基于PCA和改进AHP的输变电工程云边协同造价分析  

Cloud edge collaborative cost analysis of power transmission and transformation projects based on PCA and improved

作　　者：宋远 路瑶 李昊 赵振喜 郭晓丹 AHP SONG Yuan;LU Yao;LI Hao;ZHAO Zhenxi;GUO Xiaodan(School of Electrical Engineering,Northeast Electric Power University,Jilin 132011,Jilin,China;Construction Department,State Grid Jilin Electric Power Supply Company,Changchun 130012,Jilin,China;Economic Technology Institute,State Grid Jilin Electric Power Supply Company,Changchun 130012,Jilin,China)

机构地区：[1]东北电力大学电气工程学院,吉林吉林132011 [2]国网吉林省电力有限公司建设部,吉林长春130021 [3]国网吉林省电力有限公司经济技术研究院,吉林长春130021

出　　处：《沈阳工业大学学报》2025年第2期190-196,共7页Journal of Shenyang University of Technology

基　　金：吉林省科技计划项目(J2020RCDT2B0485);国网吉林省电力有限公司专项课题(SGJLJY00JJJS2000060)。

摘　　要：【目的】随着降本增效需求的日益迫切和电力工程管控精细化要求的不断提高,输变电基建项目在智慧基建背景下对多要素、全过程、精细化造价管理的要求日趋显著。传统工程造价方法已无法满足智慧基建项目全生命周期成本最优化的需求,因此迫切需要由单一造价控制向智慧型造价管理转型。【方法】为提升输变电工程海量数据下造价管理的精细化水平,实现全过程覆盖,在深入研究现有工程造价管理模式的基础上,针对传统方法难以精准分析智能电网基建中复杂且海量数据的问题,提出了一种基于主成分分析(PCA)法和层次分析(AHP)法的输变电工程云边协同造价分析方法。该方法设计了基于云边协同架构的输变电工程造价系统,构建了包含目标层、准则层和方案层的3层造价因素评价指标体系,用于评价造价影响因素。在边缘计算中心,通过PCA对海量工程造价数据进行降维处理,并上传至云中心;采用粒子群优化(PSO)算法优化AHP的评价指标权重,有效减小传统AHP的主观偏差,并利用优化后的AHP在云中心完成工程造价的可靠计算。【结果】基于选取的输变电工程造价数据进行实验分析,结果表明:PSO算法在45次迭代后完成AHP参数优化,具备较快的寻优速度和较高的准确性。与其他造价方法对比,本文方法的工程造价计算值与实际值误差最小,仅为4.16%,且整体误差小于7%,显著优于对比方法。【结论】本文方法基于云边协同架构,充分利用边缘计算结合PCA实现海量数据的高效降维预处理,优化后的AHP-PSO算法在合理的评价指标权重下实现更小的分析误差和更高的评价可靠性,有效满足智慧基建工程对全过程、全要素造价精细化管理的要求,为智能电网基建项目的造价优化管理提供了有效解决方案。[Objective]With the urgent need for cost reduction and efficiency improvement,as well as the increasing demand for refined control in power engineering,the importance of multi-element,whole-process,and refined cost management in infrastructure projects is becoming increasingly evident in the context of smart infrastructure.Traditional project cost methods cannot meet the requirements of optimizing the full life cycle cost of smart infrastructure projects.Simple cost management mode must shift to intelligent cost management mode.[Methods]To further improve the refinement level of cost management under the massive data of power transmission and transformation projects and achieve the goal of covering all stages of the entire process,a cloud edge collaborative cost analysis method for power transmission and transformation projects based on principal component analysis(PCA)and improved analytic hierarchy process(AHP)was proposed after in-depth research on the existing project cost management mode,which could address the problem that traditional project cost analysis methods were difficult to accurately analyze complex and massive data in smart grids projects.This method effectively improved the refinement level of project cost management.The proposed method first designed a targeted cost system for power transmission and transformation projects based on cloud edge collaborative architecture.A three-layer cost factor evaluation index system was developed,including a target layer,a criterion layer,and a scheme layer,which was used to evaluate the cost impact indicators.In the edge computing center of the system,the PCA was used to reduce the dimension of massive project cost data and upload it to the cloud center.The particle swarm optimization(PSO)algorithm optimizes the weights of evaluation indicators in the AHP,which effectively eliminates the subjective bias of the original AHP.In the cloud center,the optimized AHP was used to achieve reliable calculation of project cost.[Results]With the cost data of the selected

关 键 词：输变电工程 云边协同 工程造价分析 主成分分析法 粒子群优化算法 层次分析法 电网智能化 云计算 

分 类 号：TM726[电气工程—电力系统及自动化]