Research On Short-term Load Forecasting And Scheduling Optimization Of Micro-grid

At present,the rapid development of society has caused problems such as energy crisis and environmental pollution,and the required electricity is also rising.Micro-grids that use clean energy to generate electricity can not only alleviate energy shortages and environmental pollution pressures,but also serve as an effective complement to large power grids.However,the biggest drawback of micro-grid power generation is that it has strong randomness and it is difficult to supply power stably and reliably,also has difficulties in scheduling on the grid.It is of great theoretical value and engineering value to study short-term load forecasting and optimal scheduling technology of micro-grid.Taking the actual operation of the wind and solar storage grid as the research object,the short-term variation law of the load was analyzed to determine the dominant factors affecting the load change.Previous studies had shown that BP neural network was very suitable for solving load forecasting problems,but its existence was easy to fall into the local extremum.In this case,the Particle Swarm Optimizition with Adaptive Mutation(AMPSO)algorithm was proposed to optimize BP network parameters,with multi-peak function,the excellent search ability of the algorithm was verified.The AMPSO-BP load forecasting model was established.The network input and output were determined according to the load characteristics.The neural network was used to train the neural network and load forecasting,and compared with BP and PSOBP.The comparison results show that the AMPSO-BP method predicts the maximum relative error and the average relative error level is lower than the latter two methods,showing good accuracy and stability,which proves to be an effective short-term load forecasting method.Based on the short-term load forecasting results,the micro-grid optimization scheduling strategy was studied.The micro-grid power was preferentially utilized,and the net load was supplied by the large grid and the battery.In the algorithm,the SOC value of the storage was limited,and the operating cost of the system was minimized,and the distributed power generation was reasonably arranged.The micro-grid optimization scheduling objective function was established,and the objective function was solved by AMPSO algorithm.The simulation results of MATLAB show that the clean energy is fully utilized,the charging and discharging time of the battery is reasonable,and the operating cost is at a low level.The scheduling scheme meets the actual requirements,which proves that the proposed scheduling scheme has certain theoretical value and practical application value.Figure23;Table12;Reference 59...