In order to detect the thermal performance of the enameled wire, a temperature control system was designed. BP neural network was used to correct the PID parameters online to obtain the optimal parameters, and BP-PID control algorithm was applied to the temperature control system.The simulation experiment was carried out by using MATLAB, and the experimental results verified that the control system could achieve better control effect. Enamelled wire detectors were used for the physical experiment. The results show that when the control temperature is 120 ℃, compared with the traditional PID control, the control mode has better temperature control effect, the overshoot rate is less than 4%, the steadystate precision is less than ±2 ℃, reaching the expected goal.