基于改进的SSA−BP神经网络的矿井突水水源识别模型研究

刘伟韬; 李蓓蓓; 杜衍辉; 韩梦珂; 赵吉园

doi:10.13272/j.issn.1671-251x.2023070101

基于改进的SSA−BP神经网络的矿井突水水源识别模型研究

doi: 10.13272/j.issn.1671-251x.2023070101

刘伟韬^{1, 2,},
李蓓蓓^{1, 2, ,},
杜衍辉^{1, 2},
韩梦珂^{1, 2},
赵吉园^{1, 2}

1.
山东科技大学矿山灾害预防控制省部共建国家重点实验室培育基地，山东青岛　266590
2.
山东科技大学安全与环境工程学院，山东青岛　266590

基金项目: 山东省自然科学基金资助项目(ZR2023ME002)。

详细信息

作者简介:
刘伟韬（1970—），男，山东东明人，教授，博士，博士研究生导师，研究方向为矿井水害防治等，E-mail：skdlwt@126.com

通讯作者:
李蓓蓓（1999—），女，山东济宁人，硕士研究生，研究方向为矿井水害防治，E-mail：lieeb1999@163.com。

中图分类号: TD745
计量
- 文章访问数: 60
- HTML全文浏览量: 19
- PDF下载量: 18
- 被引次数: 0
出版历程
- 收稿日期: 2023-06-04
- 修回日期: 2023-12-25
- 网络出版日期: 2024-03-01

Research on the recognition model of mine water inrush source based on improved SSA-BP neural network

LIU Weitao^{1, 2
,},
LI Beibei^{1, 2
, ,},
DU Yanhui^{1, 2},
HAN Mengke^{1, 2},
ZHAO Jiyuan^{1, 2}

1.
State Key Laboratory of Mining Disaster Prevention and Control Co-founded by Shandong Province and the Ministry of Science and Technology，Shandong University of Science and Technology，Qingdao 266590，China
2.
College of Safety and Environmental Engineering，Shandong University of Science and Technology，Qingdao 266590，China

摘要

摘要: 机器学习与寻优算法的结合在矿井突水水源识别上得到广泛应用，但突水水样数据具有随机性且寻优算法易陷入局部最优，提高模型泛化能力和跳出局部最优需进一步研究。针对上述问题，提出了一种改进的麻雀搜索算法（SSA）优化BP神经网络模型，用于对矿井突水水源进行定量辨识。以鲁能煤电股份有限公司阳城煤矿为研究对象，通过常规离子浓度分析、Piper三线图对该煤矿水样的水化学特征进行分析，初步判断矿井水来源于奥灰含水层和三灰含水层，并确定Na⁺+K⁺浓度、Ca²⁺浓度、Mg²⁺浓度、${\mathrm{HCO}}_3^- $浓度、${\mathrm{SO}}_4^{2-} $浓度、Cl⁻浓度、矿化度、总硬度、pH值作为突水水源识别指标；建立基于改进SSA−BP神经网络的矿井突水水源识别模型：首先进行SSA参数设置，引入Sine混沌映射使麻雀种群均匀分布，然后通过计算适应度值进行麻雀种群的更新，引入随机游走策略扰动当前最优个体，如果满足终止条件，则获得最优BP神经网络权重和阈值，最后基于构建的BP神经网络，输出识别结果。研究结果表明：① 改进的SSA−BP模型在训练集上的识别准确率达95.6%，在测试集上的识别准确率达100%。② 改进的SSA−BP神经网络模型与BP神经网络模型、SSA−BP神经网络模型对比结果：BP神经网络模型误判率为5/18，SSA−BP神经网络模型的误判率为2/18，改进的SSA−BP神经网络模型误判率为0，迭代10次后趋于稳定，且与设定的目标误差相差最小，初始适应度值最优，识别结果可信度高。③ 将阳城煤矿5组矿井水水样数据作为输入层数据输入到训练好的模型中，矿井水水样的主要来源为奥灰含水层、三灰含水层和山西组含水层，模型识别结果与水化学特征分析的结论相互印证，实现了精准区分。
- 矿井突水水源识别 /
- 水化学特征 /
- 麻雀搜索算法 /
- BP神经网络 /
- 混沌映射 /
- 随机游走策略
Abstract: The combination of machine learning and optimization algorithms has been widely applied in the recognition of mine water inrush sources. However, the data of water inrush samples is stochastic and the optimization algorithm is prone to getting stuck in local optima. Further research is needed to improve the model's generalization capability and jump out of local optima. In order to solve the above problems, an improved sparrow search algorithm (SSA) is proposed to optimize the BP neural network model for quantitative recognition of mine water inrush sources. Taking Yangcheng Coal Mine of Luneng Coal and Electricity Co., Ltd. as the research object, the hydrochemical characteristics of the coal mine water sample are analyzed through conventional ion concentration analysis and Piper three line diagram. It is preliminarily determined that the mine water comes from the Ordovician limestone aquifer and the three limestone aquifers. The Na⁺+K⁺ concentration, Ca²⁺ concentration, Mg²⁺ concentration, ${\mathrm{HCO}}_3^- $ concentration, ${\mathrm{SO}}^{2-}_4 $ concentration, Cl⁻ concentration, mineralization degree, total hardness, and pH value are determined as the recognition indicators for water inrush source. The mine water inrush source recognition model is established based on an improved SSA-BP neural network. Firstly, the SSA parameters are set. Sine chaotic mapping is introduced to evenly distribute the sparrow population. Secondly, the sparrow population is updated by calculating fitness values, and a random walk strategy is introduced to perturb the current optimal individual. If the termination condition is met, the optimal BP neural network weight and threshold are obtained. Finally, based on the constructed BP neural network, the recognition results are output. The research results indicate the following points. ① The improved SSA-BP model has an recognition accuracy of 95.6% in the training set and 100% in the testing set. ② The comparison results of the improved SSA-BP neural network model with the BP neural network model and SSA-BP neural network model show that the BP neural network model has a misjudgment rate of 5/18, the SSA-BP neural network model has a misjudgment rate of 2/18, and the improved SSA-BP neural network model has a misjudgment rate of 0. After 10 iterations, it tends to stabilize and has the smallest error difference from the set target. The initial fitness value is the best, and the recognition results have high credibility. ③ Five sets of mine water samples from Yangcheng Coal Mine are inputted into the trained model as input layer data. The main sources of mine water samples are the Ordovician limestone aquifer, the three limestone aquifers, and the Shanxi formation aquifer. The results of model recognition are mutually confirmed with the conclusions of hydrochemical characteristic analysis, and precise segmentation is achieved.
- recognition of mine water inrush source /
- water chemical characteristics /
- sparrow search algorithm /
- BP neural network /
- chaotic mapping /
- random walk strategy

