The rotation speed of a mill is an important factor related to its operation and grinding efficiency. Analysis and regulation of the optimal speed under different working conditions can effectively reduce energy loss, improve productivity, and extend the service life of the equipment. However, the relationship between the optimal speed and different operating parameters has not received much attention. In this study, the relationship between the optimal speed and particle size and number was investigated using discrete element method (DEM). An improved exponential approaching law sliding mode control method is proposed to track the optimal speed of the mill. Firstly, a simulation was carried out to investigate the relationship between the optimal speed and different operating parameters under cross-over testing. The model of the relationships between the optimal rotation speed and the size and number of particles was established based on the response surface method. An improved sliding mode control using exponential approaching law is proposed to track the optimal speed, and simulation results show it can improve the stability and speed of sliding mode control near the sliding surface.