Abstract: There were many shortages in the traditional PID control systems for injection molding machines，such as great control deviation, slow rate of regulation, etc. In order to control the temperature accurately, a self-adaptive fuzzy control method was proposed in this work，which combines fuzzy control and PID control method. The traditional PID control model and self- adaptive fuzzy PID control model were established by MATLAB software; the temperature of cylinders was simulated and compared. The temperature was set at 200℃.The result of simulation indicated that the self-adaption fuzzy PID controller had realized the real-time temperature control of injection molding machines and met the requirements of smaller overshoot， quick temperature rising and no steady-state error，which compared with traditional PID controller. Above-mentioned results illustrated that the temperature control effectiveness of the self-adaptation fuzzy PID controller was better than the traditional PID controller.