HTML全文

图 1 研究区主要含水层

Figure 1. Main aquifers in the study area

下载: 全尺寸图片幻灯片

图 2 研究区水样Piper三线图

Figure 2. Piper three-line diagram of water samples in the study area

下载: 全尺寸图片幻灯片

图 3 改进的SSA−BP神经网络模型建立流程

Figure 3. Process of improved the SSA−BP neural network model

下载: 全尺寸图片幻灯片

图 4 改进的SSA−BP神经网络模型训练集识别准确率

Figure 4. Training set recognition accuracy of improved SSA-BP neural network model

下载: 全尺寸图片幻灯片

图 5 改进的SSA−BP神经网络模型测试集识别准确率

Figure 5. Testing set recognition accuracy of improved SSA-BP neural network model

下载: 全尺寸图片幻灯片

图 6 各模型准确率预测结果

Figure 6. The accuracy prediction results of each model

下载: 全尺寸图片幻灯片

图 7 各模型均方误差变化曲线

Figure 7. Mean square error change curves of each model

下载: 全尺寸图片幻灯片

图 8 适应度值变化曲线

Figure 8. Adaptability change curves

下载: 全尺寸图片幻灯片

表 1 水化学特征参数统计结果

Table 1. Statistical results of hydrochemical characteristic parameters

类别	评价指标	Ca²⁺/(mg·L^-1)	Mg²⁺/(mg·L^-1)	Na⁺+K⁺/(mg·L^-1)	Cl⁻/(mg·L^-1)	$ \mathrm{SO}_4^{2-}/(\mathrm{mg\cdot L}^{-1}) $	HCO₃⁻/(mg·L^-1)	TH/(mg·L^-1)	TDS/(mg·L^-1)	pH
奥灰含水层	最大值	1294.90	337.36	2436.35	5192.60	2346.10	644.44	529.00	615.59	8.19
	最小值	22.08	5.33	355.10	511.77	2.71	14.20	145.60	212.00	5.43
	平均值	701.12	166.15	1354.98	2527.31	1512.41	203.66	167.66	390.06	7.36
	标准差	335.52	87.68	807.54	1664.17	626.53	160.47	136.80	124.15	0.66
	变异系数	0.48	0.53	0.60	0.66	0.41	0.79	0.82	0.32	0.09
三灰含水层	最大值	583.20	306.04	3077.03	4804.09	1368.57	630.18	517.00	768.00	8.20
	最小值	21.86	13.58	1076.94	1386.83	1.10	161.64	132.56	168.00	6.60
	平均值	267.39	124.90	2063.29	3583.85	359.03	310.83	253.86	338.14	7.40
	标准差	196.03	90.00	415.58	915.46	456.63	137.36	116.07	134.90	0.41
	变异系数	0.73	0.72	0.20	0.26	1.27	0.44	0.46	0.40	0.05
山西组含水层	最大值	216.67	68.09	2279.30	3009.71	1295.72	581.32	775.61	1325.00	8.30
	最小值	9.88	6.74	399.51	387.96	42.81	238.77	196.81	602.20	7.50
	平均值	62.31	23.17	1149.90	1354.40	527.71	395.81	457.31	942.78	8.02
	标准差	71.89	21.68	572.59	817.09	385.72	113.93	125.96	189.58	0.26
	变异系数	1.15	0.94	0.50	0.60	0.73	0.29	0.35	0.20	0.03
第四系含水层	最大值	1531.58	716.69	1414.34	944.77	503.01	1815.30	1489.00	2304.00	8.84
	最小值	7.06	2.12	45.71	36.04	0.85	50.89	57.3	111.18	7.37
	平均值	488.83	158.42	506.90	250.58	203.78	722.59	606.91	634.74	7.98
	标准差	431.65	172.46	415.19	231.44	144.85	530.09	443.54	541.94	0.47
	变异系数	0.88	1.09	0.82	0.92	0.71	0.73	0.73	0.85	0.06
矿井水	最大值	954.15	230.72	2405.52	4518.95	2030.21	484.84	425.49	1040.23	8
	最小值	31.40	8.19	355.10	958.15	523.56	106.36	153.56	212.00	7.00
	平均值	562.54	159.21	1233.30	2153.93	1270.25	231.47	92.67	473.44	7.48
	标准差	375.44	88.79	741.83	1439.99	682.29	146.47	168.13	329.12	0.43
	变异系数	0.67	0.57	0.60	0.67	0.54	0.63	0.66	0.70	0.06

下载: 导出CSV

表 2 改进的SSA−BP神经网络识别结果

Table 2. Improved SSA-BP neural network recognition results

水样	得分				识别结果
水样	奥灰含水层	三灰含水层	山西组含水层	第四系含水层	识别结果
1	0.9281	0.1216	0.0015	0.0041	奥灰含水层
2	0.9901	0.0020	0.3143	0.0160	奥灰含水层
3	0.8701	0.3224	0.0015	0.0067	奥灰含水层
4	0.4085	0.7123	0.0021	−0.0025	三灰含水层
5	0.0016	0.2619	0.5394	0.0277	山西含水层

下载: 导出CSV

参考文献(19)

