[1]曹梅,杨超宇.基于小波的CNN-LSTM-Attention瓦斯预测模型研究*[J].中国安全生产科学技术,2023,19(9):69-75.[doi:10.11731/j.issn.1673-193x.2023.09.010]
CAO Mei,YANG Chaoyu.Research on CNN-LSTM-Attention gas prediction model based on wavelet[J].JOURNAL OF SAFETY SCIENCE AND TECHNOLOGY,2023,19(9):69-75.[doi:10.11731/j.issn.1673-193x.2023.09.010]
点击复制
基于小波的CNN-LSTM-Attention瓦斯预测模型研究*
CAO Mei,YANG Chaoyu.Research on CNN-LSTM-Attention gas prediction model based on wavelet[J].JOURNAL OF SAFETY SCIENCE AND TECHNOLOGY,2023,19(9):69-75.[doi:10.11731/j.issn.1673-193x.2023.09.010]
《中国安全生产科学技术》[ISSN:1673-193X/CN:11-5335/TB]
- 卷:
- 19
- 期数:
- 2023年9期
- 页码:
- 69-75
- 栏目:
- 职业安全卫生管理与技术
- 出版日期:
- 2023-09-30
文章信息/Info
- Title:
- Research on CNN-LSTM-Attention gas prediction model based on wavelet
- 文章编号:
- 1673-193X(2023)-09-0069-07
- Keywords:
- wavelet analysis; convolutional neural network (CNN); long short-term memory neural network (LSTM); attention mechanism
- 分类号:
- X936；TD712
- 文献标志码:
- A
- 摘要:
- 为提高非平稳性瓦斯浓度预测精度，基于小波分析方法，分解重构瓦斯时序数据，分离高、低频率特征子序列，组成输入特征矩阵，引入卷积神经网络（CNN）获取多特征输入空间联系，采用长短期记忆网络（LSTM）提取序列时序变化特征，添加注意力机制（Attention）为LSTM输出自适应分配权重，增强瓦斯浓度关键信息提取，构建基于小波的CNN-LSTM-Attention瓦斯浓度预测模型。研究结果表明：基于小波的CNN-LSTM-Attention模型瓦斯浓度预测误差MAE为0.001 898，RMSE为0.002 576，模型预测精度均高于其他单一模型。研究结果可为瓦斯预测预警提供参考依据。
- Abstract:
- In order to improve the prediction accuracy of nonstationary gas concentration,the time sequence data of gas was decomposed and reconstructed based on the wavelet analysis method,and the input feature matrix was formed by separating the high and low frequency feature sub-sequences.The convolutional neural network (CNN) was introduced to obtain the multi-feature input spatial association,and the long short-term memory neural network (LSTM) was applied to extract the sequence time series variation features.The attention mechanism (Attention) was added to adaptively assign weights to the LSTM output,so as to enhance the extraction of key information of gas concentration,and a CNN-LSTM-Attention gas concentration prediction model based on wavelet was established.The results show that the prediction error MAE and RMSE of CNN-LSTM-Attention model based on wavelet are 0.001 898 and 0.002 576,and the prediction accuracy of the model is higher than other single models.The research results can provide a reference basis for the gas prediction and early warning.
参考文献/References:
［1］张培森,李复兴,朱慧聪,等.2008—2020年煤矿事故统计分析及防范对策［J］.矿业安全与环保,2022,49(1):128-134.
ZHANG Peisen,LI Fuxing,ZHU Huicong,et al.Statistical analysis and prevention countermeasures of coal mine accidents from 2008 to 2020［J］.Mining Safety & Environmental Protection,2022,49(1):128-134.
［2］何俊,刘林兵,唐一举.基于线性二次指数平滑法的瓦斯含量预测［J］.煤炭科学技术,2014,42(12):48-50，61.
HE Jun,LIU Linbing,TANG Yiju.Gas content prediction based on linear double exponential smooth method［J］.Coal Science and Technology,2014,42(12):48-50,61.
［3］邓君.基于灰色理论的煤与瓦斯突出危险性预测及防控技术研究［D］.湘潭：湖南科技大学,2019.
［4］王鹏,伍永平,王栓林,等.矿井瓦斯浓度Lagrange-ARIMA实时预测模型研究［J］.煤炭科学技术,2019,47(4):141-146.
WANG Peng,WU Yongping,WANG Shuanlin,et al.Study on Lagrange-ARIMA real-time prediction model of mine gas concentration［J］.Coal Science and Technology,2019,47(4):141-146.
［5］李金生.基于ARIMA时间序列的瓦斯涌出量预测研究［J］.陕西煤炭,2020,39(S2):6-10.
LI Jinsheng.Prediction of gas emission rate based on ARIMA time series［J］.Shaanxi Coal,2020,39(S2):6-10.
［6］郭思雯,陶玉帆,李超.基于时间序列的瓦斯浓度动态预测［J］.工矿自动化,2018,44(9):20-25.
