煤矿工业数据AI模型自动推理技术

张智星; 付翔; 张小强; 李浩杰; 秦一凡; 刘萌; 孙岩; 贾一帆; 杨宇琪

doi:10.13272/j.issn.1671-251x.18181

煤矿工业数据AI模型自动推理技术

doi: 10.13272/j.issn.1671-251x.18181

张智星^1,,
付翔^{1, 2, 3},
张小强^1, ,,
李浩杰¹,
秦一凡¹,
刘萌¹,
孙岩¹,
贾一帆¹,
杨宇琪¹

1.
太原理工大学矿业工程学院，山西太原　030024
2.
山西焦煤集团有限责任公司博士后工作站，山西太原　030024
3.
智能采矿装备技术全国重点实验室，山西太原　030024

基金项目: 国家自然科学基金项目（52274157）；“科技兴蒙”行动重点专项项目（2022EEDSKJXM010）；国家重点研发计划项目（2020YFB1314004）。

详细信息

作者简介:
张智星（2000—），男，湖北襄阳人，硕士研究生，研究方向为智慧煤矿工业互联网技术，E-mail：2546581521@qq.com

通讯作者:
张小强（1984—），男，山西太原人，副教授，博士，研究方向为煤矿开采、矿山压力及其智能控制，E-mail:tyzxq2009@163.com。

中图分类号: TD67
计量
- 文章访问数: 19
- HTML全文浏览量: 4
- PDF下载量: 3
- 被引次数: 0
出版历程
- 收稿日期: 2024-02-02
- 修回日期: 2024-09-27
- 网络出版日期: 2024-10-17

Automatic reasoning technology for coal mine industrial data AI models

ZHANG Zhixing^1
,,
FU Xiang^{1, 2, 3},
ZHANG Xiaoqiang^{1
, ,},
LI Haojie¹,
QIN Yifan¹,
LIU Meng¹,
SUN Yan¹,
JIA Yifan¹,
YANG Yuqi¹

1.
College of Mining Engineering, Taiyuan University of Technology, Taiyuan 030024, China
2.
Postdoctoral Workstation, Shanxi Coking Coal Group Co., Ltd, Taiyuan 030024, China
3.
State Key Laboratory of Intelligent Mining Equipment Technology, Taiyuan 030024, China

摘要

摘要: 煤矿生产过程的智能化主要依托于人工智能（AI）技术分析煤矿工业数据，但单一应用场景AI模型无法适用于煤矿复杂的应用场景，且仅使用分布式计算来处理AI模型输入特征值会导致模型应用效率降低。针对上述问题，提出了一种煤矿工业数据AI模型自动推理技术。该技术架构包括数据层、计算驱动层和模型推理层：数据层采集各类监测数据并统一存储，为计算驱动层提供原始数据；计算驱动层将数据层采集的海量原始数据转换成煤矿应用场景AI模型输入特征值，通过煤矿应用场景AI模型输入特征值双计算引擎自动切换机制，根据数据量自动合理地选择使用基于Spark的分布式计算方式或基于Python的单机计算方式，解决了海量数据计算速度慢、数据应用延迟大的问题；模型推理层将特征值输入应用场景AI模型进行推理，引入煤矿应用场景AI模型多触发方式协同推理机制，通过定时触发、人为交互触发、信号反馈触发3种触发方式，解决了在煤矿复杂的应用条件下单一应用场景AI模型利用效果差的问题。测试和应用结果表明，该技术可实现多应用场景AI模型输入特征值的快速计算，以及不同应用场景AI模型的快速、自动、协同推理。
- 煤矿人工智能 /
- 煤矿工业数据 /
- AI模型推理 /
- 海量数据计算 /
- AI模型应用
Abstract: The automation of coal mine production processes has largely relied on artificial intelligence (AI) technology to analyze industrial data. However, AI models developed for single application scenarios prove inadequate for the complex environments in coal mining. Relying solely on distributed computing to process the input features of AI models has led to decreased application efficiency. To address these challenges, an automatic reasoning technology for AI models in coal mine industrial data was developed. The system architecture consisted of three layers: the data layer, the computation-driving layer, and the model reasoning layer. The data layer gathered and stored various types of monitoring data, supplying raw data to the computation-driving layer. The computation-driving layer converted this vast amount of raw data into input features for AI models tailored to coal mining applications. An automatic switching mechanism between two computational engines—based on the input feature values—intelligently selected either Spark-based distributed computing or Python-based local computing, depending on the data volume, thereby resolving the issues of slow processing speeds and high latency in large-scale data applications. In the model reasoning layer, the input features were fed into the AI models for reasoning. A collaborative reasoning mechanism, with multiple triggering methods—scheduled, manual, and feedback-triggered—was introduced to enhance the effectiveness of AI models in complex coal mining scenarios. The results demonstrate that this technology enables rapid calculation of input features for multiple AI models across different application scenarios, as well as fast, automatic, and collaborative reasoning.
- coal mine artificial intelligence /
- coal mine industrial data /
- AI model reasoning /
- large-scale data computation /
- AI model application