[1]	曾一凡,武强,赵苏启,等. 我国煤矿水害事故特征、致因与防治对策[J]. 煤炭科学技术,2023,51(7):1-14. ZENG Yifan,WU Qiang,ZHAO Suqi,et al. Characteristics,causes,and prevention measures of coal mine water hazard accidents in China[J]. Coal Science and Technology,2023,51(7):1-14.
[2]	王昱同,王皓,王甜甜,等. 蒙陕接壤区浅埋煤层矿井水水化学特征及来源分析[J]. 煤田地质与勘探,2023,51(4):85-94. doi: 10.12363/issn.1001-1986.22.07.0553 WANG Yutong,WANG Hao,WANG Tiantian,et al. Hydrochemical characteristics and source analysis of mine water in shallow coal seams in Shaanxi and Inner Mongolia contiguous area[J]. Coal Geology & Exploration,2023,51(4):85-94. doi: 10.12363/issn.1001-1986.22.07.0553
[3]	范立民,马万超,常波峰,等. 榆神府矿区地下水水化学特征及形成机理[J]. 煤炭科学技术,2023,51(1):383-394. FAN Limin,MA Wanchao,CHANG Bofeng,et al. Hydrochemical characteristics and formation mechanism of groundwater in Yushenfu Mining Area[J]. Coal Science and Technology,2023,51(1):383-394.
[4]	刘伟韬,刘云娟,申建军. 基于模糊物元理论的深部开采底板突水安全性评价[J]. 山东科技大学学报(自然科学版),2014,33(3):25-31. LIU Weitao,LIU Yunjuan,SHEN Jianjun. Safety evaluation of floor water inrush in deep mining based on the fuzzy matter-element theory[J]. Journal of Shandong University of Science and Technology(Natural Science),2014,33(3):25-31.
[5]	WU Qiang,MU Wenping,XING Yuan,et al. Source discrimination of mine water inrush using multiple methods:a case study from the Beiyangzhuang Mine,Northern China[J]. Bulletin of Engineering Geology & the Environment,2017(78):469-482.
[6]	邵良杉,詹小凡. 基于IWOA−HKELM的矿井突水水源识别[J]. 中国安全科学学报,2019,29(9):113-118. SHAO Liangshan,ZHAN Xiaofan. Identification method of mine water inrush source based on IWOA-HKELM[J]. China Safety Science Journal,2019,29(9):113-118.
[7]	秋兴国,王瑞知,张卫国,等. 基于PCA−CRHJ模型的矿井突水水源判别[J]. 工矿自动化,2020,46(11):65-71. QIU Xingguo,WANG Ruizhi,ZHANG Weiguo,et al. Discrimination of mine inrush water source based on PCA-CRHJ model[J]. Industry and Mine Automation,2020,46(11):65-71.
[8]	LI Xiang,DONG Donglin,LIU Kun,et al. Identification of mine mixed water inrush source based on genetic algorithm and XGBoost algorithm:a case study of Huangyuchuan Mine[J]. Water,2022(14):2150-2167.
[9]	段李宏,戴磊,张金陵. 基于Fisher判别模型的煤层底板突水水源预测[J]. 工矿自动化,2022,48(4):128-134. DUAN Lihong,DAI Lei,ZHANG Jinling. Prediction of water inrush source of coal seam floor based on Fisher discriminant model[J]. Journal of Mine Automation,2022,48(4):128-134.
