基于SVM和差分进化算法的输变电工程数据预测与评估模型设计  

Design of data prediction and evaluation model of power transmission and transformation project based on SVM and differential evolution algorithm

作　　者：沈华强 杨玲 李皓 丁云峰 范殷伟 SHEN Huaqiang;YANG Ling;LI Hao;DING Yunfeng;FAN Yinwei(State Grid Zhejiang Electric Power Co.,Ltd.,Hangzhou 310007,China;Huzhou Power Supply Company,State Grid Zhejiang Electric Power Co.,Ltd.,Huzhou 313000,China)

机构地区：[1]国网浙江省电力有限公司,浙江杭州310007 [2]国网浙江省电力有限公司湖州供电公司,浙江湖州313000

出　　处：《电子设计工程》2022年第22期93-98,共6页Electronic Design Engineering

基　　金：国家电网有限公司财务管理项目(1200-2020016131A-4-34-11)。

摘　　要：针对当前输变电工程数据繁杂且难以有效预测和评估的问题,设计了一种基于SVM和差分进化算法的输变电工程数据预测与评估模型。在对输变电工程项目的影响因素进行分析的基础上,通过随机比特森林算法筛选出输变电工程的主要影响因素并构建相应的数据特征子集,将数据特征子集作为支持向量机(SVM)的输入变量,利用差分进化算法优化SVM的核心参数,从而构建相应的预测与评估模型。将浙江省2012-2020年输变电工程造价实际数据作为数据集进行算例分析,实验结果表明,所提模型对于输变电工程数据预测与评估的平均误差均低于6%,相比于其他常用算法具有显著的优势,能够较为准确地对输变电工程数据进行预测与评估。Aiming at the problem that the current power transmission and transformation engineering data are complex and difficult to predict and evaluate effectively,a power transmission and transformation engineering data prediction and evaluation model based on SVM and differential evolution algorithm is designed.Based on the analysis of the influencing factors of power transmission and transformation project,the main influencing factors of power transmission and transformation project are selected through random bit forest algorithm,and the corresponding data feature subset is constructed.The data feature subset is used as the input variable of Support Vector Machine(SVM),and the differential evolution algorithm is used to optimize the core parameters of SVM,so as to construct the corresponding prediction and evaluation model.Taking the actual cost data of power transmission and transformation projects in Zhejiang Province from 2012 to 2020 as a data set for example analysis and experiment,the results show that the average error of the proposed model for power transmission and transformation project data prediction and evaluation is less than 6%,which has obvious advantages compared with other common algorithms,and can predict and evaluate power transmission and transformation project data more accurately.

关 键 词：支持向量机 差分进化算法 随机比特森林算法 输变电工程 特征子集 

分 类 号：TP391[自动化与计算机技术—计算机应用技术] TN99[自动化与计算机技术—计算机科学与技术]