摘要
针对海量电力工程数据处理与价值挖掘问题,文中开展了机器学习算法在电力工程数据识别与处理中的应用研究,并提出了基于GRA-PSO-SVM的电力工程静态造价预测算法。该算法利用灰色关联度分析(GRA)筛选出静态工程造价的主要影响因子,且采用粒子群(PSO)算法对支持向量机(SVM)的惩罚系数与核函数参数加以优化,以提高SVM算法的收敛速度和计算精度。同时通过引入SVM算法对主要影响因子数据进行识别与处理,实现了电力工程静态造价的精准预测。仿真分析结果表明,所提算法显著优于PSO-SVM及SVM等单一算法,且对静态造价预测的平均误差仅为4.62%。
Aiming at the problem of massive power engineering data processing and value mining,this paper studies the application of machine learning algorithm in power engineering data recognition and processing.A static cost prediction algorithm of power engineering based on GRA-PSO-SVM is proposed.The algorithm uses Grey Relation Analysis(GRA)algorithm to screen the main influencing factors of static engineering cost.Particle Swarm Optimization(PSO)algorithm is used to optimize the penalty coefficient and kernel function parameters of Support Vector Machine(SVM)algorithm,which improves the convergence speed and calculation accuracy of SVM algorithm.By introducing SVM algorithm to identify and process the data of main influencing factors,the accurate prediction of static cost of power engineering is realized.The simulation results show that the proposed algorithm is significantly better than PSO-SVM and SVM,and the average error of static cost prediction is only 4.62%.
作者
徐鑫乾
何宏杰
张华
张可抒
XU Xinqian;HE Hongjie;ZHANG Hua;ZHANG Keshu(State Grid Jiangsu Electric Power Co.,Ltd.,Nanjing 210024,China;Economic Research Institute,State Grid Jiangsu Electric Power Co.,Ltd.,Nanjing 210008,China;Changzhou Changgong Electric Power Design Institute Co.,Ltd.,Changzhou 213001,China)
出处
《电子设计工程》
2023年第22期134-138,共5页
Electronic Design Engineering
基金
国网江苏省电力有限公司管理咨询项目(B710K0223J4S)。
关键词
人工智能
电力工程
数据处理
智能感知
灰色关联度分析
artificial intelligence
power engineering
data processing
intelligent perception
Grey Rel-ation Analysis
