工矿自动化

Advancements and applications: In-situ monitoring technology for overburden movement in mining

ZHU Weibing, WANG Xiaozhen, XIE Jianlin, ZHAO Bozhi, NING Shan, XU Jialin

2023, 49(9): 1-12. doi: 10.13272/j.issn.1671-251x.18136

<Abstract>(1829) <HTML> (80) <PDF>(67)

Abstract:
The in-situ monitoring technology for overburden movement in mining has the features of adapting to complex geological conditions in deep shaft high pressure water and thick coal seam mining, multi layer dynamic monitoring, and high-precision remote real-time online transmission. It provides effective data support for mining enterprises to carry out roof disaster prevention and control. Starting from the practical background of coal mining application, this paper systematically reviews the development process, technological progress, and application effects of in-situ monitoring technology for overburden movement in mining. Based on the development history of mining pressure theory and overburden movement monitoring technology in China, this paper comprehensively introduces the important stages of in-situ monitoring technology for overburden movement in mining. It elaborates on the theoretical innovation and technological breakthroughs achieved by this technology in three aspects: multidimensional real-time collaborative monitoring, unmanned online monitoring, and deep rock movement monitoring. Based on the monitoring engineering examples of coal mines such as Bulianta Coal Mine, Tongxin Coal Mine, and Gaojiabao Coal Mine, the effectiveness of in-situ monitoring technology for overburden movement in mining is demonstrated in practical engineering applications. The application prospects of this technology in different types of mining areas and research fields are discussed. It is pointed out that the development trend of in-situ monitoring technology for overburden movement in mining is precision, intelligence, and integration, namely optimizing sensor performance and layout plans to improve monitoring precision and accuracy, using artificial intelligence, big data, and Internet of things technology to achieve automatic analysis and prediction, combining the in-situ monitoring technology with other technologies to form a complete monitoring system.

Construction and development trends of intelligent mining basic platform

YU Yang, ZHANG Shen

2023, 49(9): 13-22, 121. doi: 10.13272/j.issn.1671-251x.2023070023

<Abstract>(1278) <HTML> (79) <PDF>(68)

Abstract:
Intelligent mines are the continuation of the development of digital mines and integrated automation systems. Compared to digital mines and integrated automation systems, intelligent mines have higher requirements for basic platforms. The basic platforms of intelligent mine are divided into two parts: network platform and data platform. The network platform is divided into backbone network and access network. The backbone network has gone through the development process of industrial bus network, 100 Mbit/s industrial Ethernet, 1000 Mbit/s industrial Ethernet, and 10 Gbit/s industrial Ethernet. This study analyzes the development process, advantages, disadvantages, and applicability of the backbone network from industrial bus to 10 Gbit/s industrial Ethernet. It is pointed out that industrial bus network and 100 Mbit/s industrial Ethernet are not suitable as intelligent mine backbone networks. The 1000 Mbit/s and 10 Gbit/s industrial Ethernet are currently the preferred backbone networks for intelligent mines. By analyzing the construction requirements of intelligent mine access networks, it is pointed out that the access networks of intelligent mines should have blind spot free access and underlying computing capabilities. Currently, wireless access networks are still difficult to possess these capabilities. Leakage communication system belongs to semi wireless mode, the application situation is limited, and the speed is not high. 5G is mainly a wireless transmission network with relatively flexible rate adaptability. It is suitable to be a pure access transmission network. Due to its lack of ad hoc network and underlying computing capabilities, it is limited in blind spot free monitoring applications. WSN has a certain degree of self-sufficient network and underlying computing capabilities. But its speed is relatively low. Especially when used in underground mines with multiple hops, the speed rate significantly decreases, and the power consumption increases, thereby reducing computing capabilities and self-organizing network capabilities. Therefore, new access network devices that meet the needs of intelligent mines need to be developed. The data platform approach based on discrete servers and simple virtual servers can no longer meet the requirements of intelligent mines for data platforms. This paper analyzes the architecture and key technologies of the data platform's hyperconverged service platform, the characteristics of hyperconverged services, and its adaptability to the construction of intelligent mines. It is pointed out that hyperconverged servers are the development direction of intelligent mine data platforms.

Summary of research on health status assessment of fully mechanized mining equipment

CAO Xiangang, DUAN Yong, ZHAO Jiangbin, YANG Xin, ZHAO Fuyuan, FAN Hongwei

2023, 49(9): 23-35, 97. doi: 10.13272/j.issn.1671-251x.18143

<Abstract>(1291) <HTML> (73) <PDF>(81)

