Abstract
The water hazard of the coal seam floor is a major threat to safe coal production in China. To improve the accuracy of water hazard predictions, a water inrush risk predictive model was constructed using PSO-SVM. Historical monitoring data were added to the basic database in a timely manner to narrow the difference between the monitoring data and predicted results. The optimized database was used for neural network model training. The prediction model was improved by establishing a database self-optimization and model self-learning process (SOMSP). The PSO-SVM model and the SOMSP was used to predict the inrush risk for 23 groups of floor water inrush cases from the north China mine area. The initial accuracy of the model was only 25% for the first 19 data groups, which were used as the basic training data to predict data groups 20–23. Using the SOMSP, the accuracy of the water inrush risk of the coal seam floor was increased to 100% (3/3). Thus, the accuracy of the predictions was greatly improved by the SOMSP.
Resumen
El riesgo de irrupción de agua en el fondo de la veta de carbón es una de las principales amenazas para la producción segura de carbón en China. Para mejorar la precisión de las predicciones del peligro del agua, se construyó un modelo de predicción del riesgo de irrupción de agua utilizando PSO-SVM. A la base de datos básica, se añadieron oportunamente datos históricos de monitorización para reducir la diferencia entre los datos de monitorización y los resultados predichos. La base de datos optimizada se utilizó para entrenar el modelo de red neuronal. El modelo de predicción se mejoró estableciendo un proceso de autooptimización de la base de datos y autoaprendizaje del modelo (SOMSP). El modelo PSO-SVM y el SOMSP se utilizaron para predecir el riesgo de irrupción de 23 grupos de casos de agua en el suelo de la zona minera del norte de China. La precisión inicial del modelo fue sólo del 25% para los 19 primeros grupos de datos, que se utilizaron como datos de entrenamiento básicos para predecir los grupos de datos 20–23. Utilizando el SOMSP, la precisión del riesgo de irrupción de agua en el fondo de la veta de carbón aumentó hasta el 100% (3/3), mostrando que el uso del SOMSP aumenta notablemente la precisión de las predicciones.
Zusammenfassung
Gefährdungen durch katastrophale Liegendwasserzutritte sind eine große Bedrohung für die sichere Kohleförderung in China. Um die Genauigkeit von Wassergefährdungsprognosen zu verbessern, wurde mit Hilfe von PSO-SVM (Particle Swarm Optimization-Support Vector Machine) ein Modell zur Vorhersage des Wassereinbruchsrisikos entwickelt. Historische Überwachungsdaten sind einer Basisdatenbank hinzugefügt worden, um den Unterschied zwischen den Überwachungsdaten und den vorhergesagten Ergebnissen zu verringern. Die optimierte Datenbank wird für das Training des neuronalen Netzes verwendet. Das Vorhersagemodell ist zudem durch die Einrichtung eines Selbstoptimierungs- und Selbstlernprozesses für die Datenbank (SOMSP) verbessert worden. Das PSO-SVM-Modell und das SOMSP werden zur Vorhersage des Zutrittsrisikos für 23 Gruppen von Liegendwasserzutritten im nordchinesischen Bergbaugebiet verwendet. Die Genauigkeit des Modells für die ersten 19 Datengruppen betrug anfänglich lediglich 25%. Im Anschluss wurden dieses als Trainingsdaten für die Vorhersage der Datengruppen 20–23 verwendet. Mit Hilfe von SOMSP konnte die Genauigkeit des Wassereinbruchsrisikos über die Sohle des Kohleflözes auf 100% (3/3) erhöht werden. Die Genauigkeit der Vorhersagen wird somit durch die Nutzung von SOMSP erheblich verbessert.
Similar content being viewed by others
Data Availability
The authors confirm that the data supporting the findings of this study are available within the article and its supplementary materials.
References
Bukowski P (2011) Water hazard assessment in active shafts in Upper Silesian coal basin mines. Mine Water Environ 30:302–311. https://doi.org/10.1007/s10230-011-0148-2
Bukowski P (2015) Evaluation of water hazard in hard coal mines in changing conditions of functioning of mining industry in Upper Silesian coal basin – USCB (Poland). Arch Min Sci 60(2):455–475. https://doi.org/10.1515/amsc-2015-0030
Chen XF, Miao Y, Chen WT (2019) Study on evaluation of mine floor water inrush danger based on fuzzy set pair analysis method. Coal Sci Technol 47(2):218–223 (in Chinese)
Cherkassky V, Ma YQ (2004) Practical selection of SVM parameters and noise estimation for SVM regression. Neural Netw 17(1):113–126. https://doi.org/10.1016/S0893-6080(03)00169-2
China Coal Mine Safety Administration (2018) Provisions on the prevention and control of water in coal mines. China Coal Industry Publishing Home (in Chinese)
Gu QX, Huang Z, Li SJ, Zeng W, Wu Y, Zhao K (2020) An approach for water-inrush risk assessment of deep coal seam mining: a case study in Xinlongzhuang coal mine. Environ Sci Pollut Res 27:43163–43176. https://doi.org/10.1007/s11356-020-10225-0
Jin DW, Qiao W, Li P, Fan JA (2019) Research status and prospects on intelligent technology and equipment for mine water hazard prevention and control. Coal Sci Technol 47(3):10–17 (in Chinese)
