CLASSIFICATION OF COAL MINE PILLAR STABILITY USING EXTREME LEARNING MACHINE AND PARTICLE SWARM OPTIMIZATION ADAPTIVE WEIGHT DELAY VELOCITY

Nadhilah Farhana; Gatot Fatwanto  Hertono; Bevina Desjwiandra  Handari

doi:10.52436/1.jutif.2024.5.6.2403

Nadhilah Farhana Magister Matematika, Fakultas Matematika dan Ilmu Pengetahuan Alam, Universitas Indonesia, Indonesia
Gatot Fatwanto Hertono Magister Matematika, Fakultas Matematika dan Ilmu Pengetahuan Alam, Universitas Indonesia, Indonesia
Bevina Desjwiandra Handari Magister Matematika, Fakultas Matematika dan Ilmu Pengetahuan Alam, Universitas Indonesia, Indonesia

DOI: https://doi.org/10.52436/1.jutif.2024.5.6.2403

Keywords: adaptive weight delay velocity, extreme learning machine, particle swarm optimization, pillars stability classification, underground mining

Abstract

In underground mining, pillars are prime structural parts for supporting the overburden. Precise prediction of pillar stability is necessary because pillar failure might cause catastrophic events that could endanger mining personnel and equipment. This research aims to classify the stability of underground coal mining pillars using the Extreme Learning Machine model with Particle Swarm Optimization Adaptive Weight Delay Velocity that used to optimize the model's input weights and bias. Extreme Learning Machine is a model for training artificial neural networks using a single-layer feedforward Network architecture. Performance comparison is presented between the proposed model and the Particle Swarm Optimization-Extreme Learning Machine. The dataset originated from South African coal mining with two pillar stabilities: failed and intact. The pillar stability of the dataset expanded into five categories: failed upper, failed lower, intact upper, intact lower slender, intact lower not-slender. Out of the five pillar stability categories, the failed lower category is the most dangerous pillar category, with the rest are non-dangerous pillar category. The expanded categories are according to the Probability of Failure of the pillar and the type of pillar (slender, intermediate, and squat). The results showed that the AUC 91,4%; 74,3%; 72,6%, and G-mean 82,2% were all at least 10% higher in the proposed model. The proposed model successfully classified 91.24% of non-dangerous pillar stability category, but only 36% of the most dangerous pillar stability category. The results of this study are expected could give assistance to provide information as a consideration in predicting pillar.

Downloads

Download data is not yet available.

References

J. Brodny dan M. Tutak, “Applying computational fluid dynamics in research on ventilation safety during underground hard coal mining: A systematic literature review,” Process Saf. Environ. Prot., vol. 151, hal. 373–400, 2021, doi: 10.1016/j.psep.2021.05.029.

J. N. Van Der Merwe, “Beyond Coalbrook: Critical Review of Coal Strata Control Developments in South Africa,” in 25th International Conference on Ground Control in Mining, 1998, no. 1957. [Daring]. Tersedia pada: http://www.mineaccidents.com.au/uploads/coalbrook-mine-disaster-1960-van-der-merwe-2006.pdf

A. P. Schissler, “Coal Mining, Design and Methods of,” Encycl. Energy, vol. 1, hal. 485–494, 2004, doi: 10.1016/b0-12-176480-x/00284-9.

W. Liang, S. Luo, G. Zhao, dan H. Wu, “Predicting hard rock pillar stability using GBDT, XGBoost, and LightGBM algorithms,” Mathematics, vol. 8, no. 5, hal. 1–17, 2020, doi: 10.3390/MATH8050765.

N. Li, M. Zare, C. Yi, dan R. Jimenez, “Stability Risk Assessment of Underground Rock Pillars Using Logistic Model Trees,” Int. J. Environ. Res. Public Health, vol. 19, no. 4, 2022, doi: 10.3390/ijerph19042136.

J. N. Van Der Merwe dan M. Mathey, “Probability of failure of South African coal pillars,” J. South. African Inst. Min. Metall., vol. 113, no. 11, hal. 849–857, 2013, [Daring]. Tersedia pada: https://www.saimm.co.za/Journal/v113n11p849.pdf

A. S. Tawadrous dan P. D. Katsabanis, “Prediction of surface crown pillar stability using artificial neural networks,” Int. J. Numer. Anal. Methods Geomech., vol. 31, no. 7, hal. 917–931, 2007, doi: 10.1002/nag.566.

R. K. Wattimena, “Predicting the stability of hard rock pillars using multinomial logistic regression,” Int. J. Rock Mech. Min. Sci., vol. 71, hal. 33–40, 2014, doi: 10.1016/j.ijrmms.2014.03.015.

H. Ding, G. Li, X. Dong, dan Y. Lin, “Prediction of Pillar Stability for Underground Mines Using the Stochastic Gradient Boosting Technique,” IEEE Access, vol. 6, no. October, hal. 69253–69264, 2018, doi: 10.1109/ACCESS.2018.2880466.

