Adaptive Nonlinear Control for Multi-tank Level System

For some disturbances and model uncertainties of the multi-tank level system, we develop an adaptive nonlinear control law based on the neural network to achieve high performance control. A nonlinear model including the model uncertainties and disturbances is firstly built through carefully analyzing the multi-tank level system. Based on the model, the level controller is designed, in which the adaptive law based on the neural network is used to approach the model uncertainty and disturbances, and the nonlinear feedback control law deals with the nonlinearity of the multi-tank level system. The proposed control strategy is thoroughly tested by simulation results compared with those obtained from a normal PID controller. The comparisons demonstrate that, for the multi-tank system with the model uncertainty, the designed control strategy is more advantageous in disturbance rejection, with higher control accuracy and so on, and is far superior to the PID control method.