An approach to process modelling and control using linguistic fuzzy models with generalized rule conclusions

A linguistic fuzzy model with generalized rule conclusions is presented realizing a good compromise between interpretability and approximation capability. The data-driven model construction leads to constrained optimization problems. An algorithm for step-wise identification of input membership function parameters is developed which tackles the problem of multimodality of the objective function. In the second part a design method for fuzzy controllers based on a fuzzy plant model is described. The application of this method is demonstrated by means of an example.