Mould Bath Level State Recognition and Fuzzy Control Method Based on GA

In continuous casting process, because mould bath level can be easily affected by many complex and uncertain factors and difficult to control, a new structure-varied fuzzy controller is proposed. In this controller, a state pattern recognition method based on an improved adjustable clustering algorithm,dual-ISODATA algorithm is used. At the fuzzy controller,quantity and the content of the fuzzy rule and mumbership function are optimized by the Genetic Algorithms. The reasoning speed and the robustness of the control systems is greatly improved.The results of the simulation are satisfactory.