Research On Charging Scheduling Approach For Voltage Control In Distribution Networks By Electric Vehicle Agents

The development of electric vehicles(EVs)is an important approach for China to get rid of oil dependence,reduce carbon emissions and achieve technological leap in the automobile industry,which has important strategic significance.In recent years,China’s EV number and sales have been rising,and public charging networks in urban and rural areas are gradually being built in the direction of fast charging as the main charge mode and slow charging as a supplement,resulting in a large penetration of EV loads with spatio-temporal dual-dimensional random characteristics and impact into the distribution network system.The charging load during the peak EV charging period can cause considerable nodal voltage deviation,which makes the grid operator starts to pay attention to the impact of future large-scale EV charging on the voltage safety in the distribution network.Although distribution network expansion can solve the voltage safety problem,this measure will bring high construction cost for the grid operator.In addition,the high growth rate of EV load will accelerate the consumption of the distribution network planning margin and increase the frequency of the distribution network expansion.To reduce or postpone investment in the distribution network infrastructure,the grid operator need to establish EV charging scheduling mechanisms and reasonably control the EV charging demand power to ensure the operation safety of the distribution network.In the current research for EV charging scheduling,the voltage problem is mainly taken care of by the centralized scheduling system that relies on global information of the distribution network.However,in practice,the rolling cycle of centralized scheduling is generally over 15 minutes,and such centralized voltage control method will be difficult to guarantee voltage control accuracy under the condition that the total load of the distribution network fluctuates significantly.Considering that there is a considerable amount of distributed energy resources(DERs)(Only photovoltaic system and energy storage system are considered in this dissertation)in the smart grid,adding a collaborative voltage regulation mechanism between the EV and DER will reduce the EV charging power curtailment.Based on the above analysis,it is necessary to study a novel EV charging scheduling method that includes a collaborative voltage regulation mechanism between the EV and DER to improve the voltage control accuracy in the distribution network under the high EV load rate and improve the charging experience of vehicle users.To fill deficiencies of existing EV charging scheduling studies in terms of the voltage control,this dissertation brings a multi-agent cooperative control and a hybrid voltage control framework including centralized scheduling and regional voltage self-management into the study of the EV charging scheduling.Aggregation zone allocation,cooperative-control-based EV agent charging scheduling for zonal voltage control and hybrid distribution network voltage regulation method taking into account EVs and DERs.It provides an important theoretical support and a feasible technical route to establish the EV agent charging scheduling method for voltage control in the distribution network.The main research contents are as follows:(1)Analysis of the impact of EV charging on the nodal voltage in the distribution network.To obtain a sufficiently comprehensive research basis for the subject,the impact of multi-scene EV charging on the distribution network voltage is analyzed.Based on the real data,the EV charging characteristics in typical EV charging stations(Cab charging station,public stations for commercial areas and residential areas)are obtained.An EV load generation method considering the EV charging characteristics and charging dynamic is proposed.Based on this method,the impact of the charging station location,EV charging power level and EV charging demand on the nodal voltage in the distribution network is simulated,which contains the above charging scenarios.(2)Aggregation zone allocation method,which takes into account the uncertainty of load forecast,the density of zonal regulating resources and the regulating capacity.To solve the mismatch problem between the theoretical and realistic voltage regulation capacities of the aggregation zone due to the large load prediction error,the performance index of the aggregation zone is constructed based on the electrical distance and the probability of nodal voltage violations.A novel calculation method for voltage violations is derived to address the contradictory nature of the probabilistic calculation method in terms of assessment reliability and calculation speed.Finally,an aggregation zone allocation algorithm is designed with the objective that the aggregation zone performance index is the biggest,so that the voltage regulation capacity of the assigned aggregation zone can meet the actual voltage regulation demand as far as possible.(3)EV agent charging scheduling for zonal voltage control considering the disconnection fault,output boundaries of EV agents and information delay.Based on the clarification of the physical-information architecture of a proposed voltage control system,the cooperative voltage control strategy is studied in-depth with the research objective of solving the aggregation area low voltage problem.Firstly,the causes of disconnection faults are described,and a communication network topology reconfiguration algorithm is designed to solve the disconnection fault.Secondly,a cooperative control algorithm considering the output boundaries of the EV agent is also designed,and a control model of the voltage control system with information delay is established.Finally,a model discretization condition to ensure the convergence of the voltage control system is proposed,which is demonstrated mathematically.(4)Hierarchical voltage regulation method with EVs and DERs as control units.A hierarchical voltage control framework is introduced for solving the collaborative problem between EV agents and DERs,and their integrated scheduling method of them is studied.First,a centralized optimization model to minimize the distribution network operation cost is established.By costing the scheduling value of regulating resources and penalizing the EV charging load curtailment operation,the centralized scheduling part gives priority to using DERs to regulate the distribution network voltage,and cuts the EV charging load only when the DERs reach their regulating capacity limits and nodal voltages still cannot meet their safety constraints at this moment.Secondly,terminal control algorithms are designed for the photovoltaic system and energy storage system,among which the terminal control algorithm for photovoltaic system can enhance the control ability of photovoltaic system on the high voltage at the PCC point.Finally,a hierarchical energy management system architecture for EV agents and DERs is designed to promote the orderly participation of voltage regulation resources in the scheduling,to bring into play their collaborative voltage regulation capability,improve the voltage safety management level in the distribution network and reduce the EV load curtailment.