Li A, Xue Y, Jin CJ, Wang MH, Yao XB (2006) Prediction of Nε-acetylation on internal lysines implemented in Bayesian discriminant method. Biochem Biophysic Res Commun 350(4):818–824. https://doi.org/10.1016/j.bbrc.2006.08.199
Liu WT, Han MK, Meng XX, Qin YY (2021a) Mine water inrush risk assessment evaluation based on the GIS and combination weight-cloud model: a case study. ACS Omega 6:32671–32681. https://doi.org/10.1021/acsomega.1c04357
Liu SQ, Wu Q, Li Z, Zeng YF, Yuan QD, Yu YL (2021b) Vulnerability evaluation and application of floor water inrush in mining area with multiple coal seams and single aquifer based on variable weight. J China U Min Technol 50(3):587–597 (in Chinese)
Marini F, Walczakb B (2015) Particle swarm optimization (PSO). A Tutorial Chemometr Intell Lab Sys 149(12):153–165. https://doi.org/10.1016/j.chemolab.2015.08.020
Mark HL, Tunnell D (1985) Qualitative near-infrared reflectance analysis using Mahalanobis distances. Anal Chem 57(7):1449–1456. https://doi.org/10.1021/ac00284a061
Niknam T, Amiri B (2010) An efficient hybrid approach based on PSO, ACO and k-means for cluster analysis. Appl Soft Comput 10:183–197. https://doi.org/10.1016/j.asoc.2009.07.001
Orosun MM, Usikalu MR, Oyewumi KJ (2022) Environmental risks assessment of kaolin mines and their brick products using Monte Carlo simulations. Earth Syst Environ 6:157–174. https://doi.org/10.1016/j.sbspro.2013.12.568
Ranaee V, Ebrahimzadeh A, Ghaderi R (2010) Application of the PSO–SVM model for recognition of control chart patterns. ISA Trans 49(4):577–586. https://doi.org/10.1016/j.isatra.2010.06.005
Shi LQ, Tan XP, Wang JA, Ji XK, Niu C, Xu DJ (2015) Risk assessment of water inrush based on the PCA_fuzzy_PSO_SVC. J China Coal Soc 40(1):167–171 (in Chinese)
Shi LQ, Zhang RG, Xu DJ, Li Y, Qiu M, Gao WF (2020) Prediction of water inrush from floor based on GWO-Elman neural network. J China Coal Soc 45(7):2455–2463 ((in Chinese))
Thanh C-L, Trong N-N, Samir K, Phuoc T-N, Seyedali M, Khuong DN (2022) An efficient approach for damage identification based on improved machine learning using PSO-SVM engineering with computers. Engineering with Computers 38:3069–3084. https://doi.org/10.1007/s00366-021-01299-6
Tosun N (2006) Determination of optimum parameters for multi-performance characteristics in drilling by using grey relational analysis. Int J Adv Manuf Technol 28:450–455. https://doi.org/10.1007/s00170-004-2386-y
Wang XC, Shi F, Yu L, Li Y (2013) 43 cases of Matlab neural network. Beijing University of Aeronautics and Astronautics Press, Beijing (in Chinese)
Wu Q, Li B, Liu SQ, Zeng YF (2013) Vulnerability assessment of coal floor groundwater bursting based on zoning variable weight model: A case study in the typical mining region of Kailuan. J China Coal Soc 38(9):1516–1521 (in Chinese)
Wu Q, Liu Y, Zhou W, Wu X, Liu S, Sun W, Zeng Y (2015) Assessment of water inrush vulnerability from overlying aquifer using GIS–AHP-based ‘three maps–two predictions’ method: a case study in Hulusu coal mine, China. Q J Eng Geol Hydrogeol 48:234–243. https://doi.org/10.1144/qjegh2015-014
Wu Q, Liu YZ, Wu XL, Liu SQ, Sun WJ, Zeng YF (2016) Assessment of groundwater inrush from underlying aquifers in Tunbai coal mine, Shanxi province, China. Environ Earth Sci 75(9):737.1-737.13. https://doi.org/10.1007/s12665-016-5542-4
Wu Q, Zhao DK, Yang W, Shen J, Liu H (2017) Method for assessing coal-floor water-inrush risk based on the variable-weight model and unascertained measure theory. Hydrogeol J 25(10):1–15. https://doi.org/10.1007/s10040-017-1614-0
Yin HY, Zhou XL, Lang N, Zhang LF, Wang ML, Wu T, Li X (2021) Prediction model of water inrush from coal floor based on GA-BP neural network optimized by SSA and its application. Coal Geol Explor 49(6):175–185 (in Chinese)
Zhang FD, Shen BH (2018) PSO_SVM prediction model for evaluating water inrush risk from deep coal seam floor. Coal Sci Technol 46(7):61–67228 (in Chinese)
Zhang WQ, Zhang GP, Li W, Hua X (2013) A model of Fisher’s discriminant analysis for evaluating water inrush risk from coal seam floor. J China Coal Soc 38(10):1831–1836 (in Chinese)
Zhang GD, Xue YG, Bai CH, Su MX, Zhang K, Tao YF (2021) Risk assessment of floor water inrush in coal mines based on MFIM-TOPSIS variable weight model. J Central South Univ 28(8):2360–2374 (in Chinese)
Zhuo LW (2011) Matlab in Mathematical Modeling. Beijing Univ of Aeronautics and Astronautics Press, Beijing (in Chinese)
Acknowledgements
This work was supported by The National Natural Science Foundation of China (51704158, 51704161), Science and Technology Innovation Fund of Tiandi Science and Technology Co., Ltd (2019-TD-MS004), and Science and Technology Innovation Fund of Coal Mining and Designing Department of Tiandi Science and Technology Co., Ltd (KJ-2019-TDKCMS-02, KJ-2021-KCMS-06).
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
About this article
Cite this article
Zhang, F. Dynamic Monitoring and an Early Warning Model of a Floor Water Disaster: A Case Study. Mine Water Environ 42, 158–169 (2023). https://doi.org/10.1007/s10230-023-00925-7
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10230-023-00925-7