可靠性知识与最优评估配电网故障恢复(二)最优评估  

FAULT SERVICE RESTORATION FOR DISTRIBUTION NETWORKS BASED ON THE COMBINATION OF RELIABLE KNOWLEDGE AND OPTIMUM EVALUATED MODEL Part Two Optimum Evaluated Model

作　　者：孙雅明[1] 杜红卫[1,2] 

机构地区：[1]天津大学自动化学院 [2]国电自动化研究院,江苏省南京市210003

出　　处：《电力系统自动化》2003年第13期36-39,53,共5页Automation of Electric Power Systems

摘　　要：对配电网故障恢复 (FSR)决策解集最优评估指标进行了模糊性研究 ,形成前向神经网络(FNN)的输入矢量 ;采用统一、规范化准则 ,对各优化目标进行等级划分并进行不同的组合 ,形成FNN的学习样本集 ,其输出即决策解的综合性能的评估值。不仅可以准确地获得最优解 ,同时可以依据最优评估值的大小来估计该解的总体性能的好坏。仿真结果表明 。This paper is about optimum evaluation of decision-making solving set for fault service restoration (FSR). The optimum evaluation indices of decision-making solving set are studied based on fuzzy theory, and the input vectors of feedforward neural network (FNN)are formed. According to the unified and standardization criterion, the optimal objectives are graded and combined to form learning sample set for FNN model, and its output is the integrated evaluated value. Not only the optimal solution can be ensured, but also the optimal value can be used to estimate the integrated performance of the solution. The simulation results show that the true optimal FSR solution can be obtained according to the principle proposed in the paper.

关 键 词：故障恢复 最优评估 前向神经网络 模糊理论 

分 类 号：TM711[电气工程—电力系统及自动化]