基于IPSO-LSVM的短期负荷预测方法  

Short-term load forecasting method based on IPSO-LSVM

作　　者：程志友[1,2] 丁柏宏 余国晓 CHENG Zhi-you;DING Bai-hong;YU Guo-xiao(School of Electronics and Information Engineering,Anhui University,Hefei 230601,China;Power Quality Engineering Research Center,Ministry of Education,Anhui University,Hefei 230601,China)

机构地区：[1]安徽大学电子信息工程学院,安徽合肥230601 [2]教育部电能质量研究中心,安徽大学,安徽合肥230601

出　　处：《电工电能新技术》2020年第5期41-48,共8页Advanced Technology of Electrical Engineering and Energy

基　　金：国家自然科学基金项目(61672032);安徽省科技重大专项(18030901018)。

摘　　要：传统基于支持向量机(SVM)的短期负荷预测模型存在着输入变量不易确定以及模型参数难以最优化的问题。针对这两个缺点,本文提出了一种基于IPSO-LSVM的短期负荷预测方法。首先引入LASSO回归筛选SVM预测模型中的滞后负荷,将筛选结果和其余影响因素作为SVM预测模型中的输入变量,建立LASSO与SVM的耦合模型(LSVM);然后采用IPSO算法对LSVM预测模型中的参数进行优化,提高预测结果的准确性和稳定性;最后采用优化后的模型进行预测。算例分析表明,IPSO优化LSVM的方法能够有效提高短期负荷预测的精度。Traditional short-term load forecasting model based on SVM has problems in that input variables are difficult to determine and model parameters are difficult to optimize. Aiming at these two shortcomings,this paper proposes a short-term load forecasting method based on IPSO-LSVM. Firstly,the LASSO regression is introduced to select the lagging load in the SVM prediction model,and the screening results and the remaining influencing factors are used as input variables in the SVM prediction model to establish a coupled model of LASSO and SVM( LSVM).Secondly,the IPSO algorithm is used to optimize the parameters in the LSVM prediction model to improve the accuracy and stability of the prediction results. Finally,the optimized model is used for prediction. The analysis of numerical example shows that the IPSO algorithm for LSVM can effectively improve the accuracy of short-term load forecasting.

关 键 词：SVM 短期负荷预测 LASSO 滞后负荷 LSVM IPSO 

分 类 号：TM715[电气工程—电力系统及自动化]