Neural Networks-based Internal Model Control And Its Application

The NNs-IMC is a type of nonlinear adaptive control. In general, there are tow NNs in the system of NNs-IMC. One is neuro-controller, and another is neuro-emulator. The neuro-emulator emulates the model of the plant online, and gives the Jacobin of the plant to neuro-controller. Because this control strategy absorbs both IMC' and NNs adaptive control' merits, there has been considerable interest in exploring its applications in the control community. In the dissertation, control strategy was discussed in detail. To guarantee convergence and for faster learning, an approach that uses adaptive learning rates is developed by a Lyapunov function. At the last, a simulation has been conducted with the proposed NNs-IMC.The contributions of this dissertation includes the following parts:1. The development of the NNs-IMC and the system of level control were overviewed.2. The characteristic of the NNs-IMC was discussed in detail. Some improvement was proposed to enhance its performance. And then, convergence of the neuro-emulator and stability of the neuro-controller were proved under those supposed conditions. Last, a simulation is conducted to verify the proposed strategy.3. The development of the level control system was described. Some measures which can decrease the total cost of the development InTouch-based systems were put forward.