GUO Siwen,TAO Yufan,LI Chao.Dynamic prediction of gas concentration based on time series［J］.Journal of Mine Automation,2018,44(9):20-25.
［7］范京道,黄玉鑫,闫振国,等.ARIMA-SVM组合模型驱动下的瓦斯浓度预测研究［J］.工矿自动化,2022,48(9):134-139.
FAN Jingdao,HUANG Yuxin,YAN Zhenguo,et al.Research on gas concentration prediction driven by ARIMA-SVM combined model［J］.Journal of Mine Automation,2022,48(9):134-139.
［8］赖祥威,夏云霓,郑万波,等.基于集成学习的改进灰色瓦斯浓度序列预测［J］.中国安全生产科学技术,2021,17(7):16-21.
LAI Xiangwei,XIA Yunni,ZHENG Wanbo,et al.Improved grey prediction of gas concentration sequence based on integrated learning［J］.Journal of Safety Science and Technology,2021,17(7):16-21.
［9］肖鹏,谢行俊,双海清,等.小波-极限学习机在瓦斯涌出量时变序列预测中的应用［J］.西安科技大学学报,2020,40(5):839-845.
XIAO Peng,XIE Xingjun,SHUANG Haiqing,et al.Application of wavelet-extreme learning machine in time-varying series prediction of gas emission quantity［J］.Journal of Xi’an University of Science and Technology,2020,40(5):839-845.
［10］郝治朝.煤矿隐伏断层定量预测方法与应用研究［D］.北京:中国矿业大学(北京),2021.
［11］刘超,雷晨,李树刚,等.基于CNN-GRU的瓦斯浓度预测模型及应用［J］.中国安全生产科学技术,2022,18(9):62-68.
LIU Chao,LEI Chen,LI Shugang,et al.Prediction model of gas concentration based on CNN-GRU and its application［J］.Journal of Safety Science and Technology,2022,18(9):62-68.
［12］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(1):33-38.
LI Shugang,MA Li,PAN Shaobo,et al.Research on prediction model of gas concentration based on RNN in coal mining face［J］.Coal Science and Technology,2020,48(1):33-38.
［13］刘剑桥.基于改进LSTM循环神经网络瓦斯数据时间序列预测研究［D］.徐州：中国矿业大学,2018.
［14］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(9):145-148.
ZHANG Xinjian,LIU Feng,LI Xiangong.Coal mine gas concentration prediction based on wavelet denoising and recurrent neural network［J］.Coal Technology,2020,39(9):145-148.
［15］谢谦,董立红,吴雪菲.基于多源数据融合的煤矿工作面瓦斯浓度预测［J］.中国安全生产科学技术,2022,18(11):71-76.
XIE Qian,DONG Lihong,WU Xuefei.Prediction of gas concentration in coal mining face based on multi-source data fusion［J］.Journal of Safety Science and Technology,2022,18(11):71-76.
［16］赵辉,杨赛,岳有军,等.基于小波分解-卷积神经网络和支持向量回归的短期负荷预测［J］.科学技术与工程,2021,21(25):10718-10724.
ZHAO Hui,YANG Sai,YUE Youjun,et al.Short-term load forecasting of convolutional neural network support vector regression using wavelet decomposition［J］.Science Technology and Engineering,2021,21(25):10718-10724.
［17］梁运培,栗小雨,李全贵,等.基于CS-LSTM的工作面瓦斯浓度智能预测研究［J］.矿业安全与环保,2022,49(4):80-86.
LIANG Yunpei,LI Xiaoyu,LI Quangui,et al.Research on intelligent prediction of gas concentration in working face based on CS-LSTM［J］.Mining Safety & Environmental Protection,2022,49(4):80-86.
［18］CHAUDHARI S,MITHAL V,POLATKAN G,et al.An attentive survey of attention models［J］.ACM Transactionson Intelligent Systems and Technology,2021,12(5):53.