Abstract:
Fully mechanized mining equipment is gradually becoming larger, more complex and more intelligent. The traditional equipment management methods of regular maintenance and post maintenance are no longer able to meet the high reliability requirements of equipment operation in coal mine intelligent construction. Therefore, studying the relevant theories and technologies of fully mechanized equipment health status assessment has great practical significance for coal mine intelligent mining. This paper proposes the scope definition of fully mechanized mining equipment health status assessment and the fully mechanized mining equipment health status assessment process. This paper summarizes the research status and development trends of comprehensive mining equipment health status assessment methods from four aspects: signal acquisition, feature extraction and fusion, health status level classification, and health status assessment model establishment. The current challenges faced by fully mechanized mining equipment health status assessment related technologies are analyzed from aspects such as signal acquisition and sensor optimization layout, data processing and feature extraction, establishment of health status assessment models, and application of fully mechanized mining equipment status assessment. In response to the current research status and challenges mentioned above, the development trend of fully mechanized mining equipment health status assessment technology is discussed from the aspects of improving data collection schemes and fault mechanism research methods, building high-performance big data computing platforms, establishing deep learning assessment models, researching dynamic evaluation models for fully mechanized mining equipment health status, and developing fully mechanized mining equipment health status assessment systems. It is pointed out that in the process of coal mine intelligence, it is necessary to ensure that the theoretical research, algorithm development, and engineering application of fully mechanized mining equipment health status assessment go hand in hand.

Research on video AI recognition technology for abnormal state of coal mine belt conveyors

MAO Qinghua, GUO Wenjin, ZHAI Jiao, WANG Rongquan, SHANG Xinmang, LI Shikun, XUE Xusheng

2023, 49(9): 36-46. doi: 10.13272/j.issn.1671-251x.18134

<Abstract>(1291) <HTML> (161) <PDF>(138)

Abstract:
Traditional belt conveyor abnormal state recognition uses manual inspection or mechanical comprehensive protection system for detection. The manual inspection is labor-intensive, inefficient, and difficult to accurately detect faults. Mechanical comprehensive protection system is prone to misjudgment and poor recognition effect. The above methods can no longer meet the needs of coal industry intelligence. With the development of machine vision, deep learning, and industrial Ethernet technology, video AI technology has become a research hotspot for intelligent recognition of abnormal states of coal mine belt conveyors. This paper analyzes the current research status of using video AI technology to identify abnormal states of coal mine belt conveyors, such as belt deviation, idler failure, personnel invasion, unsafe behavior of personnel, coal stacking, and foreign objects. It is pointed out that there are three main problems in the current video AI recognition technology for abnormal states of coal mine belt conveyors: long construction time-consumption of video image datasets, low precision of abnormal state recognition, and large time delay in video information transmission. To address the issue of long construction time-consumption of video image datasets, a solution is proposed to strengthen the research on video AI recognition algorithms based on semi supervised, unsupervised, and small sample learning, and to expand the dataset based on generative models. To address the issue of low precision of abnormal state recognition, a solution is proposed to strengthen research on data deblurring methods, and to utilize algorithms such as generative adversarial networks to balance positive and negative samples, and improve AI recognition algorithms. To address the issue of large time delay in video information transmission, a solution is proposed to build a "cloud-edge-end" collaborative video AI recognition system architecture for abnormal states of belt conveyors, and to deploy a high bandwidth and low time delay network communication system. This article looks forward to the development trend of video AI recognition technology for abnormal states of belt conveyors from four aspects: high-performance video AI recognition algorithms, high bandwidth and low time delay video communication technology, "cloud-edge-end" efficient collaborative video AI recognition system, and sound video AI recognition technology standards.

Research on the dynamic law of automatic following of hydraulic support

REN Huaiwei, ZHANG Shuai, XUE Guohua, ZHAO Shuji, ZHANG Yuliang, LI Jian

2023, 49(9): 47-54. doi: 10.13272/j.issn.1671-251x.18133

<Abstract>(1047) <HTML> (113) <PDF>(40)

Abstract:
Currently, research on the automatic following of hydraulic support in fully mechanized working faces focuses on controlling the movement of hydraulic support based on external environmental variables. It cannot maintain good results during long-term operation of the working face. Considering that changes of the external environment will ultimately be reflected in the hydraulic system, it is proposed to predict the following action time of hydraulic supports based on the pressure features of hydraulic system. Taking S1204 working face of Shaanxi Coal Mining Group Shenmu Ningtiaota Mining Co., Ltd. as the engineering background, a hydraulic system model for single support and a hydraulic system model for working face are established using AMESim software. Through simulation analysis, the dynamic change laws of parameters such as pressure and stroke of the pushing hydraulic cylinder during the simultaneous movement of advancing and sliding are obtained. It is found that the inlet pressure of pushing hydraulic cylinder is linearly related to the pulling time during the automatic following process of hydraulic support . It explains that the following action time can be predicted through the inlet pressure. The hydraulic data acquisition system for the working face is developed and installed on the experimental working face. The real-time pressure at the inlet of the pushing hydraulic cylinder during the working process of hydraulic support is obtained. The corresponding advancing time is calculated. It is found that the two have a strong linear correlation, which is consistent with the simulation results. The linear fitting method is used to obtain the relationship between the inlet pressure and the pulling time, achieving the prediction of pulling time based on hydraulic system pressure. It improves the accuracy of automatic following and reduceg reduceg the manual adjustment rate.

