In order to study the interactions between electric mobility and renewable energy plants as well as the optimum installation areas, a model based on a cellular approach for an urban distribution grid at medium voltage level (approximately 30.000 inhabitants) is developed during the FFG research project “Move2Grid”. This model simplifies the complex grid structure, reduces the calculation time and enables a temporally resolved load flow calculation with annual load profiles. Based on, this model the impact of an increasing number of charging stations and the interaction with photovoltaic potentials are studied for different E-mobility and photovoltaic potentials penetration rates. The required charging profiles are modelled by using traffic analyses and probabilistic approaches. The determination of temporally resolved photovoltaic potentials are based on data from the “Styrian Solar-Roof Cadaster”  and weather data from the Austrian Zentralanstalt für Meteorologie und Geodynamik (ZAMG) .
The aim of the proposed paper is to show the impact and interaction of E-mobility and photovoltaic potentials for different penetration rates of both. Besides the determination of the self-sufficiency level, load flow calculations with annual load profiles are performed and are benchmarked in term of equipment overloads and deviations of the voltage range. This is done by using a newly developed evaluation tool. With this tool a quick determination of areas which are susceptible for equipment overloads and deviations of the voltage range, as well as the exact time of them and their duration for further analyses becomes possible. In order to avoid equipment overloads and deviations of the voltage range and the associated necessary grid expansion, demand side measures and the use of electrical storages are implemented in the power grid model, followed by load flow calculations and analyses.
While it is often assumed in the academic literature that EVs will be charged slowly overnight at home, a significant proportion of EV charging could exist as ‘destination' charging while parked during their users’ visits to workplaces or amenities such as shopping centres, supermarkets, gyms, cinemas and motorway service stations - where cars are left for durations ranging from ten minutes to three hours. A move from a solely domestic charging-based EV uptake to one focused on the widespread availability of public charging could serve to enhance the convenience of EV usership, enable EV access to those without off-street parking (which applies to 43% of households in the UK) and has the potential to reduce system cost: according to findings from the 'My Electric Avenue' project, 32% of local electricity networks across GB will require intervention when 40% - 70% of customers have at-home EV charging. By instead encouraging users to charge away from home at their place of work or other places where they leave their car, the installation of charging infrastructure can be directed towards areas of greater spare capacity or with more potential for ‘smarter’ network operation which could allow a higher penetration of EV charging.
This paper presents a Monte Carlo (MC)-based method for the characterization of likely demand profiles of EV destination charging at these locations based on smartphone users’ anonymised positional data captured in the Google Maps Popular Times feature. Unlike the majority of academic works on the subject, which tend to rely on users’ responses to surveys, these data represent individuals’ actual movements rather than how they might recall or divulge them. Through a smart charging approach proposed in this paper, likely electrical demand profiles for EV destination charging at different amenities are presented.
The method is demonstrated by way of two case studies. Firstly, it is applied to a large GB shopping centre to show how the approach can be used to derive suitable specifications for large charging infrastructure to maximise revenue or EV service provision. Secondly, it is applied to a GB supermarket in a residential area to show how the approach can be used to examine network impact for a distribution-connected destination charging facility.
The project has started in january 2018 en will run until march 2021, will yield at project end a reduction in CO2 of 2400 ton/year and 12,4 MW of extra renewable energy production.
The first chapter describes the challenges for the power system and for grid operators when electrifying the transportation sector to meet international climate goals. In the second chapter, a standard load profile for the charging process of EVs based on assumptions for the future electricity demand of the transportation sector and a standard business load profile is introduced. The third chapter reflects today’s state-of-the-art grid integration of EVSE, whereas Chapter IV describes expected future levels of the grid integration of EVSE. Chapter V analyses the behavior of EVSE and EV in case of grid faults. A summary and outlook is given in Chapter VI.
This paper presents an aggregated approach to control a compound of charging stations with respect to network congestions. The main focus of the paper will be on the automated temporal shift of the charging power of pooled charging stations in order to avoid limitation violations of the corresponding network section.
