The PID control algorithm is used for the control of almost all loops in the process industries, and is also the basis for many advanced control algorithms and strategies. For complex systems, controller's structure and parameters need to rely on experience and on line tuning to determine. Application of soft computing techniques can realize selftuning of controller parameters. In this paper, the author applies fuzzy system, neural network technique and genetic algorithm to PID controller in order to achieve intelligent adjustment the controlled variables. Simulation results show that the application of soft computing techniques in the controller has better control quality.