基于改进PSO的神经网络短期电力负荷预测模型  

Short-term Power Load Prediction Model Based on Improved PSO Neural Network

作　　者：任东红 林鹏 袁清萍[1] 邢兵锁[1] 董国贵[1] REN Donghong1 , LIN Peng2 , YUAN Qingping1 , XING Bingsuo1 , DONG Guogui1(1. Dept. of Electrical Engineering, Tongling Polytechnic, Tongling 244061, China; 2. Tongling Power Supply Company of State Grid, Tongling 244000, China)

机构地区：[1]铜陵职业技术学院电气工程系,安徽铜陵244061 [2]国网铜陵供电公司,安徽铜陵244000

出　　处：《淮海工学院学报（自然科学版）》2018年第2期26-30,共5页Journal of Huaihai Institute of Technology:Natural Sciences Edition

基　　金：安徽省教育厅高校自然科学研究重点项目(KJ2017A632);铜陵职业技术学院科研项目(KZ2016B001)

摘　　要：针对短期电力负荷预测问题,提出了一种基于改进粒子群算法(PSO)的BP神经网络预测模型.选取与电力负荷大小密切相关的季节、天气情况、当天最高温度、当天最低温度、日期性质、上年同期历史数据为神经网络的输入变量,电力负荷大小为输出变量,建立基于BP神经网络的电力负荷预测模型,同时利用改进的粒子群算法对神经网络进行训练.该训练算法具有较高的精度,能够建立较精确的短期电力负荷测量模型.仿真结果表明,基于改进粒子群算法的BP神经网络电力负荷预测模型预测精度较高,具有一定的应用价值.BP neural network based on improved particle swarm optimization（PSO）is proposed for short-term power load prediction.The season,weather,the highest temperature of the day,the lowest temperature of the day,date nature and historical data which are closely related to short-term power load are selected to be the inputs of the neural network,power load of the day is the output of the neural network.At the same time,the improved PSO is used for training the new neural network.The proposed algorithm has higher precision and a more precisely measurement model for short-term power load can be established.The simulation results show that BP neural network based on improved PSO algorithm predict short-term power load more accurately,which has certain applicability.

关 键 词：电力负荷 预测模型 BP神经网络 神经网络输入变量 改进粒子群算法 

分 类 号：TP183[自动化与计算机技术—控制理论与控制工程]