相似文献/References:
[1]吴轩.减阻剂对管道内流动结构及安全性影响研究[J].中国安全生产科学技术,2015,11(5):51.[doi:10.11731/j.issn.1673-193x.2015.05.008]
WU Xuan.Influence on flow structures and safety of pipeline by drag-reducing additives[J].JOURNAL OF SAFETY SCIENCE AND TECHNOLOGY,2015,11(9):51.[doi:10.11731/j.issn.1673-193x.2015.05.008]
[2]余琼芳,徐静,杨艺.基于CNN_LSTM模型的复杂支路故障电弧检测*[J].中国安全生产科学技术,2022,18(4):204.[doi:10.11731/j.issn.1673-193x.2022.04.029]
YU Qiongfang,XU Jing,YANG Yi.Fault arc detection of complex branch based on CNN_LSTM model[J].JOURNAL OF SAFETY SCIENCE AND TECHNOLOGY,2022,18(9):204.[doi:10.11731/j.issn.1673-193x.2022.04.029]
[3]曹亚利,李振雷,刘旭东,等.基于卷积神经网络的冲击地压预警方法研究*[J].中国安全生产科学技术,2022,18(10):101.[doi:10.11731/j.issn.1673-193x.2022.10.015]
CAO Yali,LI Zhenlei,LIU Xudong,et al.Research on early-warning method of rockburst based on convolutional neural network[J].JOURNAL OF SAFETY SCIENCE AND TECHNOLOGY,2022,18(9):101.[doi:10.11731/j.issn.1673-193x.2022.10.015]
[4]李波,李鹏,高莲,等.基于PCA-VMD-CNN的输电线路覆冰重量预测模型*[J].中国安全生产科学技术,2022,18(10):216.[doi:10.11731/j.issn.1673-193x.2022.10.032]
LI Bo,LI Peng,GAO Lian,et al.Prediction model for weight of ice coating on transmission line based on PCA-VMD-CNN[J].JOURNAL OF SAFETY SCIENCE AND TECHNOLOGY,2022,18(9):216.[doi:10.11731/j.issn.1673-193x.2022.10.032]
[5]李子奇,蒋柱虎,王力,等.基于深度学习的工程结构损伤识别研究进展[J].中国安全生产科学技术,2022,18(12):43.[doi:10.11731/j.issn.1673-193x.2022.12.006]
LI Ziqi,JIANG Zhuhu,WANG Li,et al.Research progress in damage identification of engineering structure based on deep learning[J].JOURNAL OF SAFETY SCIENCE AND TECHNOLOGY,2022,18(9):43.[doi:10.11731/j.issn.1673-193x.2022.12.006]
[6]凌晓,王昕越,郭凯,等.基于UNet模型燃气管道高后果区分割方法研究*[J].中国安全生产科学技术,2024,20(4):157.[doi:10.11731/j.issn.1673-193x.2024.04.022]
LING Xiao,WANG Xinyue,GUO Kai,et al.Research on segmentation method for high-consequence areas of gas pipeline based on UNet model[J].JOURNAL OF SAFETY SCIENCE AND TECHNOLOGY,2024,20(9):157.[doi:10.11731/j.issn.1673-193x.2024.04.022]
[7]王文标,时启衡,郝友维.基于改进SqueezeNet的火焰识别算法*[J].中国安全生产科学技术,2024,20(8):19.[doi:10.11731/j.issn.1673-193x.2024.08.003]
WANG Wenbiao,SHI Qiheng,HAO Youwei.Flame recognition algorithm based on improved SqueezeNet[J].JOURNAL OF SAFETY SCIENCE AND TECHNOLOGY,2024,20(9):19.[doi:10.11731/j.issn.1673-193x.2024.08.003]
[8]梁学栋,王霞.考虑多利益相关群体偏好冲突的应急响应协同决策研究*[J].中国安全生产科学技术,2024,20(10):46.[doi:10.11731/j.issn.1673-193x.2024.10.006]
LIANG Xuedong,WANG Xia.Research on collaborative decision-making of emergency response considering preference conflicts among multiple stakeholder groups[J].JOURNAL OF SAFETY SCIENCE AND TECHNOLOGY,2024,20(9):46.[doi:10.11731/j.issn.1673-193x.2024.10.006]
[9]解恒星,张雄,董锦洋,等.基于CNN_BiLSTM的矿井瓦斯涌出量预测模型*[J].中国安全生产科学技术,2024,20(11):53.[doi:10.11731/j.issn.1673-193x.2024.11.007]
XIE Hengxing,ZHANG Xiong,DONG Jinyang,et al.Prediction model of mine gas emission based on CNN_BiLSTM[J].JOURNAL OF SAFETY SCIENCE AND TECHNOLOGY,2024,20(9):53.[doi:10.11731/j.issn.1673-193x.2024.11.007]
[10]程磊,张俊展,景国勋,等.基于YOLOv8的煤矿安全帽和安全背心检测算法*[J].中国安全生产科学技术,2025,21(2):115.[doi:10.11731/j.issn.1673-193x.2025.02.015]
CHENG Lei,ZHANG Junzhan,JING Guoxun,et al.Detection algorithm of safety hat and safety vest in coal mine based on YOLOv8[J].JOURNAL OF SAFETY SCIENCE AND TECHNOLOGY,2025,21(9):115.[doi:10.11731/j.issn.1673-193x.2025.02.015]
备注/Memo
- 备注/Memo:
-
收稿日期： 2023-03-03
* 基金项目： 国家自然科学基金项目（61873004）
作者简介： 曹梅，硕士研究生，主要研究方向为数据分析与数据挖掘、煤矿安全管理。
通信作者： 杨超宇，博士，教授，主要研究方向为煤矿安全信息化技术应用。
