Journal article
MINE'S PUMP STATION ENERGY CONSUMPTION AND UNDERGROUND WATER DAM LEVELS MONITORING SYSTEM USING MACHINE LEARNING CLASSIFIERS AND MUTUAL INFORMATION ENSEMBLE TECHNIQUE
INTERNATIONAL JOURNAL OF INNOVATIVE COMPUTING INFORMATION AND CONTROL, v 12(6), pp 1777-1789
01 Dec 2016
Featured in Collection : UN Sustainable Development Goals @ Drexel
Abstract
This paper proposes the use of artificial intelligence techniques for monitoring and predicting the underground water dam level in a double pump station gold mine. Six single classifiers methods (support vector machine, artificial neural network, naive Bayesian classifier, decision trees, radial basis function and k nearest neighbors) are applied for this purpose. The paper further proposes a new approach to select, the most, suitable classifiers when constructing the most, accurate ensemble in multiple classifier learning. This approach is based on determining the mutual information amount between classifier pairs and is further used to determine the classifiers optimum number in order to build the most, accurate ensemble. Simulation results using underground pump station dam levels and energy consumption data show the proposed strategy as being more superior to methods such as Bagging and Boosting techniques in terms of predictive accuracy.
Metrics
5 Record Views
Details
- Title
- MINE'S PUMP STATION ENERGY CONSUMPTION AND UNDERGROUND WATER DAM LEVELS MONITORING SYSTEM USING MACHINE LEARNING CLASSIFIERS AND MUTUAL INFORMATION ENSEMBLE TECHNIQUE
- Creators
- Ali N. Hasan - University of JohannesburgBhekisipho Twala - University of Johannesburg
- Publication Details
- INTERNATIONAL JOURNAL OF INNOVATIVE COMPUTING INFORMATION AND CONTROL, v 12(6), pp 1777-1789
- Publisher
- Icic International
- Number of pages
- 13
- Resource Type
- Journal article
- Language
- English
- Academic Unit
- Engineering Leadership and Society/Engineering Technology
- Web of Science ID
- WOS:000406146300002
- Other Identifier
- 991022004632904721
UN Sustainable Development Goals (SDGs)
This publication has contributed to the advancement of the following goals:
InCites Highlights
Data related to this publication, from InCites Benchmarking & Analytics tool:
- Web of Science research areas
- Computer Science, Artificial Intelligence