Research on intelligent linkage regulation and control of local ventilation in long distance heading face

WANG Lei, WANG Kai

2023, 49(9): 55-63. doi: 10.13272/j.issn.1671-251x.18117

<Abstract>(1120) <HTML> (59) <PDF>(33)

Abstract:
The existing research on ventilation control for long-distance heading face is mostly limited to the frequency conversion of local ventilation fans themselves. There is few research on the on-demand wind supply direction for long-distance heading face. In order to solve the above problem, a design scheme for an intelligent regulation and control system for local ventilation in long-distance heading face is proposed. The system consists of an underground monitoring system, a ventilation anomaly control system, and a ground workstation. The underground monitoring system achieves early warning of abnormal working conditions of local ventilation fans, dynamic analysis of air leakage inside the air ducts, and dynamic prediction of actual supply and demand air volume by real-time monitoring of underground ventilation fans, air ducts, and working surfaces. The ventilation anomaly control system identifies underground abnormal ventilation parameters, develops risk collaborative disposal strategies for different parameters and levels, and displays the specific distribution of parameters such as gas concentration in the heading face in real-time. The ground workstation excavates the potential laws between the air flow status of the heading face and the parameters of the air duct, forming an air flow-air duct control model. Real time control of the air duct is achieved based on the distribution status of the air flow in the heading face. At the same time, the workstation establishes a variable frequency predictive air regulation model that matches the theoretical air supply, actual air supply, and actual air demand. Based on the variable frequency air regulation model and real-time ventilation parameters, the operating frequency of the fan is determined. The on-demand air supply is achieved through intelligent variable frequency of the fan. In abnormal ventilation situations, based on the prediction of gas emission and the limit capacity of air venting gas, supplemented by drilling gas extraction to control the concentration of gas in the working face, the ventilation safety guarantee for the long-distance heading face is achieved. Taking the 23303 heading face of Zhuanlongwan Coal Mine as an example, numerical simulation of air flow is conducted to study the distribution status of air flow in the face. It provides a basis for adjusting the layout of wind speed sensors in the heading face. The paper proposes two different ventilation linkage control methods, namely variable frequency wind regulation under normal conditions and regulating exhaust air under abnormal conditions. The operating frequency of the fan is determined through supply and demand matching analysis under normal conditions to achieve intelligent variable frequency air regulation of the fan. In case of abnormal ventilation, four regulation and exhaust rules are adopted to ensure the ventilation safety of long-distance heading faces, while achieving the effect of energy conservation and emission reduction. A comprehensive evaluation system for the health indicators of the local ventilation system has been constructed. Through a comprehensive evaluation model and health index, real-time health "physical examination" of the local ventilation system is achieved, and the risk levels of different indicators are quantitatively displayed to ensure that the local ventilation system is in a healthy state.

Research on the influence of roadway obstacles on the position of wind speed monitoring

ZHANG Jingzhao, XIONG Shuai, FAN Jingdao, YAN Zhenguo, HUANG Yuxin, ZHANG Yashuang

2023, 49(9): 64-72. doi: 10.13272/j.issn.1671-251x.2023020040

<Abstract>(1050) <HTML> (50) <PDF>(27)

Abstract:
The existing high-precision wind speed sensors are uniformly installed in the coal mines under normal airflow conditions. It does not consider the abnormal airflow caused by obstacles placed in the roadway. It cannot meet the wind speed precision requirements of intelligent ventilation and it is difficult to achieve safe production in the mine. In order to solve the above problems, taking the 11218 return air roadway of Xiaojihan Coal Mine as the research object, the influence of different positions and sizes of obstacles in the underground roadway on wind speed is studied. Based on on-site measured roadway basic parameters and Fluent software, a roadway model is constructed that fits the features of the mine. The influence of factors such as the distance between the obstacle placed on the floor at a distance of 10 meters from the upstream port and the two sides (referred to as the distance L), its shape, size, and position on the monitoring position of roadway wind speed is studied. ① The quantitative analysis results show that there are small reasonable wind speed regions at the right angles of the cross-section for each model. The maximum area is when L=0.5 m, followed by when L=1 m, and the minimum area is when L=0 m. As the distance L increases, the optimal placement position of the wind speed sensor follows a uniform distribution with the increase of the x-coordinate (roadway direction) - a trace distribution at the right angle of the cross-section - a hollow rounded rectangle distribution pattern. The reasonable airflow diffuses faster towards the two sides. When L=0 m, the reasonable airflow distribution of the vertical line in the roof position is at 2.59-2.78 m. When L=0.5 m, the reasonable airflow distribution of the vertical line in the roof position is between 2.59-2.80. When L=1 m, the reasonable airflow distribution of the vertical line in the roof position is 2.61-2.78 m. ② The qualitative analysis results indicate that the average wind speed in the roadway with obstacles is in a state of increase - decrease - increase - balance. The vertical placement or increase in width of obstacles has a significant impact on wind flow. The volume of obstacles is the same, and the peak wind speed is roughly the same. When the wind flow develops steadily, the wind speed reliability is highest at L=0.5 m, followed by L=1 m, and the reliability is lowest at L=0 m. ③ Through the analysis of wind speed universality, it can be concluded that under the same model, different wind speed change rates are in four stages of ascending - descending - ascending - balancing. Under the condition of model 2 and spacing L=0.5 m, the conclusion that the influence on the air flow transport law of the return air roadway is relatively small has wind speed universality.