Based on simulated driving profiles of electrical vehicles (EVs) and a perfect foresight approach the flexibility of the charging processes is determined in a way, that no user is restricted in his mobility needs. Regarding the start- and end time of each travel the ratio of parking time and required charging time is formed as an indicator of the available flexibility.
The driving profiles are used in a mixed integer linear program (MILP) to optimize the procurement of the required energy for charging the EVs. To different marketing alternatives are considered: in the first step, the optimization considers only the Day-Ahead Auction. The second step combines the Day-Ahead auction with the continuous Intraday market. This combination, denoted as Intraday Redispatch, leads to a significant reduction of costs due to the higher volatility at the continuous Intraday Market. The other major advantage of the Intraday Redispatch is the possibility to trade energy at the continuous Intraday Market until 30 Minutes before delivery, so deviations of the forecasted arrival times of the EVs can be compensated.
The simulation carried out with 50 electrical vehicles shows the realized cost savings for an optimized charging based on the considered short-term markets and the resulting effects on the distribution grid. The Day-Ahead optimizations reduces the charging costs to 30% of the average costs, with the Intraday Redispatch trading strategy the costs could be lowered to even negative costs.
This paper focusses on the use of sector coupling in the transport sector. In particular, it investigates the direct electrification of the transport sector. In addition, an analysis of the influence of the required energy supply infrastructures on the use of sector coupling technologies for decarbonisation of the transport sector in Germany is carried out. The aim is to examine not only the influence of infrastructures on the choice of sector coupling technologies, but also the impact of the sector coupling options on infrastructures.
To this end, in the framework of the German Kopernikus project ENavi, the energy system model TIMES is expanded with regard to the conceivable sector coupling technologies. In order to be able to adequately evaluate these technologies, a simplified representation of the required energy supply infrastructures will also be implemented. By varying the infrastructure parameters, scenario-based analyses will then be carried out to determine to what extent the infrastructures influence the selection of sector coupling paths and thus the possible composition of Germany’s future energy system.
The main findings indicate that the potential of sector coupling technologies is very much dependent on the choice of greenhouse gas emission reduction targets. In freight traffic in particular, these options only become attractive when ambitious targets are established. With regard to infrastructures, it can be said that a detailed assessment of the infrastructures has a great influence on the energy system. Furthermore, the impact of the use of sector coupling is far from negligible and requires more detailed research.
The project "BOB-Solingen" - the abbreviation BOB denotes the German words “Batterie-Oberleitungs-Bus” – is intended to electrify the entire public transport sector by introducing a new kind of trolleybuses, which will be able to travel regardless of the vital overhead line by means of the included battery. BOB is the result of combining the recognized trolleybus technology with the latest battery technology and the intelligent charging infrastructure, creating the next generation of buses which, as a matter of fact, are able to drive on routes with no power supply as well.
Moreover, the project is planned to integrate charging stations for electric vehicles (EV), decentralized renewable power generators such as photovoltaic (PV) systems as well as a stationary power storage system. The stationary storage will consist of used trolleybus batteries to increase their cost efficiency by establishing a second-life utilization concept.
The entire DC system will be transformed into a Smart-Trolleybus-System (STS) allowing an intelligent control and management of the power flow in the overall system. The Chair of Power System Engineering at the University of Wuppertal will develop and implement the essential automation system for the DC grid to use its existing overhead infrastructure as effective as possible within its physical limitations.
In order to realize an intelligent control of the grid, the load flow of the current grid (including the trolleybuses) as well as of the future grid (including BOB) has to be modeled and simulated. By means of the simulation, critical grid situations can be detected. These might occur more frequently in future due to the fact that e.g. the additionally implemented batteries can cause an increased number of peak loads which have to be handled.
This paper intends to simulate and evaluate the power profiles of all operating trolleybuses as well as the planned BOB based on different possible operation circumstances (e.g. stopping or not at traffic lights and bus stops, different stopping durations, traffic influences, variation in temperature and passenger numbers). The power profiles for both the buses and additional actuators within the DC grid, such as PV systems, charging stations and stationary power storage units will be presented and discussed.
