Immune PID cascade control based on neural network for main steam temperature system

Abstract

The main steam temperature control system in thermal power plant has the features of large inertia, large delay, time-varying and uncertainty, using the traditional PID cascade control strategy is difficult to obtain a satisfactory quality of control. Biological Immune System has strong robustness and self-adaptability in the environment with lots of disturbance and uncertainty. Neural network has the ability of approximating any non-linear functions, by the self-learning of neural network, the control parameters can be obtained under an optimal control rule. Learning from the feedback mechanism of biological immune and the self-learning capability of neural network, a control strategy of immune PID cascade control applied to main steam temperature control system in the thermal power plant based on self-learning of the neural network parameters has been proposed. Simulation study shown that the method of the control is better than traditional PID control, which can adapt to the changes of object parameters, and show a good control quality, and has strong robustness and self-adaptability.