To some time-varying, large time delay, nonlinear complex systems in Modern industrial control system, it is difficult for the conventional PID algorithm to set the PID parameters. So that it is more difficult to achieve the desired results. The article realizes the conventional PID algorithm and single neuron adaptive PID algorithm in the linear and nonlinear system control respectively based on the experiment platform of the MCGS. The experimental results show that the single neuron adaptive PID control algorithm in linear system and nonlinear system has better stability and dynamic performance.