| With the gradual advancement of the "double-carbon" strategy in China,renewable energy has been connected to the electrical power system on a large scale continuously,and the penetration of new energy has been increasing.Because the output of renewable energy represented by wind power and salor has fluctuations and intermittent,the source-load power of electrical system is often in an unbalanced state.With the growing of emerging loads,the user’s diversified ways of using electricity,and the increasing peak-valley difference of loads also further exacerbate the power imbalance between source and load in the electrical power system.Therefore,the study of coordinated control of Source-Grid-Load-Storage Microgrid systems has gradually become a hot spot of current research.In this paper,a PSO-LSTM algorithm is proposed to solve the problem of load side demand uncertainty in power system.The PSO-LSTM algorithm is used to predict the power load demand.The PSO algorithm is used to optimize the LSTM network’s hyperparameters in the prediction.The weights of the LSTM network structure parameters can be adjusted in time,and the global optimal solution can be found accurately.The example load is predicted,and the superiority of PSO-LSTM prediction model is verified by comparing PSO-LSTM algorithm model with LSTM and BP network algorithm model.In order to solve the problems of BP network such as falling into local extremes easily and slow convergence speed,the gray wolf algorithm is used to optimize the initial parameters of BP network,so as to improve the speed,accuracy of wind output prediction and the optimal dispatching ability of new energy power system.The example is used to predict the output of wind power and photovoltaic,and the accuracy of GWO-BP prediction ability is verified by comparing the GWO-BP algorithm model with the BP network and GA-BP algorithm model.Finally,in response to the problem of coordinated control of source-grid-loadstorage microgrid system,the wind-solar-storage economic dispatch model based on renewable energy utilization and the wind-light-water-fire economic dispatch model based on robustness are constructed,and the objective function and constraints are determined.The grey wolf optimization algorithm improved by dynamic weighting proportional mechanism is used for coordinated control.The coordinated control model is considering whether there is storage energy,whether the wind is abandoned,robustness and time-sharing electricity price,respectively.Through case analysis,the economy of the model and the stability of the multi-energy complementary system are verified. |