Visible and infrared image fusion algorithm for underground personnel detection

ZHOU Libing, CHEN Xiaojing, JIA Wenqi, WEI Jianjian, YE Baisong, ZOU Sheng

2023, 49(9): 73-83. doi: 10.13272/j.issn.1671-251x.2023070025

<Abstract>(1281) <HTML> (71) <PDF>(65)

Abstract:
The working environment and lighting conditions of mining intelligent vehicles are complex. When detecting underground personnel, infrared reflection information and detailed texture information can be fused into visible light images by fusing visible and infrared images to improve the target detection effect. Traditional visible and infrared image fusion methods can lead to blurring of image edges and textures as the number of decomposition layers increases, and the fusion time also increases. At present, deep learning based fusion methods for visible and infrared images are difficult to balance the features in visible and infrared images, resulting in blurred detail information in the fused images. In order to solve the above problems, the image fusion algorithm based on multiple attention modules (IFAM) is proposed. Firstly, convolutional neural networks are used to extract image features from visible and infrared images. Secondly, the extracted features are cross fused using spatial attention and channel attention modules. The fusion weights of the output features of the two attention modules are calculated using the gradient information in the features. The output features of the two attention modules are fused based on the weights. Finally, the image features are restored through deconvolution transformation to obtain the final fused image. The fusion results on the RoadScene dataset and TNO dataset indicate that the IFAM fused image contains both background texture information from visible light images and personnel contour feature information from infrared images. The fusion results on the underground dataset indicate that in low lighting environments, infrared images can compensate for the shortcomings of visible light and are not affected by other light sources in the environment. In low lighting conditions, the personnel contour in the fused image is still obvious. The comparative analysis results show that the information entropy (EN), standard deviation (SD), gradient fusion metric (Q^AB/F), visual information fidelity for fusion (VIFF), and the union structural similarity index measure (SSIM_u) of the image after IFAM fusion are 4.901 3, 88.521 4, 0.169 3, 1.413 5, and 0.806 2, respectively. The overall performance is superior to similar algorithms such as LLF-IOI and NDM.

Research on personnel re-recognition method in coal mine underground based on improved metric learning

ZHANG Liya, WANG Yu, HAO Bonan

2023, 49(9): 84-89, 166. doi: 10.13272/j.issn.1671-251x.18100

<Abstract>(1462) <HTML> (91) <PDF>(44)

Abstract:
In the traditional personnel re-recognition method in coal mine underground based on metric learning, because metric learning ignores the absolute distance between positive and negative samples, the gradient of the loss function disappears or disperses. This results in low recognition precision of underground personnel position information. In order to solve this problem, a personnel re-recognition method in coal mine underground based on improved metric learning is proposed. Firstly, a feature extraction method for underground personnel based on manual design features is adopted to manually process and extract features such as color space and texture space, enriching the feature dimensions. Secondly, Euclidean distance is used to calculate the similarity of high-dimensional features of personnel. Finally, an improved triple loss function is proposed. Adding adaptive weights to the traditional triple loss function increases the weight of effective samples. It solves the problem of gradient disappearance or dispersion caused by ignoring the absolute distance between positive and negative samples. The traditional recognition method is compared with the personnel re-recognition method in coal mine underground based on improved metric learning for cumulative matching feature curve verification and recognition rate verification. The results show the following points. ① The personnel re-recognition method in coal mine underground based on improved metric learning has a sample matching probability of 100% when the number of similar samples is around 50. ② The personnel re-recognition method in coal mine underground based on improved metric learning reduces the inference time of two different calibration size images by 44 ms and 68 ms, respectively, compared to traditional re-recognition methods. ③ The personnel re-recognition method in coal mine underground based on improved metric learning performs better after discarding the images of personnel heads and feet. It has a sample matching probability of 100% when the number of similar samples is around 42.

