Swarm Intelligence Based Fuzzy Controller -- A Design for Nonlinear Water Level Tank

Fuzzy direct controllers are being used widely in industry these days. One of the benefits of fuzzy controllers is their ability to control unidentified processes which lets a model free controlling scheme; but on the other hand, an efficient fuzzy direct controller design, strictly depends on human expert and knowledge of a certain process. This will limit the ability of noncontrol specialists to apply fuzzy controllers on various ranges of plants. In this paper, a fuzzy direct controller is optimized in rule base using Particle Swarm Optimization algorithm. The optimization is performed subjected to minimize the output error surface of a nonlinear water level tank process. An offline Sugeno-Fuzzy system identifier is employed to prepare the evaluation function for particle swarm algorithm. Results show that the proposed controller performance is much better than simple human knowledge tuned controller.

I. Introduction (HEADING 1)

Intelligent control techniques are nowadays well recognized tools in both academia and industry. Methodologies coming from the field of computational intelligence, such as neural networks, fuzzy systems and evolutionary computation, can be lead to accommodation of more complex processes, improved performance and considerable time saving and cost reduction[20]. The introduction of fuzzy systems was originally motivated by the need to represent (ambiguous) human knowledge and the corresponding deductive processes [23].