Skip to Main Content
Hierarchical fuzzy control can process distributed control parameters, reduce the number of fuzzy rules effectively and easily extract fuzzy rules, so it is suitable for non-linear temperature control for industrial furnaces with features, such as large capacity and long time-delay. But rule sets and membership functions in conventional fuzzy control are often pre-determined according to human experiences and will no longer be changed in whole control process. Therefore, in the case of that there are more uncertain and disturbed factors, its control effect becomes unsatisfactory. In response to this situation, we presented a GA-based two-stage fuzzy temperature control algorithm for industrial furnaces, which can greatly reduce the number of fuzzy rules by taking advantages of hierarchical fuzzy control and taking full account of impact of many procedure parameters upon controlled variables. In the basis of that the fuzzy control decision is made through expert knowledge, we optimized fuzzy control query table using genetic algorithms, which not only avoided the most unreasonable consequence produced in the process of optimization of the control rules, but also greatly increased the convergence rate. Practical application showed that the algorithm can reduce the fuel consumption and possess a high control precision and robustness. Particularly for large time delay, nonlinear systems, its quality was superior to conventional control and general fuzzy control.