Research on remote control technology of mining equipment based on 5G

LI Chenxin

2023, 49(9): 90-97. doi: 10.13272/j.issn.1671-251x.18089

<Abstract>(1807) <HTML> (105) <PDF>(83)

Abstract:
Mining 5G provides a high-speed information transmission channel for the construction of intelligent mines. The remote control application of mining equipment based on 5G is a key means to achieve less human and unmanned mine production. The paper analyzes the shortcomings of using 4G and WiFi6 in remote control of mining equipment, and points out that 5G technology is a necessary support method for achieving remote control of mining equipment. A reference architecture for remote control application system of mining equipment based on 5G is constructed using the research method of information physics system architecture. Taking 5G+coal mining machine remote control as an example, the key technologies of 5G transmission link are studied. The information flow between monitoring data and remote control data is sorted out. The current 5G network adopts a layer three communication method, and point-to-point layer two communication is required between the control system of remote control of mining equipment and the controlled equipment. In order to solve this problem, a layer two tunnel construction method and 5G LAN technology have been studied, and a stable channel for remote control information transmission has been established. In order to address the high bandwidth transmission requirements of on-site monitoring data and the low latency transmission requirements of remote control data, a flexible and adaptable over the air bandwidth allocation mechanism for resource pre-scheduling and request scheduling is proposed. The on-site test results show that a total of 13 328 data packets are transmitted through the layer two tunnel, without any packet loss or unsuccessful reception. The end-to-end delay is 11.5-23.8 ms, which can meet the transmission requirements of remote control of mining equipment. The RSRP(reference signal receiving power) distribution is between −93 dB·m and −53 dB·m, and the SINR(signal to interference plus noise ratio) distribution is between 10 dB and 38 dB, indicating good wireless coverage. The reliability, end-to-end delay, and wireless coverage of the mining 5G wireless communication system can meet the transmission requirements of remote control of shearers.

Research on data exchange and sharing standards for mining 5G intelligent terminal

CAI Feng, WANG Chenshulve, QIAO Liang, BAO Xiaobo, ZHANG Dongyang

2023, 49(9): 98-105. doi: 10.13272/j.issn.1671-251x.18109

<Abstract>(1310) <HTML> (53) <PDF>(41)

Abstract:
5G technology is the foundation for achieving intelligent mining construction. In response to the problem of difficult interoperability between mining 5G intelligent terminals and control platforms among different manufacturers, it is necessary to unify and standardize the data exchange and sharing process in the 5G system. The deployment architecture of "5G+intelligent mining" is proposed. The mining 5G intelligent terminals can interact and share data with servers containing control platforms through 5G base stations, switches, and 5G core networks in coal mines. This paper introduces the message types, message transmission formats, message composition, and message connection methods transmitted in the mining 5G system. For specific application scenarios in the mining field, the data in the mining 5G system is divided into voice data, video data, sensor data, and control data. The data types transmitted by the mining 5G intelligent terminal in different application scenarios are provided. This paper analyzes the basic transmission capabilities that mining 5G intelligent terminals and control platforms should have to meet data interaction. It is pointed out that the process of data interaction between mining 5G intelligent terminals and control platforms includes device information query and reporting, device upgrade, log collection, alarm issuance and processing, and platform configuration issuance and processing. It proposes the participating entities, security levels, and processes for data sharing of mining 5G intelligent terminals (including data aggregation, application, authorization, provision, and feedback). By implementing standardized processes, while ensuring communication reliability, the data exchange and sharing process under the "5G+intelligent mine" system architecture has been simplified to accelerate the intelligent construction of mines.

Design of all dielectric metasurface methane sensor based on Fano resonance

LIU Hai, ZHOU Tong, CHEN Cong, GAO Peng, DAI Yaowei, WANG Xiaolin, DUAN Senhao, GAO Zongyang

2023, 49(9): 106-114. doi: 10.13272/j.issn.1671-251x.18108

<Abstract>(975) <HTML> (69) <PDF>(29)

Abstract:
Compared with traditional methane sensors, metasurface methane sensors have advantages such as high sensitivity, stable performance, miniaturization, integration, and multi functional customizability. It better meets the application needs in complex environments such as coal mines. This paper proposes an all dielectric type metasurface methane sensor based on Fano resonance. The metasurface structure consists of periodic silicon nanostructures and SiO₂ substrates, consisting of four square silicon ring nanostructures and a central silicon nanoblock. By changing the geometric parameters, the effect on the Fano resonance of the all dielectric metasurface structure is observed. The results show the following points. Considering the quality factor and modulation depth of the structure, the center distance of the square silicon ring should be 1000 nm, the inner edge length of the square silicon ring should be 100 nm, and the edge length of the silicon nanoblock should be 200 nm. At this time, the quality factor is 227.60, and the modulation depth is 99.98%, which is close to 100%. By coating methane gas sensing thin films within the metasurface structure to achieve sensing and detection functions, combined with the extremely narrow linewidth Fano resonance features and significant local field enhancement effect, high-precision detection of methane gas is achieved. The simulation results show that the sensitivity of the all dielectric metasurface sensor to methane volume fraction is −0.953 nm/%. The change in methane volume fraction is linearly related to the shift of the resonance peak, indicating good monitoring performance. The refractive index sensitivity of the all dielectric metasurface sensor is as high as 883.95 nm/RIU. The resonance peak offset is linearly related to the environmental refractive index increment, which can be used to detect changes in environmental refractive index.

