Abstract
Methods currently used to identify water sources and predict mine water inrush points are mostly based on water temperatures, water levels, or hydrochemical data. However, these methods are relatively complex. Therefore, to rapidly and effectively identify the source of mine water inrush, an elite strategy was used to optimize a genetic algorithm. The optimized elitist genetic algorithm (EGA) was applied to parameter selection of a probabilistic neural network (PNN) using spectral data to construct an EGA–PNN discriminant model. Water samples were collected from four different sources in the Huangyuchuan Mine in north China, including Permian sandstone fissure water, Carboniferous coal seam roof water, Ordovician limestone water, and Permian mined-out water. Full spectral data of the water samples were obtained, and 80% of the sample data were randomly selected as the training set for the EGA–PNN model. Theoretical analysis and experimental results showed that after 100 iterations, the average recognition rate was ≈ 95.69% with a mean square error average of only ≈ 8.21 × 10−3. This study provides a novel methodology that can be used to rapidly identify mine water inrush sources in similar geological coalfields elsewhere in northern China.
Zusammenfassung
Die methodische Herangehensweise zur Identifizierung des Ursprungs und zur Vorhersage von Wassereinbrüchen in Bergwerken basiert vorwiegend auf Wassertemperatur, Wasserstand oder hydrochemischen Daten. Diese Methoden sind jedoch relativ komplex. Um den Ursprung für Wassereinbrüche in Bergwerken schnell und effektiv zu identifizieren, wurde eine Elitenselektion zur Optimierung eines genetischen Algorithmus genutzt. Der optimierte, elitenbasierte genetische Algorithmus (EGA) wurde zur Auswahl von Parametern aus einem wahrscheinlichkeitsbasierten neuronalen Netz (PNN) unter Verwendung spektraler Daten genutzt, um ein EGA-PNN Diskriminanzmodell zu entwickeln. Im Huangyuchuan Bergwerk wurden Wasserproben aus vier verschiedenen Quellen genommen: Kluftwasser aus permischem Sandstein, Wasser aus karbonischen Kohleflöz-Deckschichten, Wasser aus ordovizischem Kalkstein und permisches Grubenwasser. Von allen Wasserproben wurden vollständige Spektraldaten bestimmt, und 80% der Datenstichprobe wurden zufällig als Trainingssatz für das EGA-PNN Modell gewählt. Theoretische Analyse und experimentelle Ergebnisse zeigen, dass die durchschnittliche Wiedererkennungsrate nach 100 Iterationen bei ca. 95,69% mit einer mittleren quadratischen Abweichung von nur 8,21×10-3 lag. Die vorliegende Studie bietet somit eine neue Methode zur schnellen Identifizierung des Ursprungs von Wassereinbrüchen in geologisch vergleichbaren Kohlerevieren in Nordchina.
目前水源识别和矿井突水预测的方法多基于水温、水位或水化学数据。然而, 这些方法相对复杂。为了快速有效地确定矿井突水水源, 采用精英保留策略优化了遗传算法。应用优化精英遗传算法 (EGA) 进行光谱数据的概率神经网络 (PNN) 参数选择, 构建EGA-PNN判别模型。从华北黄玉川煤矿采集四种来源水样, 包括二叠系砂岩裂隙水、石炭系煤层顶板水、奥陶系灰岩水和二叠系矿井水。先获得水样的全频谱数据, 再随机抽取80%的样本数据作EGA-PNN模型训练集。理论分析和实验结果表明, 经过100次迭代之后, 模型的**均识别率为≈95.69%, 均方误差**均值仅≈8.21 × 10 −3。研究为**北方其它类似煤田快速识别矿井突水水源提供了一种新方法。.
