基于支持向量机的变电站工程数据预测方法研究  

Research on substation engineering data prediction method based on support vector machine

作　　者：薛礼月 刘鑫 黄亦章 阚竟生 XUE Liyue;LIU Xin;HUANG Yizhang;KAN Jingsheng(Economic and Technology Research Institute of State Grid Shanghai Electric Power Company,Shanghai 200002,China)

机构地区：[1]国网上海市电力公司经济技术研究院,上海200002

出　　处：《电子设计工程》2021年第20期147-151,共5页Electronic Design Engineering

基　　金：国网上海市电力公司2019年度科技创新项目(B3090R190000)。

摘　　要：针对变电站工程各项数据的计算过程复杂、涉及专业较多、效率较低的问题,文中综合利用现代智能算法,将粒子群算法与支持向量机算法相结合并应用到变电站工程数据预测中,以实现在短时间内相对准确地预估出变电站工程的相关数据信息。文中基于变电站三维Revit软件平台,提取出工程历史样本数据,采用粒子群算法优化了支持向量机模型的输出权值和隐层阈值。在选取合适的激活函数和隐藏层神经元个数后,能够在变电站工程开展的前期阶段,准确地预测估计出各项工程数据,为变电站工程决策阶段提供参考。Aiming at the complicated calculation process of various data of substation engineering,involving more specialties,and low efficiency,the article comprehensively uses modern intelligent algorithms,combines particle swarm algorithm and support vector machine algorithm,and applies it to substation engineering data prediction.In a short time,the relevant data of the substation project can be estimated relatively accurately.Based on the substation 3D Revit software platform,this paper extracts the historical sample data of the project,uses the particle swarm algorithm to optimize the output weight and hidden layer threshold of the support vector machine model.After selecting the appropriate activation function and the number of hidden layer neurons,in the early stage of substation project development,various engineering data are accurately predicted and estimated to provide reference for the decision-making stage of substation engineering.

关 键 词：变电站 SVM 三维 REVIT 工程数据 

分 类 号：TP181[自动化与计算机技术—控制理论与控制工程]