遗传神经网络技术在专家系统中的应用  

Application of Genetic Neural Network Technology in Expert System

作　　者：景广军[1] 梁雪梅[1] 范训礼[1] 

机构地区：[1]南京大学计算机科学与技术系,南京210093

出　　处：《计算机工程》2001年第7期10-13,共4页Computer Engineering

基　　金：国家自然科学基金资助项目(59574034);湖南省自然科学基金项目

摘　　要：讨论了神经网络在专家系统中的应用,提出了神经网络专家系统的设计方法,给出了系统的结构,论述了系统实现中的一些技术问题。分析了遗传算法及神经网络的理论基础,讨论了遗传算法、神经网络与专家系统的结合方式。在专家系统中应用遗传神经网络技术,具有独BP创性,理论和实践上都具有重要意义,进一步丰富了选矿专家系统的研究方法。建立相应的遗传神经网络模型,论述了选矿数据预处理的方法、神经网络的设计,提出了相关参数的经验选取办法。通过实例验证,模型的预测精度可以达到以上。BP90%In this paper, the application of neural network in expert system, the design methods and system structure of neural network expert system are advanced. At last, some technological details for system realization are discussed. Theoretical basis of genetic algorithm and BP neural network andintegration methods of genetic algorithm, neural network and expert system, are discussed in this dissertation. Mineral processing expert system using genetic neural network in the dissertation is an important innovation, with great theoretical and practical significance. It can enrich the research methods on mineral processing expert system. Based on genetic neural network, the idea of building model is advanced, the genetic neural network models for predictionof ore beneficiability, flow-sheet selection and flotation production targets are built. The methods of original data pre-processing, the design of BP neural network, and some parameters selection are studied. Through verification of production examples, the prediction accuracy has achieved over 90%. ;;

关 键 词：专家系统 神经网络 遗传算法 人工智能 选矿 知识库 

分 类 号：TP182[自动化与计算机技术—控制理论与控制工程] TD9[自动化与计算机技术—控制科学与工程]