G. F. Hertono, R. K. Wattimena, G. A. Mendrofa, dan B. D. Handari, “Classifying Coal Mine Pillar Stability Areas with Multiclass SVM on Ensemble Learning Models,” J. Eng. Technol. Sci., vol. 56, no. 1, hal. 95–109, 2024, doi: 10.5614/j.eng.technol.sci.2024.56.1.8.

E. Ghasemi, H. Kalhori, dan R. Bagherpour, “Stability assessment of hard rock pillars using two intelligent classification techniques: A comparative study,” Tunn. Undergr. Sp. Technol., vol. 68, no. August 2016, hal. 32–37, 2017, doi: 10.1016/j.tust.2017.05.012.

J. Zhou, X. B. Li, X. Z. Shi, W. Wei, dan B. B. Wu, “Predicting pillar stability for underground mine using Fisher discriminant analysis and SVM methods,” Trans. Nonferrous Met. Soc. China (English Ed., vol. 21, no. 12, hal. 2734–2743, 2011, doi: 10.1016/S1003-6326(11)61117-5.

M. Ahmad, N. A. Al-Shayea, X. W. Tang, A. Jamal, H. M. Al-Ahmadi, dan F. Ahmad, “Predicting the pillar stability of underground mines with random trees and C4.5 decision trees,” Appl. Sci., vol. 10, no. 18, 2020, doi: 10.3390/APP10186486.

B. Kumar, S. K. Sharma, dan G. S. P. Singh, “Enhanced prediction of hard rock pillars stability using fuzzy rough feature selection followed by random forest,” J. Mines, Met. Fuels, vol. 67, no. 11, hal. 509–514, 2019, [Daring]. Tersedia pada: https://openurl.ebsco.com/EPDB%3Agcd%3A6%3A25882662/detailv2?sid=ebsco%3Aplink%3Ascholar&id=ebsco%3Agcd%3A139929533&crl=f

X. Ying, “An Overview of Overfitting and its Solutions,” J. Phys. Conf. Ser., vol. 1168, no. 2, 2019, doi: 10.1088/1742-6596/1168/2/022022.

A. Akusok, K. M. Bjork, Y. Miche, dan A. Lendasse, “High-Performance Extreme Learning Machines: A Complete Toolbox for Big Data Applications,” IEEE Access, vol. 3, hal. 1011–1025, 2015, doi: 10.1109/ACCESS.2015.2450498.

R. Manjuparkavi, M. Shanthi, dan M. C. Bhuvaneshwari, “Extreme learning machine algorithm and its application,” 2nd Int. Conf. Latest Trends Eng. Sci. Humanit. Manag., hal. 50–58, 2017, [Daring]. Tersedia pada: http://www.arresearchpublication.com/images/shortpdf/1488120670_IF2014ijeee.pdf

C. Blum dan X. Li, “Swarm Intelligence in Optimization,” Swarm Intell., hal. 43–85, 2008, doi: 10.1007/978-3-540-74089-6_2.

M. R. Kaloop dkk., “Particle Swarm Optimization algorithm-Extreme Learning Machine (PSO-ELM) model for predicting resilient modulus of stabilized aggregate bases,” Appl. Sci., vol. 9, no. 16, 2019, doi: 10.3390/app9163221.

Z. Liu dan T. Nishi, “Adaptive heterogeneous particle swarm optimization with comprehensive learning strategy,” J. Adv. Mech. Des. Syst. Manuf., vol. 16, no. 4, hal. 1–15, 2022, doi: 10.1299/jamdsm.2022jamdsm0035.

M. Li, H. Chen, X. Wang, N. Zhong, dan S. Lu, An Improved Particle Swarm Optimization Algorithm with Adaptive Inertia Weights, vol. 18, no. 3. 2019. doi: 10.1142/S0219622019500147.

P. Refaeilzadeh, L. Tang, dan H. Liu, “Cross-Validation,” in Contemporary Interventional Ultrasonography in Urology, 2009, hal. 1–6. doi: 10.1007/978-1-84800-217-3_1.

S. Prusty, S. Patnaik, dan S. K. Dash, “SKCV: Stratified K-fold cross-validation on ML classifiers for predicting cervical cancer,” Front. Nanotechnol., vol. 4, no. August, hal. 1–12, 2022, doi: 10.3389/fnano.2022.972421.

M. S. Santos, J. P. Soares, P. H. Abreu, H. Araujo, dan J. Santos, “Cross-validation for imbalanced datasets: Avoiding overoptimistic and overfitting approaches [Research Frontier],” IEEE Comput. Intell. Mag., vol. 13, no. 4, hal. 59–76, 2018, doi: 10.1109/MCI.2018.2866730.