The grid state is expected to be strongly fluctuating due to the increased number of loads and feeders, some of which operate bidirectionally. Load flow calculations will be the key to point out the actual grid state in order to enable the essential intelligent grid control. Performance and capability will be presented and evaluated in this paper.
The study was evaluated on 10 different Swedish simulated detached houses without electric heating, and using various combinations of charging powers and fuse limits. The results show that the worst house—the house that needed smart charging the most—needed postponed charging 8 days a year to avoid breaking the fuse. Moreover, postponed charging increased the charging duration, and thus inconvenience to the EV owners, by at most 4 hours. On the other hand, the capacity-filling charging scheme could increase or decrease the charging duration—compared to the uncontrolled charging. An increase is expected if the difference between the fuse size and the house load is smaller than the uncontrolled charging power. The charging duration will be shorter if the difference between the fuse size and the house load is larger than the comparable uncontrolled charging power.
The capacity-filling scheme proved to be more convenient, as it did not increase the charging duration by more than 3 minutes. Moreover, it reduced the charging duration for at least 198 days a year.
The results indicate that charging the EVs by the available capacity—the difference between fuse size and house load—is recommended compared to constraining the charging power.
In this paper, we use first hand data of a real EV fleet of Tesla vehicles and their historical driving patterns to develop a two-stage stochastic optimization problem. This model maximizes the profit of a risk-averse EV aggregator that aims to place optimal bids on the day ahead in both energy and Frequency Containment Reserve (FCR) markets. Only uni-directional charging is examined, while we take into account uncertainty from prices and vehicle utilization. Case studies are carried out modelling individual vehicle driving behavior in different Nordic price areas in both winter and summer.
We identify a strong alignment of EV availability and periods of high FCR prices. Results show that consumption is shifted largely towards early hours of the morning. When compared to a reference "cost of charging case", up to 50% of the cost of charging can be covered in Norway, while the entire cost is met in Sweden.
Whether charging infrastructure expansion is effective will be determined by future demand, the quality of the coordination between stakeholders and exploitation of potential synergies between public, semi-public and private charging infrastructure.
Today, charging infrastructure planning must include all interested parties, many of whom had not been considered until recently. These include municipalities, public works departments, distribution system operators, transit authorities, car-sharing companies, charging infrastructure operators, supermarket chains, and landlords.
Because any one stakeholder can now play multiple roles, the traditional channels and directions of communication in the energy industry will change.
Charging infrastructure must meet conflicting criteria. Decision-makers need a tool or suite of tools that helps them to balance these competing criteria. Any such tool must make electromobility development potential apparent, deliver reliable estimates of future demand and assist in coordinating planning, implementation and operation of charging infrastructure.
So far, most planning tools do not include the full spectrum of charging powers available between 2.7 kW and 350 kW, have not considered mid- and long-term time horizons for infrastructure development, and do not have the ability to directly include all the stakeholders in the planning process. Further, they typically do not integrate real-time data from monitoring of existing charging stations, a requirement for any demand-oriented infrastructure planning.
Successful integration and use of charging points also require coordination with on-site implementers, as well as consideration of specific site conditions and information about independent activity in semi-public and private spaces not traditionally gathered by governments and planners.
We introduce a set of quantitative and qualitative methods for producing robust planning guidance based on our experience in the German state of Brandenburg.
The city of Hamburg, Germany committed to buy exclusively emission free buses by 2020. Thus, public transportation companies as Hamburger Hochbahn AG (HOCHBAHN) must build a charging infrastructure for large electric bus fleets. Currently HOCHBAHN is planning an urban charging depot for 240 electric buses (EBs). This pilot project is the first large-scale infrastructure project for EB fleets in Germany. New electrical infrastructures for bus depots must comply with local grid capabilities. Furthermore, they have to fulfill highly individual boundary conditions and operational requirements. In the close future, most public transportation companies will face the challenge of developing electric infrastructures for EB fleets. This paper identifies the key components of electric infrastructures for bus depots based on the introduced concept. It outlines decision objectives, that describe the characteristics of a concept, and confronts them among themselves. The authors apply a sensitivity analysis to evaluate how dimensioning of components affects bus charging times and operation. The developed algorithm in this work uses real data and demonstrates that components can be downsized by 8\%. Furthermore, the method is extended to evaluate needed capacity for varying module sizes of concepts.
