Wet type ball grinder load parameter integrated modeling method based on EEMD (ensemble empirical mode decomposition)

Abstract

The invention discloses a wet type ball grinder load parameter integrated modeling method based on EEMD (ensemble empirical mode decomposition) and relates to a parameter measuring method of ore grinding production. The method includes: on the basis of ensemble empirical mode self-adaption decomposition nonlinear unstable cylinder vibration and vibration sound signal intrinsic mode functions (IMFs) and frequency spectrum of IMFs, using a interval partial least squares (iPLS) method to extract IMFs local frequency spectrum reflecting ball grinder load parameter change, fusing with current signals, and building an ore grinding concentration, filling rate and material-media ratio grinder load parameter integrated model on the basis of an extreme learning machine (ELM). By the method, the problem that the load parameters of a wet type ball grinder is hard to detect is solved, the defect of steady state hypothesis during ball grinder signal analyzing is overcome, and generalization and stability of the integrated model are increased.

Description

technical field [0001] The invention relates to a method for measuring parameters in an ore grinding production process, in particular to an EEMD-based integrated modeling method for load parameters of a wet ball mill. Background technique [0002] The load of the ball mill is an important indicator closely related to the production efficiency, product quality and energy consumption of the grinding process. Reasonable mill load parameters are a necessary condition to ensure the optimal operation of the ball mill. Due to the complex impact and grinding periodic movement between steel balls and materials, steel balls and liners in the ball mill, the working environment is harsh, and it is difficult for the sensor to be installed in the closed and rotating ball mill barrel to directly measure the internal load of the ball mill. The industrial production process mainly relies on manual labor. Judging by experience, it has a certain degree of randomness and uncertainty, which ca...

AI Technical Summary

Problems solved by technology

[0005] The purpose of the present invention is to provide an EEMD-based integrated modeling method for load parameters of wet ball mills. This method is limited by linearity or stationarity assumptions in traditional data analysis methods, and adopts a nonlinear non-stationary adaptive signal processing method—ensemble empirical mode Decomposing EEMD overcomes the distortion and mode confusion problems existing in the empirical mode decomposition process of the vibration signal of the cylinder wall. The present invention is an integrated modeling method for wet ball mill load parameters based on collective empirical mode decomposition

Images