瓦斯安全事故是长期以来困扰我国煤矿井下人员安全和生产作业的重大影响因素之一，对煤矿井下人员和设备安全造成了严重威胁。综采工作面作为煤矿开采的第一现场，属于瓦斯事故高发区域，因此对综采工作面瓦斯浓度异常变化趋势进行研究，实现稳定的瓦斯浓度监测并提前对瓦斯浓度异常趋势进行预警，对保障煤矿工作人员生命安全和提高煤炭资源生产有着重要的研究价值和实践意义。
本文以综采工作面瓦斯浓度监测数据为研究对象，在对瓦斯浓度监测预警方法国内外研究现状分析的基础上，对综采工作面瓦斯浓度异常变化趋势特征进行研究，确定了综采工作面瓦斯浓度预测参数，在此基础上采用门控循环单元（GRU）神经网络进行综采工作面瓦斯浓度预测。针对GRU在预测过程中由于梯度下降算法易陷入局部最优问题，采用改进鲸鱼优化算法（IWOA）对神经网络模型参数进行寻优并进行瓦斯浓度预测。将预测结果与预测样本灰色关联度分析相结合建立了IWOA-GRU瓦斯浓度预警模型并应用于综采工作面瓦斯浓度预警中，并通过实验仿真对预警可靠性进行验证。本文在煤矿KJ95X安全监测平台的基础上设计了综采工作面瓦斯浓度预警系统，通过主机、数据库与预警模型的交互式应用完成了综采工作面瓦斯浓度实时预测，实现了综采工作面瓦斯浓度数据显示、预测和预警功能。
本文通过对综采工作面瓦斯浓度数据监测预警方法的研究，可以实现综采工作面瓦斯浓度的准确预测并对瓦斯异常上升趋势进行预警，能够对煤矿安全监控系统进行完善和补充，在一定程度上确保了煤矿安全生产，具有一定的理论研究价值和工程应用价值。

Gas safety is one of the significant factors that have plagued the safety and production operations of coal mine personnel in China for a long time and have caused serious threats to the safety of coal mine personnel and equipment. As the first site of coal mining, the fully mechanized coal mining watching belongs to the area with a high incidence of gas accidents. In light of this, the study of gas concentration abnormal change trend in the fully mechanized mining watching, to achieve stable gas concentration monitoring and early warning of the abnormal gas concentration trend, which may ensure the protection life safety of coal mine personnel and induce the production of coal resource. The abovementioned aspects have high research value and practical significance.
This thesis takes the gas concentration monitoring data of fully mechanized mining watching as the research object. Based on the analysis of the current research status of gas concentration monitoring and early warning methods domestically and abroad, this thesis study the characteristics of the abnormal change trend of gas concentration in fully mechanized mining watching, and confirming the numbers of fully mechanized mining face gas concentration prediction. Based on the said basis, a gated circulation unit (GRU) neural network is used to predict gas concentration in a fully mechanized mining watching. Aiming at the GRU prediction process which the gradient descent algorithm tends to fall into the local optimal problem, the improved Whale Optimization Algorithm (IWOA) is used to optimize the neural network model numbers and predict the gas concentration. Combining the prediction results with the gray correlation analysis of the predicted samples, the IWOA-GRU gas concentration early warning model is established and applied to the early warning of gas concentration in fully mechanized coal mining watching, and the reliability of the early warning was verified by experimental simulation. In this thesis, based on the coal mine KJ95X safety monitoring platform, a fully mechanized mining watching gas concentration early warning system is designed. Through the interactive application of the host computer, database, and early warning model, the real-time prediction of fully mechanized mining watching gas concentration is achieved, and the fully mechanized watching gas concentration. Realizing the data display, prediction, and early warning functions.
In this thesis, through the research of gas concentration data monitoring and early warning methods in fully mechanized mining watching, it can realize accurate prediction of gas concentration in fully mechanized mining face and early warning of abnormal gas rising trend, which can certainly improve and supplement the coal mine safety monitoring system. The abovementioned guarantees the safe production of coal mines and has certain theoretical research value and engineering application value.

[1]张建强, 宁树正, 陈美英， 等. 我国煤炭资源开发前景及对策[J]. 地质论评, 2020, 66(01): 143-145.

[2]仓定帮, 魏晓平, 曹明. 基于多因素分析的我国煤炭消费及节能减排潜力预测[J]. 工业技术经济, 2017, 36(12): 27-32.

[3]宋豪. 资源环境约束下我国未来的煤炭需求预测[D]. 北京: 中国地质大学, 2020.

[4]黄继广, 马汉鹏, 范春姣, 等. 我国煤矿安全事故统计分析及预测[J]. 陕西煤炭, 2020, 39(03): 34-39+6.

[5]宁小亮. 2013-2018年全国煤矿事故规律分析及对策研究[J]. 工矿自动化, 2020, 46(07): 34-41.

[6]孟远, 谢东海, 苏波, 等. 2010年-2019年全国煤矿生产安全事故统计与现状分析[J]. 矿业工程研究, 2020, 35(04): 27-33.

