2021 Vol. 47, No. 8

  Previous Issue | Next Issue 
Display Method:
Cover Catalogue
Coal mine intelligence，mine 5G and network hard slicing technology
SUN Jiping
2021, 47(8): 1-6. doi: 10.13272/j.issn.1671-251x.17821
<Abstract>(462) <HTML> (82) <PDF>(86)
Abstract:
In order to reduce the number of operators in the working face of coal mines, an unmanned 5G ground remote control method for working face is proposed. The industrial cameras and sensors are set up in the working face, which transmit video, audio and sensor signals to the ground through the 5G network. The ground operator operates the equipment remotely, which transmits control commands to the working face through the 5G network, and controls the actions of the mine equipment. The wireless transmission distance and transmission bandwidth calculation method are proposed for unmanned ground remote control of the fully mechanized working face. The wireless transmission distance between the base stations at the two ends of the fully mechanized working face should not be less than 1/2 of the length of the fully mechanized working face. The total uplink transmission bandwidth required for unmanned ground remote control of the fully mechanized working face is proportional to the length of the fully mechanized working face, inversely proportional to the center distance of the hydraulic support, inversely proportional to the number of supports and cameras, and proportional to the transmission bandwidth required for the compressed video of a single camera. In order to reduce the demand for uplink transmission bandwidth for unmanned ground remote control of fully mechanized working face, it is possible to only transmit the camera videos of the adjacent shearers. The minimum bandwidth of uplink transmission required for unmanned ground remote control the fully mechanized working face is proportional to the number of cameras of the adjacent shearers, and proportional to the required transmission bandwidth of a single camera after video compression. The wireless transmission distance test method for the unmanned ground remote control system of the fully mechanized working face is proposed. The method is applied for the wireless transmission system such as 5G for the unmanned ground remote control system of the fully mechanized working face. Under the premise of ensuring transmission bandwidth, time delay and reliability, the wireless transmission distance is tested. The test shows that the uplink wireless transmission bandwidth of the base station shall not be less than 20 Mbit/s, and the wireless transmission distance shall not be less than 150 m. This study proposes the all-mine integrated information transmission network based on network hard slicing. Through network hard slicing, different channels are allocated for coal mine safety monitoring, mine monitoring, personnel and vehicle and equipment positioning, video monitoring, voice communication, 5G communication, etc. This method guarantees the high reliability and low time delay requirements of coal mine safety monitoring and mine monitoring, and unifies the underground information transmission network of coal mines. Moreover, it integrates multiple networks such as coal mine safety monitoring network, mine industrial Ethernet and mine 5G communication network.
Construction conception of intelligent mine integrated management and control platform
HAN An, CHEN Xiaojing, HE Yaoyi, GAO Wen
2021, 47(8): 7-14. doi: 10.13272/j.issn.1671-251x.17814
<Abstract>(243) <HTML> (16) <PDF>(60)
Abstract:
In coal industry, there are problems of serious information island, insufficient business interconnection and interoperability, and poor data value mining and utilization under the existing single system and single business application mode. And there are problems of single application point and insufficient intelligent empowerment of new generation information technology such as Internet of Things, cloud computing, big data, artificial intelligence, intelligent control, etc. In order to solve the above problems, the new generation information technology is deeply integrated with coal mine safety, production and operation management and control business. The construction concept of intelligent mine integrated management and control platform is proposed. The platform is based on the basic information platform of intelligent mine, which aggregates, manages, stores and analyzes massive heterogeneous data through the coal industry big data center. The data platform, technology platform and application platform is applied to realize data fusion, collaborative control, business linkage and decision analysis in the same platform. By adopting unified portal, unified data and unified style, a centralized and unified intelligent control and business application center is built for production scheduling collaborative management and control, comprehensive risk prevention and control management, comprehensive management and control for decision analysis, and precise operation and maintenance detection. The production scheduling collaborative management and control application center applies coal mine safety production collaborative technology to realize business interconnection and hierarchical scheduling among mine leaders, scheduling command centers, departments, district teams and work groups, and integrated scheduling and collaborative control in fixed sites. The comprehensive risk prevention and control management application center takes coal mine safety risk prevention and control as the core, and realizes all-round perception, real-time monitoring, dynamic evaluation, and abnormal linkage treatment of risk and hazard elements in the whole mine to ensure efficient and orderly operation of coal mine safety production. The comprehensive management and control for decision analysis application center applies time-series data mining and analysis technology to realize the analysis and prediction of indicators such as safety, production and operation. Through the safety production big data board and personal workbench, the comprehensive display of safety, production and operation conditions is carried out. The precise operation and maintenance detection application center uses running probes to realizes the fault monitoring and abnormality tracking and processing during the operation of the platform.
