OBPES通用骨架系统的整体结构

来源期刊：中南大学学报(自然科学版)1997年第3期

论文作者：罗首章 李松仁 陈松乔

文章页码：222 - 225

关键词：预测； 专家系统； 矿石可选性； 通用骨架系统； 推理机

Key words：prediction； expert system； ore benificiability； shell system； inference engine

摘    要：矿石可选性预测专家系统(OBPES)通用骨架系统是选矿领域专家系统的构造工具，是用C^＋＋语言在Windows平台上开发的．本文分析了OBPES骨架系统的可行性，论述了 系统的结构及各组成部分的主要功能、系统的工作原理及应用步骤．系统采用面向对象的知识库，知识自学习包括基于解释的学习及神经网络自学习，推理机制封装于知识类之中，通过特性继承和消息传递来实现．OBPES骨架系统的功能齐全、界面十分友好，将会在选矿方面得到实践的检验，并不断完善．

Abstract: Ore beneficiability prediction expert system (OBPES) shell system is a construction tool for expert systems in the field of ore beneficiability prediction, which is developed on Microsoft Windows platform using C⁺⁺ language. The feasibility of OBPES shell system was analyzed. The structure of the system, main function of each parts, working principle and application steps were discussed in the paper. The knowledge base based on object is used by the system. Automatic knowledge acquisition includes learning based on explanation and neural network learning. The inference mechanism is encapsulated in knowledge class through characteristic inheritance and message transferring. The OBPES shell system has complete functions and friendly interface. The system will be further tested, verified and improved continuously in the practice of ore benificiability field.