基于改进的BP神经网络负荷预测
Load Prediction Based on Improved BP Neural Network
刘星晨 1袁一平1
作者信息
- 1. 吉林建筑大学,吉林 长春 130000
- 折叠
摘要
合理利用发电资源是减少资源浪费的一种方式,由于电能不能大量储存,需要对未来一段时间的发电量进行合理预测.以BP神经网络为研究对象,针对其容易陷入局部最优的缺点,利用遗传算法和粒子群算法分别对BP神经网络进行优化,建立BP神经网络时间序列预测模型.对比了两种群智能算法优化神经网络对电力系统负荷预测的拟合度和误差,通过Matlab进行实验,结果显示,利用遗传算法和粒子群算法优化的BP神经网络能够减小预测误差.
Abstract
Reasonable use of power generation resources is a way to reduce the waste of resources,according to the characteristics of electric energy cannot be stored in large quantities,it is necessary to make a reasonable prediction of the power generation in the future period of time.Taking BP neural network as the research object,aiming at its characteristics that it is easy to fall into local op-timal,genetic algorithm and particle swarm optimization are used to optimize BP neural network respectively,and BP neural net-work time series prediction model is established.This paper compares the fit degree and error of the optimized neural network of two population intelligent algorithms for the load prediction of the power system,and obtains the results through Matlab software ex-periment.The analysis results show that the BP neural network optimized by genetic algorithm and particle swarm optimization can reduce the prediction error.
关键词
负荷预测/BP神经网络/遗传算法/粒子群算法Key words
BP neural network/genetic algorithm/algorithm optimization/particle swarm optimization引用本文复制引用
出版年
2024
NETL
NSTL
万方数据