Comprehensive prevention and control system of coal mine safety based on risk management and control
CHEN Xiaolin, QU Shijia, SHE Jiuhua, ZHANG Yu
2021, 47(8): 15-19. doi: 10.13272/j.issn.1671-251x.17809
<Abstract>(85) <HTML> (12) <PDF>(13)
Abstract:
In order to solve the problems of low real-time, insufficient coverage and low reusability of the coal mine safety control system with hidden danger management as the core, the construction of a comprehensive prevention and control system of coal mine safety based on risk management and control is proposed. The overall management and control of coal mine risks is carried out through risk identification and implementation of risk control measures, dynamic risk monitoring and closed-loop management of hidden dangers. Based on the results of major disaster risk identification, the system integrates traditional single safety monitoring systems and intelligent ventilation, drainage and power supply systems to achieve comprehensive monitoring, data analysis, early warning and abnormal linkage treatment of major disaster risks. Taking the coal mine enterprise risk database as the core, the system forms and fits safety risk index according to the effects of different levels of safety risk control so as to realize the evaluation and the early warning of abnormalities of comprehensive prevention and control system of coal mine safety. The comprehensive prevention and control system of coal mine safety based on risk management and control helps to realize the pre-control of risks in personnel operation behavior, equipment safety status, environmental safety situation and management safety so as to achieve the goal of preventing and controlling major risks and containing major accidents.
Research status and direction of ultra-wide band radar technology for borehole rescue
ZHENG Xuezhao , SUN Ziyu, ZHANG Yanni, ZHANG Duo, XU Chengyu
2021, 47(8): 20-26. doi: 10.13272/j.issn.1671-251x.17800
<Abstract>(394) <HTML> (59) <PDF>(30)
Abstract:
In the mine rescue process, the current video and audio based borehole life information identification technology has the problems of video obstacle failure and fast speed of sound wave attenuation. In order to solve the above problems, the ultra-wide band (UWB) radar technology that can penetrate coal rock bodies, brick walls and other obstacles for non-contact measurement and realize life information identification is studied. The application status and existing problems of UWB radar technology in mine borehole rescue are analyzed from three aspects, including UWB electromagnetic wave transmission attenuation characteristics, clutter filtering and optimization, and the target life identification method. ① UWB electromagnetic waves can detect and locate the life information of people behind obstacles. Its attenuation is related to the electric field, magnetic field strength, frequency and medium. However, the current research on the attenuation characteristics of UWB electromagnetic wave transmission lacks the research on the law relationship between various coal and rock masses and UWB electromagnetic wave transmission attenuation at different temperatures, detection frequencies, and metamorphic degrees. There are few researches on the relationship between UWB electromagnetic wave frequency, incident angle, polarization form and other key parameters and change characteristics and UWB electromagnetic wave attenuation. ② The methods such as empirical mode decomposition method and MUSIC method can filter the clutter in UWB electromagnetic waves. But the methods lack accurate extraction and scientific representation of effective wave characteristics and information after clutter filtering, which makes it difficult to improve the speed and accuracy of life information characteristics extraction and identification. ③ UWB electromagnetic waves can capture the life signs of multiple target people behind obstacles, such as breathing or heartbeat. The use of multiple radar observation points can improve the accuracy of life information detection in a low signal-to-noise ratio environment. However, the current research lacks systematic identification and quantitative analysis of breathing, heartbeat, chest undulation, body temperature in multiple directions and from multiple angles. And the current research lacks an optimized method for life information identification models. In order to solve the existing problems, the research directions of UWB radar technology for borehole rescue application are pointed out as follows. ① It is proposed to study the influence of the properties of coal and rock masses and electromagnetic wave related parameters on the transmission attenuation of UWB electromagnetic wave in coal and rock masses under different conditions. The law of transmission attenuation is obtained. ② It is suggested to study on clutter filtering optimization and echo characteristic extraction algorithms suitable for mine environment, and use the effective characteristics in the echo to establish a learning sample database. ③ It is proposed to construct the personnel life information identification model applicable to the mine environment, and continuously optimize the model with the sample database to improve the accuracy and speed of life information identification.
