嵌入神经网络专家系统的智能化城市评价DSS

Neural Network Expert Systems Integrated Intelligent City Evaluation Decision Support System

本文以决策支持系统（ＤＳＳ）的理论方法为基础，综合运用人工智能、模糊集、神经网络理论．提出了嵌入神经网络专家系统的智能化城市评价决策支持系统（ＩＣＥＤＳＳ）的基本结构模式，介绍了ＩＣＥＤＳＳ各组成部分的功能，讨论了具有神经网络专家系统的综合评价子系统的工作原理，描述了多层次多目标模糊综合评价模型和基于神经网络的系统自学习原理．整个ＩＣＥＤＳＳ的原型系统已在ＩＢＭ微型机上用ＴｕｒｂｏＣ，ＦＯＲＴＲＡＮ７７和Ｆｏｘｂａｓｅ￣＋语言实现．

This paper proposes the fundamental framework of neural network expert systems integrated intelligent city evaluation support sys-tem(ICEDSS),introduces the function of each part of ICEDSS,discusses the working principles of comprehensive evaluation subsystem with neural network expert systems,describes the multilayer and multiobjective fuzzy comprehensive evaluation model and the system self-learning principles based on neural networks.The prototype system of the whole ICEDSS had been implemented on IBM microcomputers in Turbo C,FORTRAN 77,and Foxbase￣+ Languages.