A versatile expert system Seneca in chemical and system engineering

Abstract

Mathematical model (statistical, analytical, fuzzy) is certain integration of primary data into a form ensuring simple use and solving the problem under consideration. Any processing of information is an irreversible process. In order that this loss of information may be minimized, an expert system SENECA has been developed. With the aid of algorithms of artificial intelligence this system utilizes directly the primary data in order to solve the given problem. The system has been based on the fuzzy similarity. The versatility of the proposed expert system rests in the fact that in a very simple manner an expert base can be implemented into this system. This base represents a fuzzy description of the primary data, concerning the problem under consideration. The primary data may be represented by laboratory measurements, by results of passive observation of a real equipment, literature data, experience, analogies, etc. The data are therefore very diverse. The approach is demonstrated on a two-dimensional model of a fermentor.