改进粒子群优化RBF神经网络在短期电力负荷预测上的研究

Study on short-term power load prediction based on improved particle swarm optimization RBF neural network

电力负荷预测的准确性对整个电力系统的安全和经济效能起着很大的作用,为提高短期电力负荷预测的准确性,提出一种改进的粒子群优化RBF神经网络的模型。针对PSO算法其迭代后期极易深陷部分最优,收敛准确度低,容易发散等问题,提出了PSO算法自身的特性结合Levy飞行机制算法的特点进行融合,在保障算法的寻优准确度的同时也保障了寻优的速度,从而实现全局最优。利用改进的粒子群算法优化RBF神经网络,再将训练好的RBF神经网络应用到电力负荷的预测中。将此模型应用到黑龙江省某地区短期电力负荷预测中,结果表明此种方法有效提高了预测精度。

The accuracy of power load prediction plays a large role in the safety and economic efficiency of the entire power system.In order to improve the accuracy of short-term power load prediction,an improved particle swarm optimization RBF neural network model is proposed.Aiming at the problems of PSO algorithm that it is easy to sink in the later part of iteration,the convergence accuracy is low,and it is easy to diverge.The characteristics of PSO algorithm combined with the characteristics of Levy flight mechanism algorithm are proposed to ensure the accuracy of the optimization At the same time,the speed of optimizing is guaranteed,so as to achieve the global optimization.The improved RBF neural network is optimized by using the improved particle swarm algorithm,and the trained RBF neural network is applied to the prediction of power load.Applying this model to short-term power load forecasting in a region in Heilongjiang Province,the results show that this method effectively improves the prediction accuracy.