Estimating Ore Production in Open-pit Mines Using Various Machine Learning Algorithms Based on a Truck-Haulage System an
- PDF / 9,083,755 Bytes
- 33 Pages / 595.276 x 790.866 pts Page_size
- 55 Downloads / 193 Views
Original Paper
Estimating Ore Production in Open-pit Mines Using Various Machine Learning Algorithms Based on a Truck-Haulage System and Support of Internet of Things Yosoon Choi,1 Hoang Nguyen ,2,7 Xuan-Nam Bui ,3,4 Trung Nguyen-Thoi,5,6 and Sebeom Park1 Received 27 June 2020; accepted 3 October 2020
This study aimed to develop and assess the feasibility of different machine learning algorithms for predicting ore production in open-pit mines based on a truck-haulage system with the support of the Internet of Things (IoT). Six machine learning algorithms, namely the random forest (RF), support vector machine (SVM), multi-layer perceptron neural networks (MLP neural nets), classification and regression tree, k-nearest neighbors, and M5Tree model, were developed and investigated to estimate ore production at a limestone open-pit mine in South Korea. Systems of IoT were used to collect a massive database of 16,217 observations (big data). To process this big data, a downscaling method was applied to reduce the size of the original observations to improve the computational cost of the machine learning models. Subsequently, three validation datasets were selected from the original observations and used to evaluate (after downscaling the observations) the performance and accuracy of the machine learning models in practical engineering through various performance metrics. The results revealed that the models used can be potentially used for predicting ore production in open-pit mines. The SVM, MLP neural nets, and RF models demonstrated high accuracy, with the SVM model exhibiting the most superior performance and the highest accuracy. An assessment of the validation datasets also confirmed the feasibility and stability of the models for predicting ore production. Furthermore, a sensitivity analysis indicated that the relative operation start time, relative operation end time, and interval between the operation times were the most important input variables for improving ore production in practical engineering. KEY WORDS: Ore production, Open-pit mine, Truck-haulage, System of IoT, Machine learning, Downscaling technique.
1
Department of Energy Resources Engineering, Pukyong National University, Busan 48513, Korea. 2 Institute of Research and Development, Duy Tan University, Da Nang 550000, Vietnam. 3 Department of Surface Mining, Mining Faculty, Hanoi University of Mining and Geology, 18 Vien st., Duc Thang Ward, Bac Tu Liem District, Hanoi, Vietnam. 4 Center for Mining, Electro-Mechanical Research, Hanoi University of Mining and Geology, 18 Vien st., Duc Thang Ward, Bac Tu Liem District, Hanoi, Vietnam.
5
Division of Computational Mathematics and Engineering, Institute for Computational Science, Ton Duc Thang University, Ho Chi Minh City, Vietnam. 6 Faculty of Civil Engineering, Ton Duc Thang University, Ho Chi Minh City, Vietnam. 7 To whom correspondence should be addressed; e-mail: [email protected]
2020 International Association for Mathematical Geosciences
Choi, Nguyen, Bui, Nguyen-Thoi, and Park
Data Loading...
