Optimal Dispatching Of Regional Power Grid Considering The Difference Of Electric Vehicle Demand

The large-scale grid connection of wind power generation and electric vehicles has greatly increased the pressure on power system dispatching and operation.the current power system dispatching needs to consider not only the uncertainty of wind power output.it is also necessary to consider the characteristics of load volatility and randomness of large-scale electric vehicles.The efficient and reasonable optimal dispatching strategy of regional power grid energy management can take into account the demand differences of electric vehicles and improve the economy of regional power grid dispatching through the participation of electric vehicles in regional power grid dispatching.Due to the uncertainty of wind power generation and the fluctuation of load,it is necessary to forecast wind power generation and load demand on multitime scale,while due to the influence of environmental factors such as meteorology,temperature,humidity and travel demand,there are errors in wind power output forecasting and load demand forecasting.In order to eliminate the influence of prediction error on regional power grid dispatching and realize the economy of regional power grid intra-day dispatching,this paper studies the regional power grid optimal dispatching strategy considering the difference of electric vehicle demand.Firstly,the load model of electric vehicle is introduced.Electric vehicle aggregator(Electric vehicle aggregator,EVA)has the ability to coordinate the charge and discharge of large-scale decentralized electric vehicles based on EVA charging pile equipment,V2 G technology,vehicle networking technology,load forecasting technology and intelligent decision-making technology.In order to make the regional power grid dispatch electric vehicles more effectively,the intelligent charging pile collects the information of electric vehicles,and the data information of each electric vehicle is expressed as an one-dimensional state matrix.after classifying electric vehicles according to the wishes of users,the operation characteristics of electric vehicles are analyzed,different types of electric vehicles are extracted from the state matrix of electric vehicles by Monte Carlo method,and the load capacity range of all kinds of electric vehicles is predicted.Secondly,a multi-objective optimization model considering the schedulability of electric vehicles is designed.Firstly,the regional power network dispatching model is introduced.on the basis of considering the difference of electric vehicle dispatchability and time-of-use electricity price,a multi-objective regional power grid dispatching model is established with the goal of minimizing unit operation cost and electric vehicle owner paying cost.the multiobjective function is transformed.And through the simulation analysis of different types of electric vehicles connected to the regional power grid according to different proportions of the impact on the regional power grid unit operating costs,electric vehicle owners pay fees.Finally,a regional power grid economic dispatching strategy based on long-term and shortterm memory neural network(long short term memory network,LSTM)is proposed.On the basis of multi-time scale scheduling,the strategy adds the model training phase between the day-ahead scheduling phase and the intra-day scheduling phase.In the day-ahead stage,considering the unit operation cost of the regional power grid and the fees paid by the electric vehicle owners,a multi-objective optimal scheduling model is established,which outputs the dispatching results of the controllable units in the day-ahead phase and determines the multiobjective optimization weights of the model training phase and the intra-day dispatching phase.in the model training phase,the intra-day dispatching process of the regional power grid is simulated,and the intra-day dispatching model is obtained by training the LSTM network with a large amount of data generated.In the intra-day phase,the day-ahead scheduling results and the ultra-short-term prediction data of the intra-day controllable unit are input into the intra-day scheduling model,and the intra-day controllable unit scheduling plan is obtained.In order to verify the effectiveness and economy of the scheduling strategy,the off-line future scheduling is carried out according to the actual wind farm output power and load demand after the completion of the intra-day dispatching phase.the simulation results verify the effectiveness and economy of the proposed strategy.