Improved adaptive evidential k-NN rule and its application for monitoring level of coal powder filling in ball mill

Abstract To achieve improvements in the production capacity and energy efficiency of an industrial tubular ball mill, an Improved adaptive Evidence-Theoretic k -NN rule was proposed, and was applied to monitor an unmeasured parameter, i.e., level of coal powder filling in ball mill. The improved adaptive rule was realized by means of two strategies: (1) a parametric distance metric was applied instead of Euclidean distance metric, and (2) some structure parameters fixed in the original adaptive rule, as well as the parameters brought in by the parametric distance metric, were optimized based on gradient descent algorithm. Some popular data sets were used to validate the performance of the improved adaptive rule. Consequently, the improved adaptive rule was applied to monitor the level of coal powder filling in ball mill. To demonstrate the validation of the improved adaptive rule for this task, some experiments were conducted on an industrial tubular ball mill. The experimental results suggest that the improved adaptive rule was applicable for monitoring the unmeasured parameter in industry.

[1]  Uwe Kruger,et al.  Recursive partial least squares algorithms for monitoring complex industrial processes , 2003 .

[2]  Thierry Denoeux,et al.  A k-nearest neighbor classification rule based on Dempster-Shafer theory , 1995, IEEE Trans. Syst. Man Cybern..

[3]  S. Wold,et al.  PLS-regression: a basic tool of chemometrics , 2001 .

[4]  Yugang Guo,et al.  Acoustic Vibration Signal Processing and Analysis in Ball Mill , 2006, 2006 6th World Congress on Intelligent Control and Automation.

[5]  J. Kolacz.  Measurement system of the mill charge in grinding ball mill circuits , 1997 .

[6]  Weihua Li,et al.  Recursive PCA for adaptive process monitoring , 1999 .

[7]  Zhi-gang Su,et al.  Experimental investigation of vibration signal of an industrial tubular ball mill: Monitoring and diagnosing , 2008 .

[8]  Lu Zhen-zhong.  A Study of the Use of Grey Soft Measurement in the Detection of Fill-up Level of Ball Mills , 2006 .

[9]  G. Irwin,et al.  Process monitoring approach using fast moving window PCA , 2005 .

[10]  Thierry Denoeux,et al.  Analysis of evidence-theoretic decision rules for pattern classification , 1997, Pattern Recognit..

[11]  Christopher M. Bishop,et al.  Neural networks for pattern recognition , 1995 .

[12]  Tordis E. Morud,et al.  Multivariate statistical process control; example from the chemical process industry , 1996 .

[13]  Thierry Denoeux,et al.  Handling possibilistic labels in pattern classification using evidential reasoning , 2001, Fuzzy Sets Syst..

[14]  Arthur P. Dempster,et al.  Upper and Lower Probabilities Induced by a Multivalued Mapping , 1967, Classic Works of the Dempster-Shafer Theory of Belief Functions.

[15]  Thierry Denoeux,et al.  ECM: An evidential version of the fuzzy c , 2008, Pattern Recognit..

[16]  Thierry Denoeux,et al.  An evidence-theoretic k-NN rule with parameter optimization , 1998, IEEE Trans. Syst. Man Cybern. Part C.

[17]  Jie Zhang,et al.  Process monitoring using non-linear statistical techniques , 1997 .

[18]  Glenn Shafer,et al.  A Mathematical Theory of Evidence , 2020, A Mathematical Theory of Evidence.

[19]  Vladimir Vapnik,et al.  Statistical learning theory , 1998 .

[20]  Barry M. Wise,et al.  The process chemometrics approach to process monitoring and fault detection , 1995 .

[21]  C. V. R. Murty,et al.  Experimental analysis of charge dynamics in tumbling mills by vibration signature technique , 2007 .

[22]  Xun Wang,et al.  An approach to using MSPC for power plant process monitoring and fault diagnosis , 2008 .

[23]  Yu Xiang-jun.  Soft Sensor Modeling for On-line Monitoring the Capacity of Coal Pulverizing System , 2007 .

[24]  Philippe Smets,et al.  The Transferable Belief Model , 1994, Artif. Intell..