Electric Furnace Neural Network Internal Model Control

In this thesis, the temperature control, which has the nonlinear and great inertial character, is studied. The internal model control is proposed as computer control strategy. The controller is the inverse model. The model and inverse model of plant are discriminated by the characteristic BP neural network. In order to overcome the error between the plant and the model, the feedback filter is operated and designed on the feedback-loop. At the same time, in order to reach perfect control, the former filter is operated and designed. At last, the system of electric heater is simulated by the Matlab-Tool. The simulation result shows the neural network internal model structure has better control capability in the control of the electric heating stove.