E. Sutoyo, M. Asri Fadlurrahman, J. Telekomunikasi Jl Terusan Buah Batu, K. Dayeuhkolot, K. Bandung, dan J. Barat, “JEPIN (Jurnal Edukasi dan Penelitian Informatika) Penerapan SMOTE untuk Mengatasi Imbalance Class dalam Klasifikasi Television Advertisement Performance Rating Menggunakan Artificial Neural Network,” JEPIN, vol. 6, no. 3, hal. 379–385, 2020, [Daring]. Tersedia pada: https://jurnal.untan.ac.id/index.php/jepin/article/view/42896/75676587815

D. A. Nasution, H. H. Khotimah, dan N. Chamidah, “Perbandingan Normalisasi Data untuk Klasifikasi Wine Menggunakan Algoritma K-NN,” Comput. Eng. Sci. Syst. J., vol. 4, no. 1, hal. 78, 2019, doi: 10.24114/cess.v4i1.11458.

A. S. Almajid, “Multilayer Perceptron Optimization on Imbalanced Data Using SVM-SMOTE and One-Hot Encoding for Credit Card Default Prediction,” J. Adv. Inf. Syst. Technol., vol. 3, no. 2, hal. 67–74, 2022, doi: 10.15294/jaist.v3i2.57061.

G. Bin Huang, Q. Y. Zhu, dan C. K. Siew, “Extreme learning machine: A new learning scheme of feedforward neural networks,” IEEE Int. Conf. Neural Networks - Conf. Proc., vol. 2, hal. 985–990, 2004, doi: 10.1109/IJCNN.2004.1380068.

Q. Fan dan T. Fan, “A Hybrid Model of Extreme Learning Machine Based on Bat and Cuckoo Search Algorithm for Regression and Multiclass Classification,” J. Math., vol. 2021, 2021, doi: 10.1155/2021/4404088.

D. Wu, X. Wang, dan S. Wu, “A hybrid method based on extreme learning machine and wavelet transform denoising for stock prediction,” Entropy, vol. 23, no. 4, 2021, doi: 10.3390/e23040440.

X. Zhang, Z. Yang, F. Cao, J. Cao, M. Wang, dan N. Cai, “Conditioning optimization of extreme learning machine by multitask beetle antennae swarm algorithm,” Memetic Comput., vol. 12, no. 2, hal. 151–164, 2020, doi: 10.1007/s12293-020-00301-w.

I. Goodfellow, Y. Bengio, dan A. Courville, “Deep Learning,” MIT Press, vol. 521, no. 7553, hal. 785, 2017, doi: 10.1016/B978-0-12-391420-0.09987-X.

U. S. Shanthamallu dan A. Spanias, Neural Networks and Deep Learning. 2022. doi: 10.1007/978-3-031-03758-0_5.

H. He dan M. Yunqian, Imbalanced Learning. New Jersey: He, H., & Yunqian, M. (2013). Imbalanced Learning. In IEEE Press. https://doi.org/doi:https://doi.org/10.1002/9781118646106.fmatter, 2013. doi: doi:https://doi.org/10.1002/9781118646106.fmatter.

M. Grandini, E. Bagli, dan G. Visani, “Metrics for Multi-Class Classification: an Overview,” A White Pap., hal. 1–17, 2020, [Daring]. Tersedia pada: http://arxiv.org/abs/2008.05756

C. O. Truica dan C. A. Leordeanu, “Classification of an Imbalanced Data Set Using,” U.P.B. Sci. Bull., Ser. C, vol. 79, no. 4, 2017, [Daring]. Tersedia pada: https://www.researchgate.net/publication/322103560_Classication_of_an_Imbalanced_Data_Set_using_Decision_Tree_Algorithms

A. Tharwat, “Classification assessment methods,” Appl. Comput. Informatics, vol. 17, no. 1, hal. 168–192, 2018, doi: 10.1016/j.aci.2018.08.003.

J. S. Akosa, “Predictive Accuracy: A Misleading Performance Measure for Highly Imbalanced Data,” SAS Glob. Forum, vol. 942, hal. 1–12, 2017, [Daring]. Tersedia pada: https://support.sas.com/resources/papers/proceedings17/0942-2017.pdf

J. N. Van Der Merwe dan M. Mathey, “Update of coal pillar database for South African coal mining,” J. South. African Inst. Min. Metall., vol. 113, no. 11, hal. 825–840, 2013, [Daring]. Tersedia pada: https://www.saimm.co.za/Journal/v113n11p825.pdf

A. K. Verma, “A Comparative Study of Various Empirical Methods to Estimate the Factor of Safety of Coal Pillars,” Am. J. Min. Metall., vol. 2, no. 1, hal. 17–22, 2014, doi: 10.12691/ajmm-2-1-3.

J. N. Van Der Merwe dan M. Mathey, “Update of coal pillar strength formulae for South African coal using two methods of analysis,” J. South. African Inst. Min. Metall., vol. 113, no. 11, hal. 825–840, 2013, [Daring]. Tersedia pada: https://www.saimm.co.za/Journal/v113n11p841.pdf

M. A. A. Albadr, S. Tiun, M. Ayob, dan F. T. AL-Dhief, “Particle Swarm Optimization-Based Extreme Learning Machine for COVID-19 Detection,” Cognit. Comput., hal. 1858–1873, 2022, doi: 10.1007/s12559-022-10063-x.