# Fast Identification Method of Mine Water Source Based on Laser-Induced Fluorescence Technology and Optimized LSTM

^{1}

^{2}

^{3}

^{*}

## Abstract

**:**

## 1. Introduction

## 2. Materials and Methods

#### 2.1. Sample Preparation

#### 2.2. Apparatus

#### 2.3. Dimensionality Reduction

#### 2.4. Deep Learning

- Seven-dimensional vector candidate solution $x=\left({x}_{1},\dots ,{x}_{7}\right)$.
- Mayfly velocity $v=\left({v}_{1},\dots ,{v}_{7}\right)$, defined as its position change.
- Each mayfly automatically adjusts its trajectory to its current best position ($pbest$) and the best position ($gbest$) any mayfly has achieved so far.
- Calculate fitness values, sort them, and obtain $pbest$ and $gbest$.

- 5.
- Update the position of male mayflies and female mayflies in turn, and make them mate.

- (a)
- Male mayfly movement: ${\mathit{x}}_{\mathit{i}}^{\mathit{t}}$ is the position of male mayfly $\mathit{i}$ in the search space at the time step $\mathit{t}$, add ${\mathit{v}}_{\mathit{i}}^{\mathit{t}+\mathit{1}}$ to the current position to change the position, expressed as:

- (b)
- Female mayfly movement: Female mayflies do not congregate, but fly to males to reproduce. ${\mathit{y}}_{\mathit{i}}^{\mathit{t}}$ is the mayfly $\mathit{i}$ at time $\mathit{t}$. Its position is updated by increasing the speed:

- (c)
- Mayflies mating: Select a parent from the male population and female population based on fitness function, and cross to produce two offspring as follows:

- 6.
- Calculate fitness, update $\mathit{p}\mathit{b}\mathit{e}\mathit{s}\mathit{t}$ and $\mathit{g}\mathit{b}\mathit{e}\mathit{s}\mathit{t}$.

#### 2.5. Model Assessment

## 3. Results

#### 3.1. Spectral Data

#### 3.2. Dimensionality Reduction

#### 3.3. Models Performance

## 4. Discussion

## 5. Conclusions

- (1)
- The MA-LSTM mine water-source identification model built from the fluorescence spectra of water samples after PCA dimensionality reduction processing has the closest prediction to the actual value, the best identification effect, the highest training efficiency, and the best performance.
- (2)
- Under the LSTM, GA-LSTM, and MA-LSTM models, the model built from the fluorescence spectra after dimensionality reduction by PCA performed better than that after dimensionality reduction by LDA, and PCA was more suitable for dimensionality reduction in the fluorescence spectra of mine water sources than LDA, which improved the overall performance of the water-source identification model.
- (3)
- Among the PCA and LDA dimensionality reduction algorithms, the MA optimization LSTM water-source recognition model is better than GA, and the MA optimization algorithm is more suitable for optimizing the LSTM water-source recognition model and improve the generalization ability and recognition efficiency of the LSTM water-source recognition. It is verified that the MA-LSTM recognition model after PCA dimensionality reduction has the best recognition effect, the best performance, and has certain reliability for mine water-source recognition.

