Hybrid expert system for raw materials blending

The paper presents a case study on the practical implementation of a hybrid expert system for a raw materials blending process (RMBP). Based on blending mechanism and expert knowledge, a hybrid expert system for supervisory control is developed to optimize its performance. With the help of a compensation model and a prediction model, the proposed hybrid expert system can replace the human operator for most of the operations in the process. Both experiment and industrial applications show the feasibility and effectiveness of the developed system, and its bright potential in application of the control of the RMBP.