模糊C均值结合神经网络的短期电力负荷预测  

Short-term power load forecasting based on fuzzy C-means combined neural network

作　　者：王瑞 杨炜炜 逯静[2] WANG Rui;YANG Weiwei;LU Jing(School of Electrical Engineering and Automation,Henan Polytechnic University,Jiaozuo 454000,China;College of Computer Science and Technology,Henan Polytechnic University,Jiaozuo 454000,China)

机构地区：[1]河南理工大学电气工程与自动化学院,河南焦作454000 [2]河南理工大学计算机科学与技术学院,河南焦作454000

出　　处：《传感器与微系统》2022年第6期138-141,共4页Transducer and Microsystem Technologies

基　　金：河南省科技攻关项目(222102210120);河南高校重点科研项目(18A470012)。

摘　　要：为了提高电力系统负荷预测的预测精度,提出一种模糊C均值(FCM)聚类算法和改进的径向基函数(RBF)神经网络相结合的短期电力负荷预测模型。通过聚类算法对样本进行聚类处理,采用同类特征的数据作为预测输入,强化了样本规律性,提升预测精度。引入模拟退火粒子群优化(SA-PSO)算法来修正神经网络参数,克服了RBF容易陷入局部极小值的缺陷,使输出更接近实际值。利用河南省某县的实际负荷验证该模型的有效性,与传统RBF神经网络相比可有效提高神经网络的学习精度,减少训练次数,具有良好的学习性和适应性。In order to improve the forecasting accuracy of power system load forecasting,a short-term power load forecasting model combining fuzzy C-means(FCM)clustering algorithm and improved radial basis function(RBF)neural network is proposed.The samples are clustered by clustering algorithm,and the data with the same characteristics are used as the prediction input,which improves the prediction accuracy.The simulated annealing particle swarm optimization(SA-PSO)algorithm is introduced to modify the parameters of neural network,which overcomes the defect that RBF is easy to fall into local minimum,and makes the output closer to the actual value.Compared with the traditional RBF neural network,it can effectively improve the learning accuracy of neural network,reduce the number of training,and has good learning and adaptability.

关 键 词：径向基函数神经网络 C均值聚类算法 模拟退火算法 粒子群优化算法 

分 类 号：TP391[自动化与计算机技术—计算机应用技术] TP18[自动化与计算机技术—计算机科学与技术]