CHEMICAL-PLANT FAULT-DIAGNOSIS THROUGH A HYBRID SYMBOLIC-CONNECTIONIST APPROACH AND COMPARISON WITH NEURAL NETWORKS

A novel hybrid symbolic-connectionist approach to machine learning is illustrated for fault diagnosis of a hydrocarbon chlorination plant. T he learning algorithm addressed the knowledge acquisition problem by d eveloping and maintaining the knowledge base through inductive learnin g. The performance ofthe learning system is discussed in terms ofthe k nowledge extracted from example cases and its classification accuracy on the test cases. Results indicate that the introduced system is a pr omising alternative to neural networks for fault diagnosis and a compl ement to expert systems.