HTML全文

图 1 煤矿工业数据AI模型自动推理技术架构

Figure 1. Architecture of automatic reasoning technology for coal mine industrial data AI model

下载: 全尺寸图片幻灯片

图 2 煤矿应用场景AI模型输入特征值双计算引擎自动切换机制

Figure 2. Automatic switching mechanism of dual computational engines for AI model input eigenvalues in coal mine application scenario

下载: 全尺寸图片幻灯片

图 3 测试架构

Figure 3. Testing architecture

下载: 全尺寸图片幻灯片

图 4 数据处理速率对比

Figure 4. Comparison of data processing rate

下载: 全尺寸图片幻灯片

表 1 模型输入特征值计算耗时对比

Table 1. Comparison of time consumption for model input eigenvalue

模型	仅单机计算耗时/s	仅分布式计算耗时/s	双计算引擎自动切换耗时/s
模型1	344	100	100
模型2	3	18	3

下载: 导出CSV

表 2 模型部分触发推理记录

Table 2. Partially triggered reasoning record of models

模型	模型推理触发时间	触发方式	推理耗时/s
模型1	2023−09−01T16:15:04	定时触发	1.7
模型2	2023−09−01T16:15:04	定时触发	1.5
模型2	2023−09−01T16:17:36	信号反馈触发	1.5
模型1	2023−09−01T16:19:25	人为交互触发	1.7
模型2	2023−09−01T16:22:36	定时触发	1.5
模型1	2023−09−01T16:29:25	定时触发	1.7

下载: 导出CSV

表 3 应用结果

Table 3. Application results

模型	模型触发次数	特征值计算平均耗时/s	模型调用平均耗时/s
模型1	29420	100.6	1.57
模型2	57620	3.8	1.42

下载: 导出CSV

参考文献(21)

