Direct current (DC) motor has the advantages of large range speed regulation, large starting torque and easy to stop, especially suitable for high speed regulation requirements. Under the condition of load change, traditional proportional-integral (PI) algorithm DC speed regulation control needs to adjust the parameters, which is difficult to be adaptive. Aiming at this problem, a fuzzy adaptive PI controller is designed, which can perform fuzzy reasoning by identifying errors and error changes to realize the online control parameters setting and the optimal control. At the same time, Matlab is used to fit experimental data, the results show that the fuzzy PI control algorithm can much more effectively improve system dynamic and steady-state performances than the traditional PI control algorithm.