摘要
针对BP神经网络容易出现过拟合和陷入局部最优现象导致电力负荷预测准确性不高的问题,建立了在不同日期和气温等影响因素下的布谷鸟算法(CS)优化BP神经网络的电力负荷预测模型.训练BP网络确定基本结构,运用布谷鸟算法搜寻最优解替换为BP网络的最优参数,结合安徽某地区的电力负荷数据进行仿真分析,可以得出布谷鸟算法优化后的预测模型相比于单一的BP神经网络准确性得到提高,证明了所建立的CS-BP模型具有良好的预测性能.
Aiming at the problem that BP neural network was prone to over fitting and falling into local optimization,resulting in low accuracy of power load forecasting,a power load forecasting model of BP neural network optimized by cuckoo algorithm(CS)under different influencing factors such as date and temperature was established.Firstly,the BP network was trained to determine the basic structure,and then the cuckoo algorithm was used to search the optimal solution and replaced it with the optimal parameters of BP network,the simulation analysis was carried out in combination with the power load data of a region in Anhui Province.It could be concluded that the accuracy of the prediction model optimized by cuckoo algorithm was improved compared with that of a single BP neural network,It was proved that the CS-BP model had good prediction performance.
作者
汪炎
高昕
方亮
WANG Yan;GAO Xin;FANG Liang(School of Electrical and Information Engineering,Anhui University of Science and Technology,Huainan 232001,China)
出处
《哈尔滨商业大学学报（自然科学版）》
CAS
2022年第6期686-691,共6页
Journal of Harbin University of Commerce：Natural Sciences Edition
基金
安徽理工大学博士基金(11127).
关键词
电力负荷预测
布谷鸟算法
BP网络
最优参数
预测模型
预测性能
power load forecasting
cuckoo search algorithm
BP network
optimal parameter
prediction model
predictive performance
作者简介
汪炎(1996-),男,硕士,研究方向:电力负荷预测.
