基于Spearman-IPSO-LSSVM的短期电力负荷预测方法研究  

Research on Short-term Power Load Forecasting Method Based on Spearman-IPSO-LSSVM

作　　者：赵宇庆 腾志军 Zhao Yuqing;Teng Zhijun(School of Electrical Automation and Information Engineering,Tianjin University,Tianjin 300072,China;College of Electrical Engineering,Northeast Electric Power University,Jilin Jilin 132011,China)

机构地区：[1]天津大学电气自动化与信息工程学院,天津300072 [2]东北电力大学电气工程学院,吉林吉林132011

出　　处：《电气自动化》2025年第1期102-104,108,共4页Electrical Automation

摘　　要：为提高地区用电负荷预测的精度,使用斯皮尔曼相关系数法,计算出地区天气各特征因素和用电负荷的相关性大小,并选择相关性大的因数作为最小二乘支持向量机模型的输入向量。为克服最小二乘支持向量机算法模型对核函数和惩戒参数的敏感性,提高算法的泛化能力,引入一种改进的粒子群优化算法对最小二乘支持向量机模型相关参数进行寻优。最后以某地区电力负荷实际历史数据为算例,对某日负荷进行预测。结果表明,所提算法对地区短期电力负荷预测具有较好的预测精度和使用价值。To improve the accuracy of regional electricity load forecasting,the Spearman correlation coefficient method to calculate the correlation between various weather characteristics and electricity load in the region was used,and the factor with high correlation as the input vector of the least squares support vector machine model was selected.To overcome the sensitivity of the least squares support vector machine algorithm model to kernel functions and penalty parameters,and improve the generalization ability of the algorithm,an improved particle swarm optimization algorithm to optimize the relevant parameters of the least squares support vector machine model was introduced.Taking the actual historical data of power load in a certain region as an example,the load on a certain day was predicted.The results show that the algorithm proposed has good prediction accuracy and practical value for short-term power load forecasting in the region.

关 键 词：用电负荷 斯皮尔曼相关系数 最小二乘支持向量机 改进粒子群算法 预测精度 

分 类 号：TM74[电气工程—电力系统及自动化]