Research on the overall framework and key technologies of intelligent open-pit mines
FU Ensan, LIU Guangwei, ZHAO Hao, QU Yeming, DI Shuai, JIANG Lin
2021, 47(8): 27-32. doi: 10.13272/j.issn.1671-251x.2021010058
<Abstract>(248) <HTML> (14) <PDF>(33)
Abstract:
Combined with the characteristics of open-pit mines in China, the definition of intelligent open-pit mines is given. This paper introduces the overall network architecture, business architecture and collaborative process of the intelligent open-pit mine. The overall network architecture of cloud, edge and end is adopted to realize the collaborative correlation of various production links in the open-pit mine by accessing multi-source heterogeneous data and fully applying advanced technologies such as big data and cloud computing. The business architecture adopts the "5 layers + 3 system" model. The "5 layers" from bottom to top are the basic equipment layer, business data layer, analysis service layer, analysis business layer and analysis presentation layer. And the "3 systems" are data analysis standard system, safety specification standard system, metadata and code specification standard system. The bottom-up collaboration of intelligent open-pit mine should realize "basic + data" support, mine production system collaboration, disaster risk analysis collaboration and business decision-making collaboration. The key technologies such as unmanned technology, digital twin technology, big data collection and analysis technology, risk monitoring and early warning technology of intelligent open-pit mine are discussed. It is pointed out that technologies such as big data and artificial intelligence will become the core driver of open-pit mine intelligence. The platform-based architecture will become the common choice for future intelligent open-pit mine systems.
Study on long drilling long-distance fixed-point gas content measurement
ZHANG Jianguo, LI Xiyuan, GAO Jiancheng, YU Hong
2021, 47(8): 33-40. doi: 10.13272/j.issn.1671-251x.17799
<Abstract>(57) <HTML> (9) <PDF>(6)
Abstract:
The sampling time of long drilling fixed-point gas content measurement is long. And it is unable to measure the gas content in coal seams accurately due to the absence of desorption on site. In order to solve the above problems, a method of using analogy to estimate gas loss and atmospheric gas desorption during sampling is proposed. In the same geological conditions and gas geological unit, the ratio of gas loss and on-site gas desorption to the total gas content in the measurement results of the gas content of the shallow drilling air exhaust powder sampling of the common rotary drilling rig is used as the standard. The analogy is the ratio of gas loss and on-site gas desorption to the total gas content in the deep drilling fixed-point sampling of the kilometer directional drilling rig. The ratio of gas loss and on-site gas desorption is added to the final measured residual gas content. The total gas content of the deep drilling fixed-point sampling of the kilometer directional drilling rig is obtained. This method is used to measure the gas content of long drilling in the Ji16-17-24070 working face of No.11 Pingdingshan Coal Mine, and to correct the gas content measured values of the kilometer directional drilling sampling. The test results show that compared with the gas content measurement of the short drilling direct method, the average error of the long drilling gas content measurement corrected by the analogy method is reduced from 12.07% to 0.19%. Compared with the content prediction values, the average error of the long drilling gas content measured values corrected by analogy method is reduced from 12.90% to 4.71%.
Analysis and prevention of influencing factors of face end roof falling under repeated mining of short-distance coal seam group
LI Qiang, WU Guiyi, KONG Dezhong
2021, 47(8): 41-49. doi: 10.13272/j.issn.1671-251x.2021020001
<Abstract>(59) <HTML> (8) <PDF>(6)
Abstract:
The existing researches on face end roof falling are mostly focused on single coal seam, while the stability of the face end roof under short-distance coal seam group mining is different from that of a single coal seam. In the process of short-distance coal seam group mining, due to the small spacing between adjacent coal seams, the roof structure and stress environment of the lower coal seam mining area will change after the upper coal seam is mined. This easily lead to catastrophic problems such as face end roof falling, coal wall spalling and roof pressure frame. Taking the 17101 working face of a mine that has experienced 3 large-scale face end roof falling as the engineering background, based on the stability relationship between the coal wall, hydraulic support and the face end roof, the ‘face end roof-coal wall-support’ model is established under repeated mining of short-distance seam group. This paper introduces the main influencing factors of face end roof falling under repeated mining of short-distance seam group are the strength of the roof and surrounding rock, support working resistance, advancing speed and face end distance. The UDEC simulation software is used to simulate the influence of different influencing factors on the face end roof falling under repeated mining, and conclusions are summarized as follows. ① The face end roof and coal body are damaged during repeated mining, the strength of the face end roof and coal wall is reduced, and the face end roof at the front end of the support is prone to fall, causing disasters such as roof falling and coal wall spalling. As the strength of the roof and surrounding rock increases, the stability of the roof is better. ② The low working resistance of the hydraulic support is one of the important reasons for the face end roof falling of the working face. As the working resistance of hydraulic support increases, the stability of the face end roof is better. ③ The advancing speed of the working face has a significant effect on the stability of the face end roof. The slower the advancing speed of the working face is, the more serious is the roof subsidence phenomenon of the at the stope. ④ The tip-to-face distance is an important influencing factor of the face end roof falling. The face end roof falling is linearly related to the tip-to-face distance. The falling height increases with the increase of the tip-to-face distance. The smaller the tip-to-face distance is, the more stable is the face end roof. However, too short tip-to-face distance will also affect the normal mining of the working face. The tip-to-face distance should be determined according to the actual situation. In the context of the four main influencing factors, the prevention and control measures for face end roof falling under repeated mining are proposed. The measures include increasing the strength of the roof and surrounding rock, improving the working resistance of hydraulic support, controlling the advancing speed reasonably and reducing the tip-to-face distance. The above prevention and control measures can provide an effective solution for the face end roof falling under repeated mining of short-distance seam group.