For BEV-penetration levels of 5%, 10% and 21% and for charging stations (CS) of 11 and 22 kW the additional transformer loads and voltage drops for this suburban network are calculated.
The worst-case assumption of simultaneous charging of all BEVs in a string of 60 households (HH) is used as a starting point. In this case more than 5% penetration of BEVS will always lead to excessive voltage drop, if charging stations are not clustered close to the feeder of the string.
However, including the statistics of arrival time (with a maximum at 18.00 h) and the duration of individual charging events (as they depend on the statistics of daily driving distance; 70% of cars have driven less than 50 km per day in Germany), the maximum number of simultaneously charging vehicles is reduced significantly with a high confidence. For the example of 60 HH and 21% BEV penetration (which leads to 17 BEVs in the string), the number of simultaneously charging vehicles is decreased from 17 to 6 BEVS with a confidence of 99.7%.
In a final step, the impact of decentralized storage placed at each charging station is considered. With 5 kWh of local battery storage, which is assumed to cover the initial recharging effort, and considering the distribution of driving distances in Germany, only 34% of all BEVs need further recharging from the grid. For the above example of 17 BEVs in a string, a 5 kWh storage leads to a maximum of 3 simultaneously charging vehicles with more than 99.7% confidence. 10 kWh and 15 kWh of battery storage at each charging station leads to a maximum of 2 and 1 vehicle(s) with more than 99.7% confidence, respectively.
Using the same approach, 100% penetration with BEVs s and 15 kWh battery storage at each private charging station, of the 79 vehicles in the above string ouf the network, not more than 3 vehicles will be seen to charge from the grid simultaneously with a confidence of 99,7%.
Thus, it is very important to take the statistics of charging events and average usage of cars into account, to limit network-extension effort to a reasonable amount, while still providing a high service level for the customers.
Battery storage systems are installed in Germany in more than 50% of all new residential PV installation. For residential electricity consumption of 10 to 12 kWh per day, a battery of 4-6 kWh leads to an optimum in terms of profitability (250-280 equivalent full cycles per year). Increasing the battery capacity will lead to lower specific cost per kWh of the battery and higher efficiency of the overall system, but will not lead to sufficient gain in self-consumption in order to pay for the additional invest (a 10 kWh battery leads to approx.180 cycles per year). The battery utilization can be proved, if the battery provides additional services for the network or for the customer.
Large batteries in domestic applications are not fully discharged during evening hours in summer. Hence charging a BEV during evening hours can take advantage of the surplus of stored electricity for large batteries. The conference paper will provide plots of battery utilization (in terms of equivalent full cycles per year) as it depend on PV size, local demand, daily driving distance and charging patterns (during day or evening; slow or fast charging). For the aforementioned scenario (10 to 12 kWh electricity consumption of the household; plus 10 kWh per day for charging of BEV during evening hours) a battery of 10 to 12 kWh will be cycled approx. 280 times per year and is therefore utilized well.
The final analysis of opportunities (saving through own consumption) leads to an indication of the allowed cost for the battery to provide a positive business case. This paper builds on previous publication of ZSW on PV storage systems, self-consumption, level of autarky and allowed battery cost for a positive business case. Those studies have been a result of simulations taking time series of measured yearly solar radiation and consumption as input data. Those results compared well with field test data.
We proposed Watt & Volt / Var control in which EV and PV are cooperated for maintaining voltage in the distribution feeder in the solar integration workshop 2017. In this paper, by Freq / Watt control contributing to the frequency regulation in the TSO level is also installed to in the EV inverter. A smart inverter that suppresses voltage fluctuation and frequency fluctuation by autonomous distributed control is to be successfully realized by the proposed control.
