基于DEA和遗传BP神经网络的电网技术改造造价预测  

Cost Forecast of Power Grid Transformation Project Based on DEA and Genetic BP Neural Network

作　　者：董祯 王艳芹 王勇 赵贤 容春艳 聂婧 DONG Zhen;WANG Yanqin;WANG Yong;ZHAO Xian;RONG Chunyan;NIE Jing(State Grid Hebei Electric Power Co.,Ltd.,Shijiazhuang 050021,China;State Grid Hebei Electric Power Co.,Ltd.Economic and Technical Research Institute,Shijiazhuang 050021,China)

机构地区：[1]国网河北省电力有限公司,石家庄050021 [2]国网河北省电力有限公司经济技术研究院,石家庄050021

出　　处：《河北电力技术》2020年第1期25-29,共5页Hebei Electric Power

摘　　要：针对电网工程技术改造项目工程造价的重要性,分析技术改造造价预测可使用的模型,提出了一套基于数据包络分析(DEA)和遗传BP神经网络模型的电网工程技术改造项目的组合预测方法。该方法利用DEA筛选输入指标和样本数据,做到数据层面的优化,通过遗传算法(GA)的全局寻优能力,寻找并优化神经网络的权值和阈值,利用神经网络(BPNN)的局部寻优能力强的特性,快速收敛至最优解,获得技术改造项目造价预测的能力,仿真算例证明,组合预测模型准确地反映了实际工程造价,对电网工程技术改造项目的造价预测具有参考价值。In order to fill the research blank of the cost forecast of technological transformation project,this paper proposes a Combination forecasting model of technological transformation project of power grid based on Data Envelopment Analysis(DEA),Genetic Algorithm(GA)and Back Propagation Neural Network(BPNN)model.First of all,the input indicators and sample data is filtered by DEA for optimization in data level.Then,through the global optimization ability of Genetic Algorithm(GA),the weights and thresholds of BP neural network are searched and optimized by using GA.Finally,through the powerful local optimization ability,BPNN will converge to the optimal solution rapidly for cost forecast of technical transformation projects.The simulation results show that the combined model can reflect the actual cost of the project accurately,and there is a certain reference value to the cost forecast of the power grid project.

关 键 词：电网工程 技术改造 造价预测 数据包络分析 遗传算法 人工神经网络 

分 类 号：TM744[电气工程—电力系统及自动化] TP39[自动化与计算机技术—计算机应用技术]