Supervised adaptive control for temperature regulation

In this paper, we propose an adaptive control scheme for a class of nonlinear time-delay systems. To alleviate the requirement for accurate modeling of the plant, a two-level robust control scheme was proposed. The inner-level adaptive neural controller is responsible for accurate servo control, while the outer-level supervising controller compensates for un-modeled system dynamics and bounded disturbances. Besides, each part of the proposed control law can be independently designed satisfying its own specifications. The number of unknown parameters is significantly reduced by using the neuron-like models. Moreover, the supervising control law and adaptation gains are designed based on stability conditions in the sense of Lyapunov. The scheme is readily applicable to temperature control systems where heat transfer and mass transportation cause significant delay effects. Experimental results collected from a prototype temperature regulation system verify effectiveness of the proposed scheme. Performance improvement of feed-forward control term is also clearly demonstrated.