Research On Microgrid Optimization Scheduling Considering Uncertainties Of Source And Load

At present,due to the rapid development of society,the reserves of primary energy are getting less and less and the problem of environmental pollution is becoming more and more serious.In this case,distributed generation has been used more and more widely because of its advantages such as reliable operation,low loss and environmental protection.However,the uncertainty and difference of power generation have an impact on power quality and dispatching,and also lead to the reduction of power supply reliability.Thus,the micro-grid came into being.At present,the research on micro-grid mainly focuses on the operation optimization scheduling of internal micro-power for the purpose of improving its economy and environmental protection.In order to achieve this goal,the most important findings of this paper are as follows:(1)The operating principle of micro-power sources commonly used in micro-grid,such as wind power units,photovoltaic power units,micro gas turbines,fuel cells and storage batteries are introduced,and their mathematical models are established respectively.In order to improve the economic and environmental benefits of the micro-grid,a function is established to reduce the total operating cost of the system and the cost of emission pollution treatment.Meanwhile,the output of each micro-power source in the micro-grid is constrained.According to the above content,the operation strategy of micro-grid based on two different modes of on-grid and off-grid is proposed.As a result,a relatively complete micro-grid optimized dispatching system is formed.(2)Research is carried out on the indispensable load forecasting link in the optimal dispatch of micro-grid.The curve similarity comparison method based on fuzzy C-means clustering is used to detect anomalies in the data set and correct the abnormal data.KPCA is used to reduce the dimension of complex feature parameters,and principal component parameters are extracted and integrated into the prediction model.A load forecasting model based on wavelet decomposition is proposed.The prediction data set is divided into low-frequency categories and high-frequency categories by wavelet decomposition,and BAS-BP forecasting model and BAS-LSSVM forecasting model combine the classification data with the principal components to predict separately,and obtain the final forecast load by reorganizing the sequence.Comparing the prediction results of the combined model with those of the two single prediction models,BAS-BP and BAS-LSSVM,it is found that the former is more accurate.(3)An improved beetle swarm optimization algorithm is proposed,which is an improvement of population expansion of beetle swarm optimization algorithm by particle swarm optimization.The idea of grouping chaos is integrated into beetle swarm optimization algorithm.The BSO algorithm is used to search in the global scope,and the TS algorithm is used to search in the local scope,and the Kent chaos mechanism is added to complete the algorithm improvement.The algorithm is verified and compared with PSO,ACO,BSO and GA algorithm.The results prove that the method has better global convergence and local search capabilities.(4)The purposed TS-CBSO algorithm is applied to the micro-grid optimization dispatch and the load demand response model is added.The on-grid and off-grid operation strategies are formulated respectively,and the optimal scheduling of the output of each micro-power in the micro-grid before and after the response under different strategies is made with the goal of realizing the optimal economic and environmental benefits.The final dispatch result can be used as a reference for the micro-grid’s power output in the next day,which can effectively improve the overall benefits of micro-grid operation.