Process monitoring and supervision in batch plant based on recipe adaptation

This paper propose the novel methodology for process monitoring and supervision in batch process by using the case based approach for retrieve, reuse, revise and retain the recipes. The simulation results show the learning mechanism of the recipe after solving the new problem and retaining the recipe into case based. This paper discusses the utilization of safety related data (HAZOP result, accident database, operating manuals, etc.) for supporting the run-time processes. This proposed system offers more efficient, low cost and accurate solution for handling any abnormal situations. It gives significant impact to batch processes which are partially or fully controlled by plant operators. A set of new problems can be solved by reuse of proven process solutions. The proposed system integrates quantitative and qualitative parameters for the adaptation of cases. A technique to generate the accurate recipes and adapt to the processes is introduced during normal and abnormal situation. In particular, the different between current process variables with the references (recipes) should be managed to prevent from any hazardous conditions. The processes are evaluated using similarity assessment to the case based. This intelligent approach can distinguish the plausible cases and generate the accurate recipes and adapt to the new situation. The aim is to use the offline historical process data and safety related information in order to propose changes and adjustment in the process.