Research progress on neural network algorithms for mixed gas detection in coal mines

JIAO Mingzhi, SHEN Zhongli, ZHOU Yangming, HE Xinjian, HE Yaoyi

2023, 49(9): 115-121. doi: 10.13272/j.issn.1671-251x.18105

<Abstract>(1024) <HTML> (58) <PDF>(63)

Abstract:
When coal mine gas sensors are used for mixed gas detection, there is cross interference between measurement signals. It is difficult to ensure detection accuracy. For the same gas to be identified, the recognition precision of traditional gas recognition algorithms is lower than that of gas recognition algorithms based on neural networks. Neural networks achieve higher gas recognition accuracy by adjusting their network layers, the number of neurons in each layer, the activation function of neurons, and the weights between each layer of networks. This paper introduces the structure of a coal mine mixed gas detection system. By constructing a gas sensor array, utilizing its multi-dimensional gas response mode, and combining specific gas recognition algorithms, the qualitative and quantitative recognition of mixed gases is achieved. Several neural network algorithms for mixed gas detection in coal mines are analyzed and compared. The algorithms mainly include backpropagation (BP) neural network, convolutional neural network (CNN), recurrent neural network (RNN), and radial basis function (RBF) neural network. BP neural network can usually achieve high classification precision, but requires training a large number of parameters and a long training time. Usually, in order to reduce time and improve precision, BP neural networks can be combined with other algorithms. CNN can automatically extract data features, with better precision and training speed than BP neural networks. But it is prone to falling into local optima. RNN can use less data and extract more effective features, but it is prone to problems such as gradient vanishing. RBF neural networks have strong robustness and online learning capability, but they usually require a large amount of data to complete model training. The application of neural network algorithms will significantly improve the detection precision of mixed gases in coal mines, ensuring the implementation of intelligent coal mines.

Summary of research on artificial intelligence empowerment technology for new generation intelligent coal mine

FU Xiang, QIN Yifan, LI Haojie, NIU Penghao

2023, 49(9): 122-131, 139. doi: 10.13272/j.issn.1671-251x.18113

<Abstract>(2122) <HTML> (162) <PDF>(172)

Abstract:
The deep integration of the coal industry and artificial intelligence (AI) is an important path for modern mines to achieve intelligent personnel reduction, cost reduction, and efficiency improvement. AI empowerment in the entire process and business application scenarios of the coal industry is a specific technical measure to achieve coal mine intelligence. In the context of the current development of intelligent coal mines, a basic paradigm for the evolution of primary intelligent coal mines to new generation intelligent coal mines has been proposed. The composition, functions, and technical connotations of primary intelligent coal mines and new generation intelligent coal mines have been compared and analyzed. It is pointed out the importance of AI empowerment technology for new generation intelligent coal mine and its two key applications and implementation: the coal mine industry mechanism AI model and the coal mine Industry internet platform. The paper summarizes the current research status of industrial mechanism AI models for complex operations such as coal mine geology, mining, excavation, and safety monitoring. The paper clarifies the rapid development trend of industrial mechanism AI analysis in intelligent coal mine construction. A new generation of intelligent coal mine multi-level cloud edge collaborative industrial Internet platform architecture is designed. Using industrial information software and hardware facilities such as group data center, mine data center, production system centralized control center, and combining the features of massive data cloud computing and small amount of data edge computing, a multi-level cloud edge collaborative mechanism of group cloud, mine cloud and link edge, scene edge is proposed. It is pointed out that further research directions in the future should continue to strengthen the development and software research of AI models for coal mining industry mechanisms. Gradually a knowledge software system empowered by AI throughout the entire process of coal mining will be formed. It is suggested to fully utilize the digital resources and information facilities of the coal mining industry Internet platform to gradually realize the AI technology support of the coal mining industry Internet platform.

Research on mechanical response of artificial dam under gas explosion in roadway

QU Shijia, YANG Huan

2023, 49(9): 132-139. doi: 10.13272/j.issn.1671-251x.2023040078

<Abstract>(788) <HTML> (59) <PDF>(11)

Abstract:
When a gas explosion occurs in a mine, the explosion shock wave can damage the water storage dam, leading to a large amount of water gushing out of the goaf, and even causing gas water coupling disasters. Therefore, the stability of artificial dams under extreme conditions is of great significance for mine safety. Currently, there's a lack of research on the mechanical response features of underground artificial dams propagating with gas explosion shock waves. In order to solve the above problems, the LS-DYNA software is used to simulate the impact of gas explosion in roadways on the mechanical properties of artificial dams. The stress state, deformation, and stress features of the explosion facing side, loess interlayer, and explosion backing side are studied. The dynamic response process of artificial dams under the action of gas explosion shock waves in roadways is analyzed. The analysis results of the load distribution on the surface of the artificial dam indicate that when an explosion occurs inside the roadway, the explosion load on the explosion facing surface of the artificial dam is unevenly distributed. At the same time, in the intersection area of various underground structures, the reflected overpressure has a significant strengthening effect due to the convergence and superposition of reflected shock waves. With the rapid release of explosive energy, the impulse loading time history curve of the central measuring point on the explosion facing surface exhibits a three-stage change feature. When the gas volume is 200 m³, the maximum impulse of the central measuring point on the explosion facing surface can reach 0.04 MPa·s within 500 ms of explosion. The results of deformation and stress analysis on the surface of the artificial dam indicate that within 0-500 ms, the central part of the explosion facing surface is always under compressive stress. The maximum lateral displacement of the central node is 0.319 mm. Due to the effect of cutting, the artificial dam is subjected to tensile stress around it, where the maximum tensile and shear stresses occur. The dynamic response of the loess interlayer is in the order of "compression - compaction - plastic deformation - pressure transfer", during which the loess plays a buffering role, with a maximum displacement of 0.067 5 mm. The wall of the explosion backing side undergoes mechanical response due to the compression of the loess interlayer. But the stress is relatively small, and the outer wall is basically in a safe state.

