Fault detecting intelligent system for Baker's yeast fermentation

Baker's yeast fermentation consists of a number of upstream and downstream processes and is sensitive to contamination, composition of medium, changes in process parameters like pH, temperature and aeration. Any deviation fro m optimum process condition or any leakage or breakdown in process equipmen t may lead to huge financial loss. The main problems are maintenance of ste rility, avoiding equipment faults and thus keeping maximum production rate. An intelligent diagnostic system for prompt detection of any malfunction o f such process will be helpful. Intelligent diagnostic systems have been ex tensively studied by many research workers. The underlying concept of such a system is to find a collection of disorders (causes) that can be the best explanation for the present set of symptoms (manifestation), which may be considered as a deviation from normal behaviour. The reverse process of rea soning to reach the conclusion without using any deductive rule is known as 'abductive reasoning' A prototype expert system for diagnosis of Baker's y east fermentation was developed on the expert system shell Knowledge Engine ering Sytsem.HT (KES.HT) based on abductive reasoning.