Research On Neuron Control Methods And Applications

The neuron-model-free control is one of important development branches in intelligent control. The design theory and applications of neuron-model-free control systems are discussed in this thesis. Several neuron-model-free control methods are proposed and the simulation experiments of controlling industrial plants with uncertainties such as pH processes, hydroelectric generating units are made. The results show the efficiency of the proposed algorithms. The main contributions given in this thesis are as follows:1. A fuzzy-neuron control scheme is presented for nonlinear plants. In order to improve the performance of the controller, a fuzzy-neuron controller is constructed by combining the fuzzy controller and the neuron controller. The neuron controller gain is tuned according to the error, the change in error and the setpoint of the systems. With an example of a pH process, the experiments are made. The simulation results demonstrate that the proposed method has satisfactory performance, strong robustness and adaptability.2. To the hydroelectric generating units with uncertainty, a fuzzy neuron control method based on immune tuning gain is proposed. The hybrid neuron controller is combined with the fuzzy controller, and the gain of neuron controller is tuned by the immune scheme. Simulation results show the efficiency of this method.3. The neuron controller is used to optimize a nonlinear PID controller designed with fuzzy logic, and then the nonlinear PID controller with neuron optimization parameters is put forward. The scheme is used to control the hydroelectric generating unit with uncertainties, the simulation results show that this controller has good performance, and strong robustness.