基于随机森林优化RL的电力工程造价预测

Power project cost prediction based on random forest optimization RL

下载PDF

导出

摘要针对现阶段电力工程造价数据预测准确度较低的问题,提出了一种基于随机森林优化RL的相关预测算法。通过分析影响电力工程造价的因素,将其按照影响程度的大小进行分级,利用随机森林算法加以筛选,并采用筛选后的影响因素特征向量作为预测模型的训练和测试数据。同时,将电力工程造价预测问题转化为输电线路规划问题,再使用强化学习优化蚁群算法的参数来构建电力工程造价预测模型。经过实验对照,综合两种栅格尺寸结果,所提方案比两种对照组算法的电子工程造价分别降低了2.97%、3.78%。 Aiming at the problem of lower accuracy of power project cost data prediction at the present stage,a power project cost prediction algorithm based on random forest optimization RL was proposed.By analyzing the factors that affect the cost of power projects,the random forest algorithm was used to screen the factors according to the degree of impact on the cost of power projects.The filtered factor eigenvector was used as the training and testing data of the power project cost prediction model.The problem of power project cost prediction was transformed to the problem of transmission line planning,and the parameters of ant colony algorithm were optimized by reinforcement learning to build the power project cost prediction model.Through the contrast experiment,in combination with the results for two grid sizes,the power project cost described in this paper is reduced by 2.97%and 3.78%,respectively,compared with the two contrast group algorithms.

作者张文静刘云周波洪崇王立功 ZHANG Wenjing;LIU Yun;ZHOU Bo;HONG Chong;WANG Ligong(School of Electrical and Electronic Engineering,Power and Mechanical Engineering,North China Electric Power University,Baoding 071003,Hebei,China;School of Energy,Power and Mechanical Engineering,North China Electric Power University,Baoding 071003,Hebei,China;Department of Internet,State Grid Hebei Electric Power Co.,Ltd.,Shijiazhuang 050001,Hebei,China;Economic and Technological Research Institute,State Grid Hebei Electric Power Co.,Ltd.,Shijiazhuang 050001,Hebei,China;School of Electrical and Automation,Wuhan University,Wuhan 430072,Hubei,China;Department of Software Cost,Hebei SECPT Computer Consulting Service Co.,Ltd.,Shijiazhuang 050081,Hebei,China)

机构地区华北电力大学电气与电子工程学院华北电力大学能源动力与机械工程学院河北省电力有限公司互联网部河北省电力有限公司经济技术研究院武汉大学电气与自动化学院河北赛克普泰计算机咨询服务有限公司软件造价部

出处《沈阳工业大学学报》 CAS 北大核心 2024年第6期754-759,共6页 Journal of Shenyang University of Technology

基金河北省自然科学基金项目(F2021210005) 国网河北省电力有限公司科技项目(5204JY22000L)。

关键词输变电工程工程造价随机森林强化学习蚁群算法输电线路规划数据预测特征向量 power transmission and transformation project project cost random forest reinforcement learning ant colony algorithm transmission line planning data prediction eigenvector

分类号 TM726 [电气工程—电力系统及自动化]