Resumen
Los métodos utilizados actualmente para identificar las fuentes de agua y predecir los puntos de entrada de agua de la mina se basan principalmente en las temperaturas del agua, los niveles de agua o los datos hidroquímicos. Sin embargo, estos métodos son relativamente complejos. Por lo tanto, para identificar rápida y eficazmente la fuente de irrupción de agua en la mina, se utilizó una estrategia de reserva elitista para optimizar un algoritmo genético. El algoritmo genético elitista optimizado (EGA) se aplicó a la selección de parámetros de una red neuronal probabilística (PNN) utilizando datos espectrales para construir un modelo discriminante EGA-PNN. Se recogieron muestras de agua de cuatro fuentes diferentes en la mina de Huangyuchuan, en el norte de China, que incluían agua de fisuras de arenisca del Pérmico, agua del techo de la veta de carbón del Carbonífero, agua de caliza del Ordovícico y agua de minas del Pérmico. Se obtuvieron los datos espectrales completos de las muestras de agua y el 80% de los datos de las muestras se seleccionaron al azar como conjunto de entrenamiento para el modelo EGA-PNN. El análisis teórico y los resultados experimentales mostraron que después de 100 iteraciones, la tasa de reconocimiento promedio fue ≈ 95,69% con un error cuadrático medio promedio de sólo ≈ 8,21×10 -3. Este estudio proporciona una metodología novedosa que puede utilizarse para identificar rápidamente las fuentes de irrupción de agua de la mina en yacimientos geológicos de carbón similares en otros lugares del norte de China.
References
Balasubramanian S, Pugalenthi V (1999) Determination of total chromium in tannery waste water by inductively coupled plasma-atomic emission spectrometry, flame atomic absorption spectrometry and UV–visible spectrophotometric methods. Talanta 50:457–467. https://doi.org/10.1016/S0039-9140(99)00135-6
Blanco A, Delgado M, Pegalajar MC (2000) A genetic algorithm to obtain the optimal recurrent neural network. Int J Approx Reason 23:67–83. https://doi.org/10.1016/S0888-613X(99)00032-8
Chen LW, Yin XX, **e WP, Feng XQ (2014) Calculating groundwater mixing ratios in groundwater-inrushing aquifers based on environmental stable isotopes (D, 18O) and hydrogeochemistry. Nat Hazards 71:937–953. https://doi.org/10.1007/s11069-013-0941-2
Cui FP, Wu Q, Lin YH, Zhao SQ, Zeng YF (2018) Prevention and control techniques and methods for water disasters at coal mines in China. J Min Sci Technol 3:219–228. https://doi.org/10.19606/j.cnki.jmst.2018.03.002[inChinese]
Gero MBP, Bello-García A, del Coz Díaz JJ (2005) A modified elitist genetic algorithm applied to the design optimization of complex steel structures. J Constr Steel Res 61:265–280. https://doi.org/10.1016/j.jcsr.2004.07.007
Guan ZL, Jia ZF, Zhao ZQ, You QY (2019) Identification of inrush water recharge sources using hydrochemistry and stable isotopes: a case study of Mindong no 1 coal mine in north-east Inner Mongolia China. J Earth Syst Sci. https://doi.org/10.1007/s12040-019-1232-4
Hoya T (2003) On the capability of accommodating new classes within probabilistic neural networks. IEEE Trans Neural Netw 14:450–453. https://doi.org/10.1109/TNN.2003.809417
Hu F, Zhou MR, Yan PC, Li DT, Lai WH, Bian K, Dai RY (2019) Identification of mine water inrush using laser-induced fluorescence spectroscopy combined with one-dimensional convolutional neural network. RSC Adv 9:7673–7679. https://doi.org/10.1039/c9ra00805e
Huang PH, Yang ZY, Wang XY, Ding FF (2019) Research on Piper-PCA-Bayes-LOOCV discrimination model of water inrush source in mines. Arab J Geosci. https://doi.org/10.1007/s12517-019-4500-3
Ling SH, Leung FHF, Lam HK, Yim Shu L, Tam PKS (2003) A novel genetic-algorithm-based neural network for short-term load forecasting. IEEE Trans Ind Electron 50:793–799. https://doi.org/10.1109/tie.2003.814869
Mao KZ, Tan KC, Ser W (2000) Probabilistic neural-network structure determination for pattern classification. IEEE Trans Neural Netw 11:1009–1016. https://doi.org/10.1109/72.857781