[7]中国煤矿安全生产网. 事故快报[EB/OL]. [2020-08-20]. http://www.mkaq.org/sggl/shigukb/.

[8]王佳奇, 卢明银. 基于数字孪生的煤矿瓦斯事故安全管理[J]. 煤矿安全, 2020, 51(08): 251-255.

[9]张莉. 煤矿安全生产监控系统云联网平台关键技术研究[D]. 北京: 中国矿业大学, 2019.

[10]Ma Yunlong. Design of Coal Mine Power Grid Parameter Monitoring System Based on CAN Bus[J]. IOP Conference Series: Earth and Environmental Science, 2021, 651(03): 317-319.

[11]王学琛. 煤矿安全生产风险评价与预警系统研究[D]. 武汉: 武汉理工大学, 2017.

[12]李益鸣. 煤矿监测监控系统综合评价与应用研究[J]. 中国设备工程, 2019(16): 93-95.

[13]郑学召, 童鑫, 郭军, 等. 煤矿智能监测与预警技术研究现状与发展趋势[J]. 工矿自动化, 2020, 46(06): 35-40.

[14]陈臻, 付文俊. 煤矿巷道掘进作业智能监测与安全预警系统设计[J]. 煤炭技术, 2018, 37(02): 214-216.

[15]李爽, 李丁炜, 犹梦洁. 煤矿安全态势感知预测系统设计及关键技术[J]. 煤矿安全, 2020, 51(05): 244-248.

[16]王道元, 王俊, 孟志斌, 等. 煤矿安全风险智能分级管控与信息预警系统[J/OL]. 煤炭科学技术: 2021, 1-12. http://kns.cnki.net/kcms/detail/11.2402.TD.20200210.0920.004.html.

[17]Cheng Lianhua, Guo Huimin, Lin Haifei. Evolutionary model of coal mine safety system based on multi-agent modeling[J]. Process Safety and Environmental Protection, 2021,147(01): 257-259.

[18]Andrew T. Jebb, Louis Tay. Introduction to Time Series Analysis for Organizational Research[J]. Organizational Research Methods, 2017, 20(01): .

[19]王丽. 时间序列的函数型变系数自回归预测方法研究[D]. 太原: 山西大学, 2016.

[20]李松. 浅析煤与瓦斯突出预警技术及其应用[J]. 当代化工研究, 2021(06): 73-74.

[21]Cao Shugang, Liu Yanbao, Wang Yanping. A forecasting and forewarning model formethane hazard in working face of coal mine based on LS-SVM[J]. International Journal of Mining Science and Tec hnology, 2008, 18(02): 172-176.

[22]关维娟, 张国枢, 赵志根, 等. 煤与瓦斯突出多指标综合辨识与实时预警研究[J]. 采矿与安全工程学报, 2013, 30(06): 922-929.

[23]李振兴, 王小研. 基于BP神经网络的采煤工作面瓦斯涌出量预测[J]. 煤炭工程, 2016, 48(01): 98-100+103.

[24]刘帅, 朱永利, 张科, 等. 基于误差修正ARMA-GARCH模型的短期风电功率预测[J]. 太阳能学报, 2020, 41(10): 268-275.

[25]施人杰, 卢科学. 时间序列分析引论[M]. 西安: 西安电子科技大学出版社, 1988.

[26]周鑫隆, 章光, 吕辰, 等. 深部煤层瓦斯含量的差值GM-RBF预测模型及其应用[J]. 安全与环境学报, 2017, 17(06): 2050-2055.

[27]Xiaoguang Zhao, Yaodong Gu, Jiabin Yu. The Influence of Gender, Age, and Body Mass Index on Arch Height and Arch Stiffness[J]. The Journal of Foot and Ankle Surgery, 2020, 59(02): 112-114.

[28]白梁军, 黄萌, 饶臻, 等. 基于GARCH模型的IGBT寿命预测[J]. 中国电机工程学报, 2020, 40(18): 5787-5796.

[29]陈易平, 俞龙, 谌颃. 大数据环境下基于小波神经网络和ARMA模型的流量异常检测[J]. 重庆理工大学学报(自然科学), 2019, 33(10): 149-154.

[30]樊娇, 冯昊, 牛东晓, 等. 基于小波分析和GM-ARIMA模型的月度售电量预测[J]. 华北电力大学学报(自然科学版), 2015, 42(04): 101-105.

[31]刘彦邦. 煤矿采煤工作面安全管理研究[J]. 中国石油和化工标准与质量, 2020, 40(14): 63-64.

[32]张新建, 刘锋, 李贤功. 基于小波降噪和循环神经网络的煤矿瓦斯浓度预测[J]. 煤炭技术, 2020, 39(09): 145-148.

[33]李绍良, 王茜, 张毅, 等. 基于混沌相空间重构的IGA-LSSVM在线煤矿工作面瓦斯浓度预测模型的研究[J]. 中国煤炭, 2019, 45(05): 41-45+54.

[34]毛毅. 黄陵一号井综采面瓦斯趋势及其融合分析方法研究[D]. 西安: 西安科技大学, 2019.

