基于改进混沌粒子群算法的最小二乘支持向量机短期负荷预测
首发时间:2018-05-15
摘要:针对目前采用最小二乘支持向量机进行电力系统短期负荷预测时,运用混沌粒子群算法选取参数不能完全解决早熟,导致精度下降的问题,本文基于Logistic函数提出一种粒子惯性权重模型对现有混沌粒子群算法权重调节方式进行改进。把提出的惯性权重模型、适应度方差和平均粒矩引入混沌粒子群算法中增强其寻优能力。将改进混沌粒子群算法用于最小二乘支持向量机的电力系统短期负荷预测方法中。最后用华东某地区实际电网的历史负荷数据和气象数据验证方法的可行性和有效性。
关键词: 电力系统 负荷预测 Logistic函数
For information in English, please click here
Short Term Load Forecasting of Least Squares Support Vector Machine Based on Improved Chaotic Particle Swarm Optimization Algorithm
Abstract:The problem of early-maturing can not be completely solved by using chaotic particle swarm optimization algorithm, which leads to the decrease of precision, when using least square support vector machine to do the short-term load forecasting of power system. In this paper, based on the logistic function, a particle inertia weight model is proposed to improve the weight adjustment of the existing chaotic particle swarm optimization algorithm. Then, the proposed inertial weight model, fitness variance and mean particle moment are introduced into the chaotic particle swarm optimization algorithm to enhance its optimization ability. The improved chaotic particle swarm optimization algorithm is used to predict the short-term load of power system based on least squares support vector machine. The feasibility and validity of the method are verified by the historical load data and meteorological data of a real power grid in East China.
Keywords: Power system Load forecasting Logistic function
基金:
引用
No.****
动态公开评议
共计0人参与
勘误表
基于改进混沌粒子群算法的最小二乘支持向量机短期负荷预测
评论
全部评论0/1000