Naik SM, Jagannath RPK, Kuppili V (2019) Estimation of the smoothing parameter in probabilistic neural network using evolutionary algorithms. Arab J Sci Eng 45:2945–2955. https://doi.org/10.1007/s13369-019-04227-5
Oh SK, Park HS, Jeong CW, Joo SC (2009) GA-based feed-forward self-organizing neural network architecture and its applications for multi-variable nonlinear process systems. KSII T Internet INF 3:309–330. https://doi.org/10.3837/tiis.2009.03.006
Ojeda CB, Rojas FS (2004) Recent developments in derivative ultraviolet/visible absorption spectrophotometry. Anal Chim Acta 518:1–24. https://doi.org/10.1016/j.aca.2004.05.036
Oulhote Y, Bot BL, Deguen S, Glorennec P (2011) Using and interpreting isotope data for source identification. Trends Analyt Chem 30:302–312. https://doi.org/10.1016/j.trac.2010.10.015
Pajares G, Cruz JMDL (2002) A probabilistic neural network for attribute selection in stereovision matching. Neural Comput Appl 11:83–89. https://doi.org/10.1007/s005210200020
Piper M (1944) A graphic procedure in the geochemical interpretation of water-analyses. Am Geophys Union Trans 25:914–928. https://doi.org/10.1029/TR025i006p00914
Roy A, Das BK, Bhattacharya J (2011) Development and validation of a spectrophotometric method to measure sulfate concentrations in mine water without interference. Mine Water Environ 30:169–174. https://doi.org/10.1007/s10230-011-0140-x
Sedki A, Ouazar D, El Mazoudi E (2009) Evolving neural network using real coded genetic algorithm for daily rainfall–runoff forecasting. Expert Syst Appl 36:4523–4527. https://doi.org/10.1016/j.eswa.2008.05.024
Shi Z, Chow CWK, Fabris R, Liu J, ** B (2020) Alternative particle compensation techniques for online water quality monitoring using UV–Vis spectrophotometer. Chemometr Intell Lab Syst 204:104074. https://doi.org/10.1016/j.chemolab.2020.104074
Specht DF (1990) Probabilistic neural networks. Neural Netw 3:109–118. https://doi.org/10.1016/0893-6080(90)90049-Q
Sui WH, Liu JY, Yang SG, Chen ZS, Hu YS (2011) Hydrogeological analysis and salvage of a deep coalmine after a groundwater inrush. Environ Earth Sci 62:735–749. https://doi.org/10.1007/s12665-010-0562-y
Wu Q, Mu WP, **ng Y, Qian C, Shen JJ, Wang Y, Zhao DK (2019) Source discrimination of mine water inrush using multiple methods: a case study from the Beiyangzhuang Mine, northern China. Bull Eng Geol Environ 78:469–482. https://doi.org/10.1007/s10064-017-1194-1
Xu Z, Sun Y, Gao S, Zhao X, Duan R, Yao M, Liu Q (2018) Groundwater source discrimination and proportion determination of mine inflow using ion analyses: a case study from the Longmen coal mine, Henan Province, China. Mine Water Environ 37:385–392. https://doi.org/10.1007/s10230-018-0512-6
Yang Y, Yue JH, Li J, Yang Z (2018) Mine water inrush sources online discrimination model using fluorescence spectrum and CNN. IEEE Access 6:47828–47835. https://doi.org/10.1109/access.2018.2866506
Yang J, Dong S, Wang H, Li G, Wang T, Wang Q (2020) Mine water source discrimination based on hydrogeochemical characteristics in the northern Ordos Basin, China. Mine Water Environ 40:433–441. https://doi.org/10.1007/s10230-020-00723-5
Zhang J, Yao DX, Su Y (2018) Multivariate matrix model for source identification of inrush water: a case study from Renlou and Tongting coal mines in northern Anhui province, China. IOP Conf Ser: Earth Environ Sci. https://doi.org/10.1088/1755-1315/113/1/012212
Acknowledgements
This study was financially supported by the National Natural Science Foundation (41972255), the National Natural Science Foundation (U1710258), and the Ministry of Science and Technology of China (2017YFC0804104).
Author information
Authors and Affiliations
Corresponding author
Supplementary Information
Below is the link to the electronic supplementary material.
Rights and permissions
About this article
Cite this article
Ji, Y., Dong, D., Gao, J. et al. Source Discrimination of Mine Water Inrush Based on Spectral Data and EGA–PNN Model: A Case Study of Huangyuchuan mine. Mine Water Environ 41, 583–593 (2022). https://doi.org/10.1007/s10230-021-00840-9
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10230-021-00840-9