A fault warning method for scraper conveyor chain transmission system based on LSTM-Adam

LI Bo, GUO Xingran, LI Juanli, WANG Xuewen, XIA Rui

2023, 49(9): 140-146. doi: 10.13272/j.issn.1671-251x.18086

<Abstract>(850) <HTML> (79) <PDF>(42)

Abstract:
The scraper conveyor chain transmission system is prone to frequent faults due to its complex load bearing capacity. However, traditional fault diagnosis requires a large amount of prior knowledge and subjective intervention, which requires high technical personnel. In order to achieve the autonomy, accuracy, and efficiency of fault warning for the scraper conveyor chain transmission system, a fault warning method for the scraper conveyor chain transmission system based on LSTM-Adam is proposed using the powerful data mining capability of deep learning. Firstly, a monitoring system for the working conditions of the scraper conveyor is built based on configuration technology. The system collects real-time operating data of the scraper conveyor, such as the torque and speed of the output shaft of the reducer, the pressure of the middle groove plate, the vibration acceleration in the vertical direction of the scraper, and the strain in the running direction of the scraper chain. The data is cleaned and normalized in min-max to provide data support for fault warning. Secondly, a prediction model is built based on LSTM and trained and optimized using the Adam optimization algorithm to obtain the optimal LSTM Adam prediction model. Finally, the real-time operating data of the scraper conveyor is imported into the LSTM-Adam prediction model to obtain the predicted values of the scraper conveyor operating parameters. The sliding weighted average method is used to calculate the residual between the predicted value and the true value. The maximum residual of the same type of data under normal operating conditions is used as the warning threshold. When the residual exceeds the warning threshold, an early warning is given. The experimental results show that the LSTM-Adam prediction model can accurately predict the trend of strain data of the scraper chain and provide accurate warnings for stuck chain and broken chain faults.

Research progress and development path of intelligent mining comprehensive control platform

XING Zhen

2023, 49(9): 147-154. doi: 10.13272/j.issn.1671-251x.2023070013

<Abstract>(1185) <HTML> (90) <PDF>(94)

Abstract:
The intelligent mine comprehensive control platform is an important support for achieving safe and efficient production in coal mines. However, there are still problems in the research and practice process, such as inconsistent conceptual descriptions of the control platform, and deviation from the core development direction in the development and construction of the control platform functions. In order to clarify the development direction and path of the later functional development and on-site application process of the control platform, the essence of the control platform is analyzed. It is concluded that the essence of the control platform centered on data flow involves three aspects: data access, data fusion, and data comprehensive application. By comparing and integrating the existing intelligent mine acceptance management measures of the National Energy Administration, major coal producing provinces, and major energy central enterprises, the functional requirements of the control platform are subdivided into basic support functions, expanded supplementary functions, and key core functions. Taking the three major functions as the main line, this paper summarizes the research and application progress of control platforms and proposes a development path. The basic support functions have problems such as insufficient data access stability, maturity, and practicality. The functions should be developed towards standardization and reliability. Expanding supplementary functions can meet the personalized needs of enterprises, but it is not the core development direction of intelligent comprehensive control. It should be developed towards practicality and complementarity to assist in achieving key core functions such as global interconnection, collaborative control, and independent decision-making. The key core functions should be developed towards standardization and stability. Currently, research on key core functions lacks specific standards and details. In order to achieve the goal of intelligent mine construction, it is necessary to solve the practical difficulties and problems in coal mine sites, and gradually explore new application scenarios and potential needs.

