Adaptive control of the ore crushing process in cone crushers based on nonlinear predictive model

Abstract

The paper deals with the development adaptive control system of the ore crushing process based on nonlinear block-oriented dynamic models. It is define that the best dynamic approximation quality according to the minimum coefficient of variation of the root-mean-square error and identification time are provided by applying a hybrid structure that combines the Wiener model, the Hammerstein-Wiener model and the Laguerre orthonormal functions. Using the recursive least squares algorithm for parametric identification enables to adapt the disturbances caused by changes in the mining mass characteristics. A nonlinear model predictive control system of the ore crushing process is also developed. A method of control formation is proposed and based on the block-oriented model static nonlinearities inversion. The obtained system demonstrated high dynamics quality and low computational load on the digital controller.