PROJECT TITLE :

Online Optimization of Fuzzy Controller for Coke-Oven Combustion Process Based on Dynamic Just-in-Time Learning

ABSTRACT:

To ensure the control performance of a fuzzy management system for the combustion process in a very coke oven, the parameters of the fuzzy controller would like to be optimized therefore that the controller can handle large changes in the operating state of the oven. This paper describes an online optimization method for this purpose. During this technique, the gap and angle of the trend of the change are used to pick out knowledge, and just-in-time learning is used to form a dynamic sample base and to build a radial-basis-operate neural-network model of the process. A variable-universe fuzzy logic controller controls the method, and an adaptive differential evolution algorithm optimizes the universe parameters. This enables the controller to adapt to changes in the operating state in a very timely fashion. Simulation results demonstrate the effectiveness of the strategy.