基于改进混沌搜索的AMPSO-BP短期负荷预测

Short-term Load Forecasting Based on Improved Chaotic Search AMPSO-BP Neural Network

为了精准预测微电网短期负荷,采用模糊聚类方法选择相似日粗集,用灰色关联分析法选取相似日,并针对神经网络易陷入局部极小值的缺点,提出基于混沌搜索的自适应变异粒子群优化算法(AMPSO)获得神经网络最佳参数,建立AMPSO-BP神经网络短期负荷预测模型。对收集的电网数据进行试验仿真结果显示,所提方法有很高的预测精度和稳定性,在实际中有一定的应用价值。

In order to accurately predict the short-term load of the microgrid,a fuzzy clustering method is used to select a rough set of similar days,and a gray correlation analysis method is adopted to select similar days.Aiming at the characteristics of the neural network easily falling into the local extremum,chaotic search based adaptive mutation particle swarm optimization algorithm(AMPSO)is proposed to obtain the optimal parameters of the neural network and establish the short-term load forecasting model based on the AMPSO-BP neural network.Finally,the experimental simulation is carried out with the collected grid data.The results show that the proposed method has high prediction accuracy and stability,and it has certain application value in practice.