Research On Resources Prediction And Optimization Scheduling Of Micro Grid

Growing energy consumption was created by the progress of human society,energy depletion and environmental pollution have become two big problems of the world because of the use of fossil energy.Inexhaustible renewable energy sources have become a necessary way to solve the problem of energy depletion and environmental pollution,are one of the newest concerns for countries and research institutions.Renewable energy sources,storage,and load are connected by advanced power electronic technology in micro grid,and the problem of generation randomness,volatility and difficult to control in using renewable energy sources are solved by micro grid.The inherent complexity of micro grid is a big challenge for existing control strategy,so it will become the trend of the development of the micro grid in the future by using intelligent algorithm in resources prediction and optimization scheduling,and the research of this two problems will have vita significance.The key point in this thesis is to solve the problem of resources prediction and optimization scheduling by using intelligent algorithm.Analyzing and modeling the overall framework of the micro electric system and part of it.Analyzing characteristics of uncertainties in micro grid such as solar power,wind power and load.And using adaptive clustering algorithm in RBF neural network algorithm is verified by simulation,and predicting uncertainties by using this algorithm.At the same time,the prediction simulation results of solar power,wind power and load will be the actual example of this algorithm in actual usage.The objective function,constraint condition and scheduling policy of optimization scheduling are determined,and the self-adaptive mutation particle swarm optimization which have been verified through the optimization of solving typical problems is put forward by using the thought of self-adaptive and mutation.The optimization scheduling of micro grid system which is running in a different mode of operation optimization is solved by using self-adaptive mutation particle swarm optimization algorithm.Using the principle of the above,writing algorithm in the MATLAB,the simulation results of interconnection and islands modes aregiven respectively,determine the accuracy of the algorithm through the analysis of the results.The main work point lies in adding the ideas of modern artificial intelligence to the control of micro grid in this thesis.Using adaptive clustering algorithm in order to improve the RBF neural network algorithm in clustering analysis,and the improved RBF neural network algorithm is applied to the micro grid resource prediction;Overcoming the slow convergence and shortcoming of local optimum easily in particle swarm optimization by using the idea of adaptive and mutation,using adaptive mutation particle swarm optimization algorithm to complete the micro grid scheduling optimization.Improving the previous micro grid control algorithm and providing better solutions for solving the problems of micro grid resource prediction and optimization scheduling by using these algorithms.