[35]吴娇娇. 基于时空神经网络模型的瓦斯浓度预测研究[D]. 北京: 中国矿业大学, 2015.

[36]王海波. 大采高综采工作面环境安全智能化监测[J]. 工矿自动化, 2020, 46(07): 12-15.

[37]Shuanfeng Zhao, Pengfei Wang, Shijun Li. Study on the Fault Diagnosis Method of Scraper Conveyor Gear under Time-Varying Load Condition[J]. Applied Sciences, 2020, 10(15): 107-110.

[38]彭林军, 岳宁, 李申龙, 等. 超大采高综采工作面回撤通道支护研究[J/OL]. 煤炭科学技术, 2020, 1-7. http://kns.cnki.net/kcms/detail/11.2402.td.20201217.1411.005.html.

[39]方宇生. 1000kW刮板机减速器箱体设计优化研究[D]. 北京: 中国矿业大学, 2020.

[40]高志扬, 雒赵飞, 景国勋, 等. 综采工作面环境状况灰色熵权聚类分析[J]. 煤炭技术, 2016, 35(06): 145-147.

[41]杨晓东. 保德煤矿81305工作面瓦斯浓度影响因素分析[J]. 煤炭工程, 2015, 47(03): 74-76.

[42]董丁稳. 基于安全监控系统实测数据的瓦斯浓度预测预警研究[D]. 西安: 西安科技大学, 2012.

[43]Bing Qing Tan, Su Xiu Xu, Ray Zhong, et al. Sequential auction based parking space sharing and pricing mechanism in the era of sharing economy[J]. Industrial Management & Data Systems, 2019, 119(08): 257-260.

[44]廖英雷. 基于逻辑回归模型的瓦斯浓度异常值检测与预警[J]. 陕西煤炭, 2020, 39(04): 13-18+44.

[45]李树刚, 马莉, 潘少波, 等. 基于循环神经网络的煤矿工作面瓦斯浓度预测模型研究[J]. 煤炭科学技术, 2020,48(01): 33-38.

[46]李实秋, 李文娟. 信号与系统[M]. 西安: 西安电子科技大学出版社, 2015.

[47]张景贵, 李勇帆, 赵晋琴. 基于短时傅里叶变换分析超短光脉冲的传输特性[J]. 激光技术, 2013, 37(01): 52-55.

[48]俞一彪, 孙兵. 数字信号处理[M]. 南京: 东南大学出版社, 2017.

[49]董田田. 基于改进粒子群小波神经网络的瓦斯的研究及预测[D]. 西安: 西安科技大学, 2016.

[50]Baldazzi Giulia, Solinas Giuliana, Del Valle Jaume, et al. Systematic analysis of wavelet denoising methods for neural signal processing[J]. Journal of Neural Engineering, 2020, 17(06): 335-338.

[51]陈良健, 许建秋. 基于门控循环单元模型的在线路网匹配算法[J]. 华东师范大学学报(自然科学版), 2020, (06): 63-71.

[52]Dutta Shawni, Mandal Jyotsna Kumar, Kim Tai Hoon, et al. Breast Cancer Prediction Using Stacked GRU-LSTM-BRNN[J]. Applied Computer Systems, 2020, 25(02): 277-279.

[53]许飞. 基于机器学习的短期负荷预测[D]. 广州: 华南理工大学, 2020.

[54]谢国民, 单敏柱, 付华. 基于IsoMap和MBFO-SVR的瓦斯涌出量动态预测研究[J]. 传感技术学报, 2016, 29(07): 1115-1120.

[55]Du Pengzhen, Cheng Weiming, Liu Ning, et al. A Modified Whale Optimization Algorithm with Single-Dimensional Swimming for Global Optimization Problems[J]. Symmetry, 2020, 12(11): 331-334.

[56]石旭东, 姜鸿晔. 基于鲸鱼优化小波神经网络的飞行航迹缺失数据预测方法[J]. 计算机应用与软件, 2020, 37(07): 200-205.

[57]Du Pengzhen, Cheng Weiming, Liu Ning, et al. A Modified Whale Optimization Algorithm with Single-Dimensional Swimming for Global Optimization Problems[J]. Symmetry, 2020, 12(11): 228-230.

[58]Chen Yang, Pi Dechang, Xu Yue. Neighborhood Global Learning based Flower Pollination Algorithm and its Application to Unmanned Aerial Vehicle Path Planning[J]. Expert Systems with Applications, 2021(prepublish).

[59]冀伟朋. 煤矿材料库信息管理系统的设计与应用[D]. 郑州: 郑州大学, 2018.

[60]孙岩, 李斌, 罗群. 煤田地质勘探地理信息系统空间数据库管理系统设计[J]. 科学技术与工程, 2021, 21(01): 95-102.

[61]余峥, 鲍正德, 李晨曦. 浅析JDBC连接MySQL的增删改查操作[J]. 计算机系统网络和电信, 2019, 1(02): 333-337.

王诚. 研究JDBC连接MySQL数据库技术[J]. 网络安全和信息化, 2018, (12): 68-69.