Multi-model neural network-based intelligent controller

A new multi-model intelligent control scheme based on an artificial neural network is presented in this paper. The proposed controller is used to improve the system performance of a complex control system of known or unknown model by driving the process to follow the desired trajectory. The multi-model control scheme depends on the multiple representation of a process using different models, involving two phases; (i) locating the model that best matches the process, and (ii) generating the basic control signal that will drive the system to follow the desired trajectory according to the located model. A multilayer neural network is used to implement both phases of the multi model scheme. The proposed controller can accommodate any deviation from the prescribed models. Simulation results are presented to show the potential of developed scheme.<
>