## Author Contributions

## Funding

## Data Availability Statement

## Acknowledgments

## Conflicts of Interest

## References

- Zhang, J.; Xu, K.; Reniers, G.; You, G. Statistical analysis the characteristics of extraordinarily severe coal mine accidents (ESCMAs) in China from 1950 to 2018. Process. Saf. Environ. Prot.
**2019**, 133, 332–3400. [Google Scholar] [CrossRef] - Bi, Y.; Wu, J.; Zhai, X.; Wang, G.; Shen, S.; Qing, X. Discriminant analysis of mine water inrush sources with multi-aquifer based on multivariate statistical analysis. Environ. Earth Sci.
**2021**, 80, 1–17. [Google Scholar] [CrossRef] - Zhang, H.; Xing, H.; Yao, D.; Liu, L.; Xue, D.; Guo, F. The multiple logistic regression recognition model for mine water inrush source based on cluster analysis. Environ. Earth Sci.
**2019**, 78, 1–15. [Google Scholar] [CrossRef] - Wang, H.; Dong, S.; Ji, Y.; Qiao, W.; Shang, H.; Zhu, K.; Zhou, Z.; Ning, D. Key technology and platform development of intelligent prevention and control on coal mine water disaster. Meitan Xuebao/J. China Coal Soc.
**2022**, 47, 883–892. [Google Scholar] - Yan, P.; Shang, S.; Zhang, C.; Yin, N.; Zhang, X.; Yang, G.; Sun, Q. Research on the Processing of Coal Mine Water Source Data by Optimizing BP Neural Network Algorithm with Sparrow Search Algorithm. IEEE Access
**2021**, 9, 108718–108730. [Google Scholar] [CrossRef] - Zhao, X.; Xu, Z.; Sun, Y. Identification of Mine Water Source Based on AHP-Entropy and Set Pair Analysis. Geofluids
**2022**, 2022, 3862746. [Google Scholar] [CrossRef] - Zhang, J.; Chen, L.; Chen, Y.; Ge, R.; Ma, L.; Zhou, K.; Shi, X. Discrimination of water-inrush source and evolution analysis of hydrochemical environment under mining in Renlou coal mine, Anhui Province, China. Environ. Earth Sci.
**2020**, 79, 1–13. [Google Scholar] [CrossRef] - Yan, P.-C.; Zhou, M.-R.; Liu, Q.-M.; Zhang, K.-Y.; He, C.-Y. Research on the source identification of mine water inrush based on LIF technology and SIMCA algorithm. Guang Pu Xue Yu Guang Pu Fen Xi/Spectrosc. Spectr. Anal.
**2016**, 36, 243–247. [Google Scholar] - Remmer, C.R.; Owca, T.; Neary, L.; Wiklund, J.A.; Kay, M.; Wolfe, B.B.; Hall, R. Delineating extent and magnitude of river flooding to lakes across a northern delta using water isotope tracers. Hydrol. Process.
**2019**, 34, 303–320. [Google Scholar] [CrossRef] - Huang, P.; Hu, Y.; Gao, H.; Su, Q. Dynamic identification and radium–radon response mechanism of floor mixed water source in high ground temperature coal mine. J. Hydrol.
**2021**, 603, 126942. [Google Scholar] [CrossRef] - Inostroza, M.; Moune, S.; Moretti, R.; Bonifacie, M.; Robert, V.; Burtin, A.; Chilin-Eusebe, E. Decoding water-rock interaction and volatile input at La Soufriere volcano (Guadeloupe) using time-series major and trace element analyses in gas condensates. J. Volcanol. Geotherm. Res.
**2022**, 425, 107517. [Google Scholar] [CrossRef] - Barral, N.; Husillos, R.; Castillo, E.; Cánovas, M.; Lam, E. Hydrochemical evolution of the Reocín mine filling water (Spain). Environ. Geochem. Health
**2021**, 43, 5119–5134. [Google Scholar] [CrossRef] [PubMed] - Barral, N.; Maleki, M.; Madani, N.; Cánovas, M.; Husillos, R.; Castillo, E. Spatio-temporal geostatistical modelling of sulphate concentration in the area of the Reocín Mine (Spain) as an indicator of water quality. Environ. Sci. Pollut. Res.
**2021**, 29, 86077–86091. [Google Scholar] [CrossRef] [PubMed] - Hu, F.; Zhou, M.; Yan, P.; Li, D.; Lai, W.; Bian, K.; Dai, R. Identification of mine water inrush using laser-induced fluorescence spectroscopy combined with one-dimensional convolutional neural network. RSC Adv.
**2019**, 9, 7673–7679. [Google Scholar] [CrossRef] [PubMed] - Ai, S.; Fan, Y.; Li, Y.; Gong, Y.; Ding, P.; Ma, F.; Li, X.; Wang, Z. The Diagnosis of Metal Vapor Density After Arc Extinction By Plane Laser-Induced Fluorescence. In Proceedings of the 2021 International Conference on Electrical Materials and Power Equipment (ICEMPE), Chongqing, China; 2021; pp. 1–4. [Google Scholar]
- Wan, M.; Sun, M.; Wu, G.; Gao, Q.; Li, B.; Wang, H.; Sun, Y.; Zhu, J. Quantitative feature extraction of turbulent premixed flames by photofragmentation laser-induced fluorescence. Opt. Eng.
**2021**, 60, 023108. [Google Scholar] [CrossRef] - Taylor, A.T.; Lai, E.P.C. Current State of Laser-Induced Fluorescence Spectroscopy for Designing Biochemical Sensors. Chemosensors
**2021**, 9, 275. [Google Scholar] [CrossRef] - Sherstinsky, A. Fundamentals of Recurrent Neural Network (RNN) and Long Short-Term Memory (LSTM) network. Phys. D Nonlinear Phenom.
**2020**, 404, 132306. [Google Scholar] [CrossRef] - Khandelwal, P.; Konar, J.; Brahma, B. (Eds.) Training RNN and it’s Variants Using Sliding Window Technique. In Proceedings of the 2020 IEEE International Students’ Conference on Electrical, Electronics and Computer Science (SCEECS), Bhopal, India, 22–23 February 2020. [Google Scholar]
- Srinivasu, P.N.; SivaSai, J.G.; Ijaz, M.F.; Bhoi, A.K.; Kim, W.; Kang, J.J. Classification of skin disease using deep learning neural networks with MobileNet V2 and LSTM. Sensors
**2021**, 21, 2852. [Google Scholar] [CrossRef] - Panja, P.; Jia, W.; McPherson, B. Prediction of well performance in SACROC field using stacked Long Short-Term Memory (LSTM) network. Expert Syst. Appl.
**2022**, 205, 117670. [Google Scholar] [CrossRef] - Lee, G.S.; Bang, S.S.; Mantooth, H.A.; Shin, Y.-J. Condition Monitoring of 154 kV HTS Cable Systems via Temporal Sliding LSTM Networks. IEEE Access
**2020**, 8, 144352–144361. [Google Scholar] [CrossRef] - Tang, G.; Lei, J.; Shao, C.; Hu, X.; Cao, W.; Men, S. Short-Term Prediction in Vessel Heave Motion Based on Improved LSTM Model. IEEE Access
**2021**, 9, 58067–58078. [Google Scholar] [CrossRef] - Mirza, A.H.; Kerpicci, M.; Kozat, S.S. Efficient online learning with improved LSTM neural networks. Digit. Signal Process. A Rev. J.
**2020**, 102, 102742. [Google Scholar] [CrossRef] - Bhattacharyya, T.; Chatterjee, B.; Singh, P.K.; Yoon, J.H.; Geem, Z.W.; Sarkar, R. Mayfly in Harmony: A New Hybrid Meta-Heuristic Feature Selection Algorithm. IEEE Access
**2020**, 8, 195929–195945. [Google Scholar] [CrossRef] - Zervoudakis, K.; Tsafarakis, S. A mayfly optimization algorithm. Comput. Ind. Eng.
**2020**, 145, 106559. [Google Scholar] [CrossRef] - Hu, A.; Deng, Z.; Yang, H.; Zhang, Y.; Gao, Y.; Zhao, D. An Optimal Geometry Configuration Algorithm of Hybrid Semi-Passive Location System Based on Mayfly Optimization Algorithm. Sensors
**2021**, 21, 7484. [Google Scholar] [CrossRef] [PubMed] - Rajakumar, M.; Sonia, R.; Maheswari, B.U.; Karuppiah, S. Tuberculosis detection in chest X-ray using Mayfly-algorithm optimized dual-deep-learning features. J. X-Ray Sci. Technol.
**2021**, 29, 961–974. [Google Scholar] [CrossRef] - Liu, Y.; Chai, Y.; Liu, B.; Wang, Y. Bearing fault diagnosis based on energy spectrum statistics and modified mayfly optimization algorithm. Sensors
**2021**, 21, 2245. [Google Scholar] [CrossRef] - Tang, J.; Zeng, J.; Wang, Y.; Yuan, H.; Liu, F.; Huang, H. Traffic flow prediction on urban road network based on License Plate Recognition data: Combining attention-LSTM with Genetic Algorithm. Transp. A Transp. Sci.
**2020**, 17, 1217–1243. [Google Scholar] [CrossRef] - Gupta, A.; Gupta, K.; Saroha, S. Short Term Solar Irradiation Prediction Framework Based on EEMD-GA-LSTM Method. Strat. Plan. Energy Environ.
**2022**, 41, 255–280. [Google Scholar] [CrossRef]