[10]	施龙青,曲兴玥,韩进. 黄土梁峁地貌矿井水水质时空变异评估与关键控制因子水源识别[J]. 煤田地质与勘探,2023,51(2):195-206. doi: 10.12363/issn.1001-1986.22.12.0943 SHI Longqing,QU Xingyue,HAN Jin. Evaluation on spatiaotemporal variations of mine water quality and water source identification based on key dominant factors in loess hilly and gully region[J]. Coal Geology & Exploration,2023,51(2):195-206. doi: 10.12363/issn.1001-1986.22.12.0943
[11]	尹会永,周鑫龙,郎宁,等. 基于SSA优化的GA−BP神经网络煤层底板突水预测模型与应用[J]. 煤田地质与勘探,2021,49(6):175-185. YIN Huiyong,ZHOU Xinlong,LANG Ning,et al. Prediction model of water inrush from coal floor based on GA-BP neural network optimized by SSA and its application[J]. Coal Geology & Exploration,2021,49(6):175-185.
[12]	黄敏,毛岸,路世昌,等. 矿井突水水源识别的主成分分析−混沌麻雀搜索−RF模型[J]. 安全与环境学报,2023,23(8):2607-2614. HUANG Min,MAO An,LU Shichang,et al. Identification of mine water inrush source based on PCA-CSSA-RF model[J]. Journal of Safety and Environment,2023,23(8):2607-2614.
[13]	LI Qiang,SUI Wanghua. Risk evaluation of mine-water inrush based on principal component logistic regression analysis and an improved analytic hierarchy process[J]. Hydrogeology Journal,2021(29):1299-1311.
[14]	李海祥,曹志国,王路军,等. 台格庙矿区地下水水化学特征与演变规律研究[J]. 煤炭科学技术,2023,51(9):284-291. LI Haixiang,CAO Zhiguo,WANG Lujun,et al. Study on chemical characteristics and evolution law of groundwater in Taigemiao Mining Area[J]. Coal Science and Technology,2023,51(9):284-291.
[15]	曾一凡,梅傲霜,武强,等. 基于水化学场与水动力场示踪模拟耦合的矿井涌(突)水水源判识[J]. 煤炭学报,2022,47(12):4482-4494. ZENG Yifan,MEI Aoshuang,WU Qiang,et al. Source discrimination of mine water inflow or inrush using hydrochemical field and hydrodynamic field tracer simulation coupling[J]. Journal of China Coal Society,2022,47(12):4482-4494.
[16]	姜子豪,胡友彪,琚棋定,等. 矿井突水水源判别方法[J]. 工矿自动化,2020,46(4):28-33. JIANG Zihao,HU Youbiao,JU Qiding,et al. A discrimination method of mine water inrush source[J]. Industry and Mine Automation,2020,46(4):28-33.
[17]	闫鹏程,尚松行,张超银,等. 改进BP神经网络算法对煤矿水源的分类研究[J]. 光谱学与光谱分析,2021,41(7):2288-2293. YAN Pengcheng,SHANG Songxing,ZHANG Chaoyin,et al. Classification of coal mine water sources by improved BP neural network algorithm[J]. Spectroscopy and Spectral Analysis,2021,41(7):2288-2293.
[18]	XUE Jiankai,SHEN Bo. A novel swarm intelligence optimization approach:sparrow search algorithm[J]. Systems Science & Control Engineering,2020,8(1):22-34.
[19]	李光阳,潘家文,钱谦,等. 融合学习机制的多混沌麻雀搜索算法[J]. 计算机科学与探索,2023,17(5):1057-1074. LI Guangyang,PAN Jiawen,QIAN Qian,et al. Multi-chaotic sparrow search algorithm based on learning mechanism[J]. Journal of Frontiers of Computer Science and Technology,2023,17(5):1057-1074.