The diagnostic monitoring of the acoustic emission from a laboratory ball mill

[摘要] ENGLISH ABSTRACT: The harsh interior environment of mills makes on-line monitoring of these grinding systems difficult. Notonly are conventional contact sensors expensive, but the nature of the grinding process makes theirapplication impractical. Unfortunately few accurate quantitative measures are in place in industry todescribe or assist in the operation and diagnosis of ball mills. In the South African context operatorslearn to control the mill based on a priori knowledge of the system gathered from years of processexperience. It is common knowledge in industry that these operators associate the sound emissionfrom the system with certain process conditions, and adjust the mill set points to obtain optimalgrinding conditions. Unfortunately the high turnover of manpower in the mining industry has led to adrain of knowledge from many operations, leading to a loss of valuable control information.In this work the acoustic emission from a ball mill was studied making use of a laboratory ball mill,acoustic microphones and a personal computer, equipped with a sound card. The mill signal wasrecorded for a series of batch experiments. These consisted of single parameter experiments wheresingle parameters such as percentage filling, mill speed, percentage water and percentage chargemass were varied, while keeping all other parameters constant. A second series of experiments wereconducted with two platinum ore types, namely UG2 and Merensky, to study the influence of changingparticle size on the acoustic emission from the mill. The acoustic signal was transformed into the frequency domain from the time domain by using Welch'saveraged periodogram method. Hereby the power spectral density function for each acoustic samplewas obtained and used as the basis for further data analysis. The structure of the data wasinvestigated with a Sammon map obtained from the power spectral density data. This methodconfirmed that specific conditions in the mill each had a unique fingerprint which enabled differentiationof the acoustic information.Feature vectors were obtained by principal component analysis of the power spectrum density functionextracted from the original mill signal. These feature vectors were used for the modelling of differentdata sets. Linear regression was applied to the Single parameter experiments yielding modelling resultswith r² values above 0.95. With the platinum- ore data both linear regression and feed forward neuralnetworks were used for modelling. However, the linear regression model was unable to predict the oreparticle size from the acoustic data. The non-linear neural network models achieved accurate particlesize predictions for both ore types on both known and unknown validation data sets. r² values greaterthan 0.93 for the test data and 0.97 for the training data were obtained.

[发布机构] Stellenbosch University