摘要
为提高协调发展评估模型对电网投资决策过程进行量化评估的精度,在保留以往距离协调发展评估模型全部基础数据处理模块的基础上,使用模糊多列神经网络分析法,对传统欧氏距离协调发展评估模型进行拆分,在将神经网络用于欧氏距离加权系数后,将其与因子归一化结果进行逐一融合,最终将数据输送到传统欧氏距离协调发展评估模型的X值计算模块中。仿真结果表明:因为革新模型对ω因子进行了基于多列神经网络的强化迭代回归数据挖掘,所以二者出现了显著的统计学差异,传统模型的误差率为革新模型的4.32倍;最终X因子的评价结果中,同样存在显著统计学差异,传统模型的误差率为革新模型的5.13倍。
This paper aims to improve the quantitative evaluation precision of the coordinated development evaluation model for power gird investment decision-making process.On the basis of retaining all the basic data processing modules of the evaluation model of coordinated development of Euclidean distance,the traditional evaluation model of coordinated development of Euclidean distance is divided by using fuzzy multi column neural network analysis method.After the neural network is applied to the weighted coefficient of Euclidean distance,it is fused with the factor normalization results one by one,and finally the data is transmitted to the X value calculation module of the traditional European distance coordination.The simulation results show that the innovation model performs enhanced iterative regression data mining based on multi-column neural networks on theωfactor,so there is a significant statistical difference between the two,and the error rate of the traditional model is 4.32 times that of the innovation model.The error rate of the traditional model is 5.13 times that of the innovative model.The improved European distance coordinated development evaluation model has higher data analysis accuracy.
作者
袁天梦
Yuan Tianmeng(State Grid Jibei Electric Power Co.,Ltd.,Tangshan Power Supply Company,Hebei Tangshan,063000,China)
出处
《机械设计与制造工程》
2021年第5期47-51,共5页
Machine Design and Manufacturing Engineering
关键词
欧氏距离因子
协调发展模型
电网投资决策
机器学习
数据偏差率仿真
Euclidean distance factor
coordinated development model
power grid investment decision
machine learning
data deviation rate simulation