frequencies (i.e. up to several kilohertz). However, no commonly agreed methods are available for the analysis of potential sub-synchronous and harmonic (super-synchronous) stability problems. Hence, there is a need to provide a general overview of the topic, highlighting the root-cause of sub-synchronous and harmonic stability issues of grid-connected power electronic devices supported by a state-of-the-art literature survey as well as industrial experience.
This paper will present the overview, status and outline of stability analysis in converter-based power systems which is investigated within the CIGRE working group entitled “Multi-frequency stability of converter-based modern power systems”. It will provide concise definitions, a literature review covering state-of-the-art contributions and aspects of modelling and analysis methods are discussed and aided by examples. The paper will describe the phenomena, consolidate definitions and explain available methods for analyses with their advantages and disadvantages as well as providing a common understanding on modelling, analysis, evaluation and mitigation techniques. Guidelines regarding the general approach to such studies and the availability as well as choice of tools will be also suggested in the paper.
Synchronous generators have served as the bedrock of electrical power systems by providing voltage and angle stability at the fundamental frequency. They “form” the grid by behaving as a voltage source and ensuring that the frequency and magnitude of voltages at different nodes in a power system network stay within a tolerable range from their nominal values. On the other hand, inverter-based generation from renewable energy sources such as wind and PV operate in a grid-following mode by behaving a current sources feeding into the nodes of the power system network. As the penetration of inverter-based resources increases to very high levels, it is necessary for wind and PV generators to operate as grid-forming sources for the stable operation of the bulk power system. This paper discusses control of doubly-fed induction generator (DFIG) based Type III wind turbines for grid-forming operation mode. The paper compares the sequence impedance responses of Type III wind turbines for grid-following and grid-forming operation modes. It is shown through impedance analysis that a Type III wind turbine has superior stability properties when it is operated in the grid-forming mode and it is less prone to subsynchronous resonance (SSR) as compared to conventional grid-following Type III wind turbine. The paper also shows that while a grid-forming Type III wind turbine can operate stably in very weak grid conditions, it may suffer from stability issues during operation under extremely strong grids.
The full paper will review scalar and vector control methods for the operation of Type III wind turbines in the grid-forming operation mode. For further evaluation of the control and operation of Type III wind turbines in the grid-forming mode using EMT simulation studies, a vector control method is selected that is most robust and requires minimal modifications for converting a grid-following wind turbine to a grid-forming wind turbines. It is shown that the grid-forming control of a Type III wind turbine can use identical vector current control approach as the grid-following turbine. The fundamental difference between the two operation modes, however, is in the implementation of the outer control loops built on top of the inner vector current control.
The motivation behind the grid-forming control of wind turbines is to improve fundamental frequency and voltage stability characteristics of the power system networks. However, it is also important to evaluate the stability behavior of grid-forming wind turbines at other frequencies. For instance, it is well-known that conventional grid-following Type III wind turbines are prone to subsynchronous resonance (SSR) when they supply to a series-compensated transmission line. This is owing to their negative damping characteristics at subsynchronous frequencies because of the induction-generator effect and active controls. To evaluate how this negative damping characteristics is influence by the grid-forming control of Type III wind turbines, this paper will compare the sequence impedance response of the grid-following and grid-forming Type III wind turbines. It is found that the grid-forming control of Type III wind turbines has much better stability behavior at subsynchronous frequencies compared to grid-following wind turbines. The impedance analysis that will be presented in the full paper will highlight fundamental differences between the impedance characteristics of the wind turbines when it is operated in the grid-following and grid-forming operation modes.
The electrical infrastructure is becoming more complex due to the introduction of long HVAC cables, HVDC connections, widespread penetration of renewable energy sources (e.g. PV plants, wind power plants) and offshore electrical network development. The number of power electronic converters (PV-systems, wind turbines, STATCOMs, HVDC transmission systems, etc.) in modern power systems is rapidly increasing. This trend creates challenges such as security of power supply from renewable generation, power system resilience and small-signal stability both in the sub-synchronous and harmonic (super-synchronous) frequency regions. This is mainly because such systems being characterized by low inertia and lack of centralized conventional generation providing system restoration after a blackout.
The increased use of power electronic converters in electric grids has led to several incidents in the past related to the security of power supply and power system resilience. Therefore, there is a need to provide frequency support as ancillary service to TSOs as well as black start service from renewable sources of electricity. Furthermore, there is a need to enhance the frequency support as well as black start services from wind power plants by battery energy storage system to improve the quality and reliability of the services. In order to see how these functions may meet currently discussed frequency support services, we focus on the situation in UK.
This paper will provide a numerical method to estimate the minimum required battery capacity given historical wind data for a reference site while upholding the service availability requirements. In the paper it is assumed that black start availability is a function of the time instance for which the service was initiated and governed by the power balance between generation and black start service consumption. The two main parameters for adjustment as design criteria are (i) battery initial charge and (ii) participating turbines. The paper will show the studies leading to the following observations:
Moreover, the paper will show that the disconnection of synchronous generators will lead to decrease of kinetic energy in the power system. It will require advanced control operation of renewable-based generation to compensate the lack of natural inertia. The implementation of two frequency support functions will be discussed in the paper, namely inertial response control and frequency containment (droop) control will be discussed and analyzed in this paper.
This paper presents an overview of the overvoltage ride through (OVRT) requirements and their influence on the grid. In this context, different decentralized generators – such as synchronous machines, wind turbines and PV inverters – have been modeled and examined with regard to their reactions during an overvoltage. Furthermore, overhead lines, cables and transformers were also modeled and their influence on the overvoltage was investigated. Additionally, a cost function for an extension of the OVRT capability of certain components is proposed. With these results, the topic OVRT can be better understood and can provide assistance for the future grids.