On-Line Elicitation of Mamdani-Type Fuzzy Rules via TSK- Based Generalised Predictive Control

Many synergisms have been described in the past between soft computing techniques such as Neural Networks (NN), Fuzzy Logic (FL) and Genetic Algorithms (GA) which have not only shown that such hybrid structures can work well but also add more robustness to the control system under consideration. In this paper, a new control architecture is proposed whereby the 2 on-line generated fuzzy rules relating to the Self-Organising Fuzzy Logic Controller (SOFLC) are obtained via integration with the popular Generalised Predictive Control (GPC) algorithm using a Takagi-Sugeno Kang (TSK) based CARIMA model structure. In this approach, GPC replaces the Performance Index (PI) table which, as an incremental model, is traditionally used to find new rules, delete rules and amend existing rules. Because the GPC sequence is computed using predicted future outputs, the new hybrid approach rewards the time-delay very well. The new generic approach, named Generalised Predictive Self-Organising Fuzzy Logic Control (GPSOFLC), is applied to a well-known non-linear chemical process, the distillation column, and is shown to lead to the elicitation of an effective fuzzy rule-base in both qualitative and quantitative terms.