Abstract: The mill has the characteristics of non-linearity, large inertia and large random disturbance during operation, which makes the conventional PID method unsuitable for load control to a certain extent.In this paper, an intelligent control method of mill load based on adaptive limit learning machine is proposed, based on the analysis of the operation process and characteristics of the mill, the design idea of the control strategy for the grinding process of the mill is put forward, the control system model of the mill is built by using the adaptive limit learning mechanism, and the intelligent optimal control of mill load is realized.The application results show that the control method can effectively eliminate the influence of disturbance on mill load, actively searching for the optimum load of the mill makes the mill run steadily.It has good adaptive ability and certain significance for improving grinding efficiency and classification effect.