Research on the stability control of surrounding rock in the roadway with dynamic pressure and high slope
SHEN Binxue, YUAN Chaofeng, GU Wenzhe, LIU Zhicheng, SONG Tianqi, PAN Hao
2021, 47(8): 50-55. doi: 10.13272/j.issn.1671-251x.2021030023
<Abstract>(54) <HTML> (10) <PDF>(7)
Abstract:
The roadway with dynamic pressure and high slope refers to the roadway which is seriously affected by dynamic pressure and has a high slope. This type of roadway not only experiences disturbances in the excavation process, but also has to experience the disturbance of the adjacent working face and this working face in the later mining process. This makes the deformation and destruction process of the surrounding rock of this kind of roadway different from that of the conventional mining roadway, and the stability of the surrounding rock with high slope is relatively weak. Taking the air intake roadway of a mine's roadway with dynamic pressure and high slope of 15312 working face as the research object, the deformation and damage mechanism of the roadway with dynamic pressure and high slope is studied by using a combination method of numerical simulation and on-site monitoring. And the support parameters of the roadway with dynamic pressure and high slope are optimized accordingly. By analyzing the damage characteristics of the air intake roadway of 15312 working face, it is concluded that the factors affecting the stability of the surrounding rock of the air intake roadway mainly include the strength of the surrounding rock, the size of the roadway section, the stress environment of the surrounding rock and the strength of the surrounding rock support. It is difficult to control the stability of the roadway from the perspective of the low strength of the surrounding rock itself, the large size of the roadway section and the complex stress environment in which the surrounding rock is located. It is proposed to improve the overall stability of the roadway from the perspective of optimizing the roadway support parameters, which is taking measures to increase the support strength and support range of the two sides of the roadway. Numerical simulation results show that the surrounding rock support stress field after the optimization of the support parameters is approximately circular, and the surrounding rock support stress field of the side wall is larger. The result is more suitable for the stability control of the surrounding rock of of the air intake roadway of the 15312 working face. The monitoring results of on-site anchor rods and anchor cables show that after the optimization of the support parameters, the stability control of the surrounding rock of the roadway with dynamic pressure and high slope can be achieved, the deformation of the surrounding rock is smaller, and the control effect is better.
.Research on the coal pillar width of gob-side entry retaining under the damage effect of heterogeneous rock
XIE Zhenhua, MA Tianhu, FAN Zhanglei, FAN Chaojun, LIU Husheng, HU Jiang, LI Yunfei
2021, 47(8): 56-62. doi: 10.13272/j.issn.1671-251x.2020120037
<Abstract>(80) <HTML> (8) <PDF>(7)
Abstract:
At present, the research on the coal pillar width of gob-side entry retaining focuses on the coal pillar strength, coal pillar load and coal pillar stability. There are few researches on the impact of rock heterogeneity and damage effects on the coal pillar width. Taking the gob-side entry retaining of No.12 coal in the fourth panel of Wulanmulun Coal Mine as the engineering background, the coal pillar width is studied in the context of rock heterogeneity-damage effect and the comprehensive use of theoretical analysis, numerical calculation and field measurement. The research results show that the rock heterogeneity-damage effect model can better reflect the fracture characteristics of the rock. Only a small number of particles are damaged in the elastic stage. Fractures begin to develop and penetrate in the plastic stage. Macro shear fractures along the diagonal direction are formed in the damage stage. As the coal pillar width increases, the overall deformation of the roadway first decreases and then increases. When the coal pillar width is 6 m, a sudden change occurs. The range and degree of damage around the coal pillar continue to decrease. The deformation of the roadway gang on the mining side is larger than that on the non-mining side. However, the amount of change with the coal pillar width is smaller than that of the non-mining side, and the surrounding rock damage distribution on the goaf side is larger than that of the solid coal side. According to the theoretical analysis and numerical calculation, it is proposed that the coal pillar width of gob-side entry retaining is 5 m. The result is applied to the engineering site. The overall deformation of the roadway in front of the working face is not large, and the movement and deformation of the overburden rock after 60 m behind the working face are basically stable, which verify the reasonableness of the coal pillar width of gob-side entry retaining.