Voltage and power flow profile of a distribution feeder in which massive PVs and EVs are interconnected and global frequency deviation based on the supply and demand calculation are emulated in a real-time simulator at the same time. HIL test is carried out by use of the real-time simulator and two smart inverters of the PV and the EV. On the real-time simulator, Opal-RT, the typical distribution feeder in Japan was modified at 6600V level. The feeder length is set at 5km. 720 houses are interconnected to the feeder. In this paper, it is assumed that PVs and EVs are interconnected to all houses, and all the interconnection inverters have control capability. A thermal power generation, an electric power demand, power output of the PVs based on actual measurement are modified for supply and demand imbalance calculation. Calculated frequency deviation is applied to the modified voltage source of the distribution substation at the top of the distribution feeder. Two different type inverter system, TriphaseNV: PM15 and MITSUBISHI ELECTRIC: Smart V2H, are used as actual smart inverters installing proposed control strategies.
Effectiveness of autonomous distributed control by PVs and EVs on the distribution feeder was confirmed through the HIL tests. And, there ware no interference and unstable phenomena caused by multiple smart inverter control. It is said that smart inverters of PVs and EVs is very effective as control devices for TSO and DSO.
We have developed some V2G control schemes through the HIL(Hardware In the Loop) tests using experimental batteries and stand-alone bi-directional inverters on control response, communication delay in the case of remote control, and power system frequency dynamics.
In this paper, the proposed V2G control schemes is implemented to the actual PEV and the V2G capable charging system as a first V2G system in Japan. Accuracy of the grid frequency and voltage measurements, response of the system, and communication capability, and so on, are verified on two different type PEV system. Effectiveness of the PEV-FFR(Fast Frequency Response) featuring quick response of the PEV and inverter system, PEV-LFC(Load Frequency Control) coordinating large-scale thermal generations, and combination control of the FFR and LFC dispatched to the multiple PEVs are evaluated. PEV-SIR(Synthetic Inertia Response), in which very quick system response in the frequency detection, control initiation, and physical responses is required, is also evaluated by the HIL test.
Overview of the HIL is as follows. The frequency fluctuations under massive PVs and PEVs integration are emulated by the power system real-time simulator (OPAL-RT Technology, OP5600). The power system model is based on a prefecture level with a population of about 9 million people. The power capacity of the PV is 20% of the total electricity demand at peak time, and the number of PEVs is 480,000, these values are target in 2030.
One PEV is Nissan LEAF(30[kWh]), the other is Nissan e-NV200(24[kWh]). These cars are connected to two different bi-directional power conditioners (Nichicon Corporation, NECST-TD1 & Mitsubishi Electric Corporation, SMART V2H System). It is possible to control charging / discharging of 3 [kW] by Nichicon’s one and 6 [kW] by Mitsubishi Electric’s one.
The frequency command value calculated by the real-time power system simulator is transmitted to the power amplifier (California Instruments, MX15, rated: 15 kVA). Then, the power amplifier outputs instantaneous voltages corresponding to the frequency deviations. The PEV controller (dSPACE, Micro Auto Box II) measures the frequency deviations and determines a charge / discharge power command to the PEV. Then the PEVs output the V2G power to the power amplifier via the PEV power conditioners. By feeding back V2G power measurements to the real-time simulator, the frequency fluctuations in the next step can be calculated in the real-time simulator. The FFR, LFC, and SIR control schemes are verified on the PEV and charger system.
In this paper we present a method for conducting grid integration studies in real integrated low- and medium-voltage grid models in the context of eMobility. A special focus is the comparison of a probabilistic distribution approach for BEV charging in low-voltage grids vs. the usage of simultaneity factors. The probabilistic method uses a pool of BEV charging profiles and places them randomly in an LV grid to derive worst-case situations. Necessary grid reinforcement and expansion as well as their cost are then estimated with an automated approach based on a heuristic optimization algorithm. Additionally we compare different charging scenarios like residential charging, public fast charging and dedicated grids for charging stations.
