Prediction of the Types of Binary Azeotropes in an Expert System, AZEOPERT

다국어 초록 (Multilingual Abstract)

This article presents the problem solving strategy for predicting the types of binary azeotropes in a hierarchical knowledge-based expert system, AZEOPERT. The use of expert system concepts is examined to develop a consultant for predicting the types of binary azeotropes in organic mixtures; especially the binary systems that until now have been not reported. The knowledge base for the prediction of the types of binary azeotropes consists of two levels of domain-specific heuristics structured hierarchically. The predictive capabilities and generality of AZEOPERT can be highly enhanced through the integration of different kinds of domain knowledge into the hierarchical structure. AZEOPERT has been implemented with the artificial intelligence programming language IF/Prolog, developed in a UNIX environment. About 60 heuristic rules for predicting the types of binary azeotrope have been implemented in AZEOPERT. The required input data for consultation are only names of homologous groups or occasionally, class names, classified by the tendency of hydrogen bondings for consulted binary components which can usually be provided easily. The heuristic rules in each level were formulated by a combination of theoretical considerations, practical experience, and case studies through a binary azeotropic data bank. Prediction of the types of binary azeotropes using AZEOPERT may give chemical engineers a good contribution to perform the process synthesis, moreover, to select the proper separating agents for separation processes and the proper refrigerants in the area of air conditioning and refrigeration.

목차 (Table of Contents)

• 1. Introduction
• 2. The Types of Binary Azeotropes
• 2.1 Theoretical Basis
• 2.2 studies on the Types of Binary Azeotropic System
• 3. Prediction of Types of Binary Azeotropes using Heuristic Rules
• 3.1 Konwledge Representation
• 3.2 Heuristic Rules for the Prediction of the Types of Binary Azotropes
• 3.3 Problem-Solving Strategy
• 4. Knowledge Implementation
• 5. conclusions
• References