Short-term Power Load Forecasting And Optimal Scheduling Of Microgrid Methods Research

With the accelerating process of industrialization in modern society,a large amount of fossil energy is consumed,and human society is facing a serious crisis of energy depletion and environmental degradation.For the sustainable development of human society,countries around the world are advocating the search for new energy sources to replace traditional energy sources.Therefore,as a small power generation and distribution system,microgrid is regarded as the key research object of national energy sustainable development by scholars all over the world with its characteristics of efficient utilization of clean energy power generation.Reasonable optimal dispatching can effectively guarantee the economy,environmental protection and stability of microgrid operation,and reliable load forecasting data is the most powerful basis for formulating optimal dispatching.Therefore,the accuracy of microgrid load forecasting directly determines the feasibility of dispatching scheme.In this context,this paper studies the short-term load forecasting and optimal scheduling of microgrid.The main research contents are as follows:Firstly,aiming at the problem that the traditional load forecasting model has poor performance in processing time series modeling,a TCN-Attention short-term load forecasting model is proposed.Since most of the load forecasting models are based on Long Short-Term Memory(LSTM),this paper takes Temporal Convolutional Networks(TCN)as the basic model.Therefore,the origin and structure of LSTM network and TCN network are analyzed in detail,and the TCN prediction model and LSTM prediction model are built for comparative analysis.The simulation results show that the TCN model has higher prediction accuracy.In order to further improve the accuracy of the TCN model,in view of the fact that the Attention mechanism adaptively allocates weights according to the importance of features,a TCNAttention short-term load forecasting model is constructed and simulated.The results show that the accuracy of the TCN-Attention model is improved by 0.49%compared with the single TCN model,reaching 96.42%.Secondly,aiming at the problem that the load data information is mixed and the change rule is not obvious,a data decomposition method based on WOA-VMD is proposed.The minimum envelope entropy of the variational mode decomposition(VMD)algorithm is used as the fitness function of the whale optimization algorithm(WOA).Through iterative optimization,the optimal parameter combination of the modal number k and the quadratic penalty factor a of the VMD algorithm is obtained and substituted into the VMD algorithm.The original one-dimensional load data is decomposed into multiple Intrinsic Mode Functions(IMF)with different frequencies but regular changes,and each IMF component is used as the input load feature of the TCN-Attention model.After that,the above two models are combined.At the same time,in order to further improve the prediction accuracy of the model,Pearson correlation coefficient is used to screen out the daily maximum temperature,minimum temperature and average temperature as meteorological feature input,and the date type factor is reconstructed as the date type feature input.Finally,based on all the above methods,a shortterm load forecasting method based on WOA-VMD-TCN-Attention model is proposed.Through experimental simulation,the results show that the MAE,RMSE,MAPE and R2 indexes of the proposed model are 87.11 MW,111.16 MW,1.01%and 99.26%respectively.Compared with other models,the prediction accuracy of the proposed model is the highest.Finally,an improved adaptive genetic algorithm is proposed to solve the problem that it is difficult to obtain an economic optimal scheduling scheme for microgrid in grid-connected mode.At the same time,considering the time-of-use electricity price of the large power grid,the charging and discharging cost of the energy storage battery and the operating constraints of each distributed power supply,an optimal scheduling model with the lowest daily operating cost as the objective function is constructed.Combined with the real parameters of the microgrid,the genetic algorithm,the adaptive genetic algorithm and the improved adaptive genetic algorithm are used respectively.The simulation results show that the daily operating cost obtained by this method is the lowest,which proves the effectiveness of the improved method in this paper.