[1]	王国法,庞义辉,任怀伟. 智慧矿山技术体系研究与发展路径[J]. 金属矿山,2022(5):1-9. WANG Guofa,PANG Yihui,REN Huaiwei. Research and development path of smart mine technology system[J]. Metal Mine,2022(5):1-9.
[2]	王国法,赵国瑞,任怀伟. 智慧煤矿与智能化开采关键核心技术分析[J]. 煤炭学报,2019,44(1):34-41. WANG Guofa,ZHAO Guorui,REN Huaiwei. Analysis on key technologies of intelligent coal mine and intelligent mining[J]. Journal of China Coal Society,2019,44(1):34-41.
[3]	张建中,郭军. 智慧矿山工业互联网技术架构探讨[J]. 煤炭科学技术,2022,50(5):238-246. ZHANG Jianzhong,GUO Jun. Discussion on industrial Internet technology framework of smart mine[J]. Coal Science and Technology,2022,50(5):238-246.
[4]	付翔,秦一凡,李浩杰,等. 新一代智能煤矿人工智能赋能技术研究综述[J]. 工矿自动化,2023,49(9):122-131,139. FU Xiang,QIN Yifan,LI Haojie,et al. Summary of research on artificial intelligence empowerment technology for new generation intelligent coal mine[J]. Journal of Mine Automation,2023,49(9):122-131,139.
[5]	付翔,李浩杰,张锦涛,等. 综采液压支架中部跟机多模态人机协同控制系统[J]. 煤炭学报,2024,49(3):1717-1730. FU Xiang,LI Haojie,ZHANG Jintao,et al. Multimodal human-machine collaborative control system for hydraulic supports following the shearer in the middle range of fully mechanized mining face[J]. Journal of China Coal Society,2024,49(3):1717-1730.
[6]	贾思锋,付翔,王然风,等. 液压支架时空区域支护质量动态评价[J]. 工矿自动化,2022,48(10):26-33,81. JIA Sifeng,FU Xiang,WANG Ranfeng,et al. Dynamic evaluation of support quality of hydraulic support in space-time region[J]. Journal of Mine Automation,2022,48(10):26-33,81.
[7]	张锦涛,付翔,王然风,等. 智采工作面中部液压支架集群自动化后人工调控决策模型[J]. 工矿自动化,2022,48(10):20-25. ZHANG Jintao,FU Xiang,WANG Ranfeng,et al. Manual regulation and control decision model of middle hydraulic support cluster automation in the intelligent working face[J]. Journal of Mine Automation,2022,48(10):20-25.
[8]	付翔,王然风,赵阳升. 工作面支架液压系统仿真与稳压供液技术[J]. 煤炭学报,2018,43(5):1471-1478. FU Xiang,WANG Ranfeng,ZHAO Yangsheng. Investigation of hydraulic system simulation and fluid feeding technology with steady pressure of working face supports[J]. Journal of China Coal Society,2018,43(5):1471-1478.
[9]	付翔,王然风,赵阳升. 液压支架群组跟机推进行为的智能决策模型[J]. 煤炭学报,2020,45(6):2065-2077. FU Xiang,WANG Ranfeng,ZHAO Yangsheng. Intelligent decision-making model on the of hydraulic supports group advancing behavior to follow shearer[J]. Journal of China Coal Society,2020,45(6):2065-2077.
[10]	夏向学,晏涛,连会青,等. 基于改进DK算法的煤矿水害逃生三维仿真平台研究[J]. 煤炭技术,2022,41(10):203-206. XIA Xiangxue,YAN Tao,LIAN Huiqing,et al. Research on 3D simulation platform of coal mine water disaster escape based on improved DK algorithm[J]. Coal Technology,2022,41(10):203-206.
[11]	梁耍,王世博,谢洋,等. 基于LSTM的煤层厚度动态预测方法研究[J]. 煤炭科学技术,2021,49(增刊1):150-157. LIANG Shua,WANG Shibo,XIE Yang,et al. Study on dynamic prediction method of coal seam thickness based on LSTM[J]. Coal Science and Technology,2021,49(S1):150-157.
[12]	刘力涛,董淑棠. 基于BP神经网络的采煤机截割自适应调速控制[J]. 煤矿机械,2020,41(8):197-199. LIU Litao,DONG Shutang. Adaptive speed regulation control of shearer cutting based on BP neural network[J]. Coal Mine Machinery,2020,41(8):197-199.
[13]	王星,高峰,陈吉,等. 基于GAN网络的煤岩图像样本生成方法[J]. 煤炭学报,2021,46(9):3066-3078. WANG Xing,GAO Feng,CHEN Ji,et al. Generative adversarial networks based sample generation of coal and rock images[J]. Journal of China Coal Society,2021,46(9):3066-3078.
[14]	李波,丁剑明. 基于大数据分析的煤矿综采生产数据应用[J]. 煤矿机械,2017,38(8):170-172. LI Bo,DING Jianming. Application of fully mechanized mining data in coal mine based on large data analysis[J]. Coal Mine Machinery,2017,38(8):170-172.
[15]	乔伟,靳德武,王皓,等. 基于云服务的煤矿水害监测大数据智能预警平台构建[J]. 煤炭学报,2020,45(7):2619-2627. QIAO Wei,JIN Dewu,WANG Hao,et al. Development of big data intelligent early warning platform for coal mine water hazard monitoring based on cloud service[J]. Journal of China Coal Society,2020,45(7):2619-2627.
[16]	申琢,谭章禄. 基于数据挖掘的煤矿大数据可视化管理平台研究[J]. 中国煤炭,2016,42(12):86-89,128. doi: 10.3969/j.issn.1006-530X.2016.12.017 SHEN Zhuo,TAN Zhanglu. Research on big data visual management platform of coal mine based on data mining[J]. China Coal,2016,42(12):86-89,128. doi: 10.3969/j.issn.1006-530X.2016.12.017
[17]	崔卫锋,田野,李旭,等. 煤矿综采工作面智能服务大数据决策平台[J]. 煤矿机械,2022,43(10):193-195. CUI Weifeng,TIAN Ye,LI Xu,et al. Intelligent service big data decision-making platform for fully mechanized mining face in coal mine[J]. Coal Mine Machinery,2022,43(10):193-195.
[18]	周江,王伟平,孟丹,等. 面向大数据分析的分布式文件系统关键技术[J]. 计算机研究与发展,2014,51(2):382-394. doi: 10.7544/issn1000-1239.2014.20120863 ZHOU Jiang,WANG Weiping,MENG Dan,et al. Key technology in distributed file system towards big data analysis[J]. Journal of Computer Research and Development,2014,51(2):382-394. doi: 10.7544/issn1000-1239.2014.20120863
[19]	姚军,管米利,乔帆. 基于Spark的煤矿安全预警系统[J]. 现代电子技术,2022,45(12):49-54. YAO Jun,GUAN Mili,QIAO Fan. Coal-mine safety warning system based on Spark[J]. Modern Electronics Technique,2022,45(12):49-54.
[20]	吴海波,施式亮,念其锋. 基于Spark Streaming流回归的煤矿瓦斯浓度实时预测[J]. 中国安全生产科学技术,2017,13(5):84-89. WU Haibo,SHI Shiliang,NIAN Qifeng. Real-time prediction of gas concentration in coal mine based on Spark Streaming Linear Regression[J]. Journal of Safety Science and Technology,2017,13(5):84-89.
[21]	邵严. 不定期受控RS485总线通信方式在煤矿安全监控系统中的应用[J]. 煤矿安全,2020,51(7):117-120. SHAO Yan. Application of communication mode of irregularly controlled RS485 bus in monitoring system for coal mine safety[J]. Safety in Coal Mines,2020,51(7):117-120.