Sequential Load Modeling Of Large-scale Electric Vehicles Plugged In Islanded Microgrid And Its Application In Short-term Reliability Evaluation

Since the increasing exhaustion of fossil energy and the pressure of environmental governance at present,all countries in the world have turned their attention to the promotion and application of renewable energy and electric vehicles(EVs).Islanded microgrid,as a small power system,can effectively integrate renewable energy and traditional units to realize effective control and efficient utilization of renewable energy.As an alternative to traditional fuel vehicles,EVs can effectively reduce the dependence on fossil fuels.However,the contribution of renewable energy sources in islanded micro-grid including wind and solar have the characteristics of strong uncertainty and temporal fluctuation,consequently large-scale EVs access would inevitably lead to the volatility of system net load.Moreover,if EVs have the tendency to charge at peak time of load it would further increase the peak valley load difference which brings huge challenge to the safety and reliability of islanded microgrid.Therefore,it is necessary to consider the schedulability of EV load and guiding the EVs avoid charging at load peak time.Therefore,the time sequence load modeling of the large-scale EVs access scenario is studied and applied in the reliability evaluation.The main research contents are as follows:Combining domestic and overseas research status,the main factors that affect private EVs charging load including battery capacity,charging power and user driving characteristics are analyzed.Based on the Monte Carlo method,daily mileage of EVs is extracted,and disordered charging model is built.Example analysis on different brand of EVs are performed,a 24-hour time-series load curve is established.Also,the effects of battery capacity,charging power and size of EVs on the load curve is analyzed.Considering the randomness of private EVs departure time,mileage and parking duration in work units,and the schedulability of charging and discharging at parking time,the time-of-use price period which worked out according to the net load guides the EVs to charge orderly.Based on the minimum operating cost of micro-grid in next 24 hours,the optimization model of EV time-series load is established.In the example analysis,cutting plane method in GUROBI solver is applied to calculate the optimization model for the next 24-hour EV sequential load,also the influence of the access and permeability ratio of wind and solar,the penetration ratio of EVs and reserve driving capacity on the load optimization results are analyzed,respectively.Based on sequential Monte Carlo method,the random failure and repair process of wind turbines,photovoltaic units and diesel units in one day are extracted,the reliability of islanded micro-grid in the following situations are respectively evaluated and compared: no EV access,EVs charge disorderly,and EVs charge orderly.For reliability evaluation of EVs charge orderly,circumstances of known and unknown unit fault transfer are analyzed.If unit fault transfer is known and one of the units change its status at the certain time,optimization would directly perform to all the time of the day.If unit fault transfer is unknown,the first optimization is performed based on the system status before fault transfer,optimization result of the unit before fault transfer time remains constant,the second optimization is performed based on the system status after fault transfer.The direct influence of wind turbine generation and solar photovoltaic cells failure on time-of-use price is considered in optimization progress.Example analysis shows that relative to disordered charge,EVs using orderly way to charge could effectively reduce the loss of load,load lose frequency and the load lose duration,which significantly improves the reliability of micro-grid.Compared to the situation that units known fault transfer circumstances,the micro-grid reliability level declined slightly in unknown fault transfer circumstances.