Prediction model of slope deformation in open pit mines based on GJO-MLP

LIU Guangwei, GUO Zhiqing, LIU Wei

2023, 49(9): 155-166. doi: 10.13272/j.issn.1671-251x.2023070017

<Abstract>(924) <HTML> (57) <PDF>(31)

Abstract:
The deformation of open-pit mine slopes is influenced by various factors such as geological structure, hydrogeological conditions, mining activities, etc., making the prediction model complex. It is difficult to accurately capture all influencing factors. At present, a large number of monitoring devices are deployed around the slope of open-pit mines to record real-time displacement data of open-pit mine slopes. The data has the features of high-dimensional, temporal correlation, and nonlinear. Traditional slope stability analysis methods cannot effectively predict slope deformation without knowing other conditions and only data, it is necessary to use a data-based model to predict the displacement data of open-pit mine slopes in advance for slope stability analysis. In order to solve the above problems, a deformation prediction model for open-pit mine slopes based on the golden jackal optimized multilayer perception machine (GJO-MLP) is proposed. Each agent in GJO is independent of each other and can accelerate the training process of optimizing MLP through parallel computing. GJO can combine the nonlinear modeling and feature extraction capabilities of MLP, making the optimized MLP more advantageous in dealing with complex problems. To test the feasibility and effectiveness of GJO-MLP, GJO-MLP is compared and analyzed with ant colony algorithm optimization based MLP (ACO-MLP), gravity search algorithm optimization based MLP (GSA-MLP), and differential evolution algorithm optimization based MLP (DE-MLP). The simulation results on six datasets show that under the same experimental conditions, GJO-MLP shows better optimization performance compared to the other three algorithms. The slope deformation prediction model based on GJO-MLP is applied to the slope deformation prediction of Baorixile open-pit mine and Huapingzi slope deformation prediction. The results show that under the same conditions, compared to the other three algorithms, the slope deformation prediction model based on GJO-MLP not only show better predictive performance in predicting slope deformation data, but also has better feasibility and robustness.

Research progress on insulation aging mechanism and condition evaluation technology of mining EPDM high-voltage cables

LEI Zhipeng, JIANG Wanting, MEN Rujia, ZHANG Jianhua, LI Yuanyuan, HE Qinghui, LI Wei

2023, 49(9): 167-177. doi: 10.13272/j.issn.1671-251x.18150

<Abstract>(1026) <HTML> (55) <PDF>(31)

Abstract:
Insulation is considered the weakest link in electrical equipment. The combined effects of special working conditions in coal mines and aging factors such as electrical, thermal, and mechanical stresses make it difficult to determine the aging mechanism and evaluate the condition of EPDM insulation for high-voltage cables used in mines. This paper introduces the basic performance and aging factor types of EPDM insulation for high-voltage mobile flexible cables used in coal mines. Based on the physical, chemical, mechanical, and electrical properties of EPDM under the influence of multiple aging factors, the aging mechanism of EPDM is proposed. This paper summarizes the basic principles and existing problems of online monitoring methods for insulation resistance, partial discharge, dielectric loss factor, and temperature of mining high-voltage cables. The paper summarizes the current research status of insulation status evaluation methods for mining high-voltage cables. The paper introduces the evaluation methods for insulation status of multi parameter based on improved radar map and single parameter based on dielectric loss mining high-voltage cables. To cope with the development of coal mine intelligence, on the one hand, it is suggested to do research on intelligent perception and control of mining electrical equipment to compensate for the lack of state perception and state evaluation feature quantities. On the other hand, it is necessary to study lightweight models or algorithms to reduce the computational complexity, parameter quantity, and analysis time of intelligent terminals near devices. It improves the feasibility of state evaluation technology, and lays the foundation for achieving intelligent analysis and decision-making.

Research on ultra wideband radar life detection technology

ZHAO Youxin, YAO Haifei, LI Jiahui, PENG Ran, LI Xun

2023, 49(9): 178-186. doi: 10.13272/j.issn.1671-251x.18111

<Abstract>(1495) <HTML> (89) <PDF>(60)

Abstract:
Ultra wideband (UWB) radar life detection technology has the advantages of low power consumption, good penetration, and high confidentiality. It is beneficial for improving the survival rate of trapped personnel after disasters. This paper systematically summarizes the research progress and current status of UWB radar life detection technology both domestically and internationally. According to the different forms of transmitted signals, UWB radar life detection technology is divided into continuous wave radar life detection technology and pulse wave radar life detection technology. The principles and application advantages of the two detection technologies are introduced respectively. Based on the respective features of continuous wave radar life detection technology and pulse wave radar life detection technology, this paper analyzes the key technologies of UWB radar life detection from three perspectives: detection signal transmission, echo signal preprocessing, and life signal extraction and analysis, and summarizes the research status of the three key technologies. The paper proposes prospects for the research on UWB radar life detection technology. The technology breaks through the hardware performance of life detector transceivers, improves the transmission signal bandwidth, and optimizes RF power amplification technology to increase the detection distance through walls. The technology comprehensively utilizes multiple feature extraction methods and intelligent pattern classification methods, as well as new generation information technologies such as artificial intelligence, big data, and cloud computing, to improve the precision of target recognition. The technology develops a human target recognition and positioning equipment based on multi input multi output radar and a high-precision distributed networked fully polarized UWB radar life detector to enhance the dimension of detection results.

2023 Vol. 49, No. 9

Submission System

Latest AcceptedMore>