**Figure 4.**Prediction results for the validation set of different models: (

**a**) Prediction results of different models after LDA dimensionality reduction; (

**b**) Prediction results of different models after PCA dimensionality reduction.

Identification Models | Precision | Recall | F1-Score | Support |
---|---|---|---|---|

LDA-Original-LSTM | 0.97 | 0.97 | 0.97 | 70 |

LDA-GA-LSTM | 0.99 | 0.99 | 0.99 | 70 |

LDA-MA-LSTM | 1.00 | 1.00 | 1.00 | 70 |

PCA-Original-LSTM | 1.00 | 1.00 | 1.00 | 70 |

PCA-GA-LSTM | 1.00 | 1.00 | 1.00 | 70 |

PCA-MA-LSTM | 1.00 | 1.00 | 1.00 | 70 |

Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content. |

© 2023 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).

## Share and Cite

**MDPI and ACS Style**

Yan, P.; Zhang, X.; Kan, X.; Zhang, H.; Qi, R.; Huang, Q.
Fast Identification Method of Mine Water Source Based on Laser-Induced Fluorescence Technology and Optimized LSTM. *Water* **2023**, *15*, 701.
https://doi.org/10.3390/w15040701

**AMA Style**

Yan P, Zhang X, Kan X, Zhang H, Qi R, Huang Q.
Fast Identification Method of Mine Water Source Based on Laser-Induced Fluorescence Technology and Optimized LSTM. *Water*. 2023; 15(4):701.
https://doi.org/10.3390/w15040701

**Chicago/Turabian Style**

Yan, Pengcheng, Xiaofei Zhang, Xuyue Kan, Heng Zhang, Runsheng Qi, and Qingyun Huang.
2023. "Fast Identification Method of Mine Water Source Based on Laser-Induced Fluorescence Technology and Optimized LSTM" *Water* 15, no. 4: 701.
https://doi.org/10.3390/w15040701