The results show, that the application of simultaneity factors can cause large deviations in regard to violations, as well as necessary grid reinforcement and expansion compared to the probabilistic approach. Especially local weak spots in LV feeders often cannot be identified when using common simultaneity factors for all BEVs in a low-voltage grid. This leads to a possible underestimation of reinforcement and expansion cost. Furthermore the cost of integrating charging infrastructure into the grid varies widely between the different scenarios considered.
The main challenge for many Distribution System Operators (DSOs) when it comes to the integration of Electric Vehicle (EV) charging on their grids is not a problem of energy but rather of power. Critical peak power (CPP), already a difficulty during winter months, is exacerbated by the increasing presence of EV charging stations as the use of electric mobility becomes widespread. As a result, DSOs are showing an increased demand for peak-shaving and peak-shifting technologies.
The project "Coordinating Power Control" ("Växlande Effektreglering" in Swedish, referred to as "VäxEl" in this paper), which began in January of 2017, is based on the increased interest of rural households for solar panels, home batteries, EVs and other "smart home" equipment. The ambition of the VäxEl project is to create a cost-effective optimization of a distribution grid by addressing various technical, regulatory, and psychological challenges.
By bringing together a community of market actors, including smart grid service providers, governmental regulatory bodies, research departments, and a local grid owner (DSO), VäxEl seeks to uncover and propose solutions to such challenges. In May of 2018, the International Smart Grid Action Network (ISGAN), a cooperation within the International Energy Agency (IEA), presented the VäxEl project with its prestigious Award of Excellence for world-leading activities in the area of smart grids for power flexibility.
Through partial funding provided by the Swedish Energy Agency and by assisting homeowners in applying for the economic assistance available to purchasers of EVs, home batteries, and solar arrays, VäxEl has been able to martial a formidable amount of power flexibility, including the installation of 500 connected water-based heating systems, 60 sites with rooftop solar panels (providing 200 kW of production), 36 kW of Electric vehicle charging, and 70 kWh of home energy storage (providing 60 kW of instant power flexibility).
This paper presents some of the progress made within the VäxEl project, primarily focusing on two key aspects: the modeling and design of an optimization algorithm for integrating the resources within the project for the purpose of reducing CPP, and the reduction in CPP achieved during February of 2018 by connected heating systems within the Upplands Energi electric grid.
E-mobility creates new and mobile volatile loads and feeds that will place an additional load on the grid and could require a large investment in network expansion. This makes e-mobility an important driver for forward-looking development of the grid in addition to the energy transition and the implementation of the European Single Market. Like other network users, e-mobility should also contribute to the provision of system services such as balancing power, frequency maintenance and reactive power supply. Smart charging is one of the key capabilities of e-mobility to provide flexibility, for example, by using electric vehicles as temporal energy storage that can feed energy back into the energy system when required.
The integration of e-mobility will occur mostly at low-voltage level. To avoid their overload, charging stations with capacities of 350 kW and over should be connected to the medium or higher voltage grids. If connection is only possible to the low-voltage grid, this type of charging stations should be combined with energy storage. In addition, three-phase charging system are preferred to avoid voltage imbalances in the network at every level. Market incentives or support programs should be based on three-phase charging.
A planning corridor is necessary for the ramp-up of e-mobility. An increasing number of electric vehicles translate into a greater likelihood of simultaneous charging that could cause bottlenecks in the network. Market-driven charging processes can lead to mass effects and create peaks of demand. As found by the meta-study commissioned by VDE|FNN and BDEW, grid-focused charging control is more efficient than market-led charging. Market signals must therefore be coupled with network-relevant parameters to find the most cost-efficient solution.
This paper summarizes current network conditions in Germany, explains the effects on the power grids and outlines the necessary next steps to facilitate e-mobility integration. Key findings from the meta-study commissioned by VDE|FNN and BDEW also provide a description of the critical factors for the integration of e-mobility and offer suggestions for future research.