PWM rectifier LCL filter design for underground mine
LI Shan, LIU Xiaodong, SU Changbo
2021, 47(8): 63-68. doi: 10.13272/j.issn.1671-251x.17683
<Abstract>(125) <HTML> (20) <PDF>(14)
Abstract:
The dual PWM variable frequency speed regulation system applied to scraper conveyors, coal shearers and other mechanical equipments needs to be installed a filter on the rectifier grid-side to suppress harmonic currents. The existing LCL filter design method relies on experience and has low efficiency, and the damping loss problem is not considered enough. It is easy to cause device heating and the design is not suitable for underground applications. By analyzing the LCL filter model of the rectifier of the double PWM variable frequency speed regulation system, the mathematical relationship among the inductance ratio, damping resistance and current attenuation coefficient, resonance frequency, and damping loss is derived, and then a PWM rectifier LCL filter design method suitable for underground coal mines is proposed. In order to reduce the volume of the rectifier, a smaller total inductance should be selected on the basis of considering the filtering effect, the current fast tracking ability, and the fast response ability of the system. On the premise of ensuring the filtering effect, it is suggested to increase the filter capacitance. Within the allowable range of total harmonic distortion rate of the current, it is proposed to select the minimum value of inductance ratio and damping resistance. A three-phase voltage PWM rectifier control model is built in Matlab/Simulink for simulation and verification. The results show that the LCL filter designed by this method can make the PWM rectifier have a small overshoot, fast response and low total harmonic distortion rate. The prototype experiment verifies that the LCL filter can better suppress the grid-side current of the PWM rectifier and improve the quality of the grid-connected current waveform. It is verified by comparison that the filter inductance and damping resistance designed by this method are significantly smaller than those designed by the traditional method, which can reduce the size of the filter and reduce the cost effectively.
Particle size change of solid residue and flammability analysis of gaseous residue of lignite coal dust explosio
ZHU Chao, WANG Hao
2021, 47(8): 69-76. doi: 10.13272/j.issn.1671-251x.2021040040
<Abstract>(135) <HTML> (19) <PDF>(5)
Abstract:
The solid and gaseous residues of coal dust explosion are the physical evidence for analyzing the detonation site and propagation path and the basis for deploying rescue work. The 20 L spherical explosion system is used to conduct explosion experiments on lignite coal dust with different concentrations and particle sizes. And the particle size change of the solid residue and the flammability of the gaseous residue of the lignite coal dust explosion are analyzed. The results show that the particle size range of lignite coal dust explosion solid residues is larger than that of the original lignite coal dust. When the coal dust concentration is constant, as the particle size decreases, the D10 change rate of lignite coal dust explosion solid residues tends to increase negatively, the D50 change rate fluctuates in the range of 10% to 20%, and the D90 change rate tends to increase positively. When the coal dust particle size is constant, as the coal dust concentration increases, the D10 of the lignite coal dust explosion solid residue first decreases and then increases, while the D50 and D90 continue to increase. The lignite coal dust explosion gaseous residue mainly contains O2, CO, H2,CO2,CH4,C2H6,C2H4,C2H2 and C3H6. As the particle size of coal dust decreases or the concentration increases, the content of O2 and CO2 in the gaseous residue decreases continuously, and the content of hydrocarbon gases such as CO, H2 and CH4 increases continuously. When the coal dust particle size is constant, the flammability index of lignite coal dust explosion gaseous residue within the concentration range of 100-200 g/m3 is less than 1, indicating that the gaseous residue is not combustible. The flammability index of lignite coal dust explosion gaseous residue within the concentration range of 300-600 g/m3 is greater than 1, indicating that the gaseous residue is combustible. When the coal dust concentration is constant, the flammability index of lignite coal dust explosion gaseous residue with the particle size of 180-250 μm is less than 1, indicating that the gaseous residue is not combustible. The flammability index of lignite coal dust explosion gaseous residue with the particle size of 38-180 μm is greater than 1, indicating that the gaseous residue is combustible. The content of H2 and CO in the gaseous residue is the key factor affecting the flammability of lignite coal dust explosion gaseous residue. The flammability is mostly affected by H2, followed by CO.
Coal mine belt conveyor foreign object detectio
DU Jingyi, CHEN Rui, HAO Le, SHI Zhimang
2021, 47(8): 77-83. doi: 10.13272/j.issn.1671-251x.2021040026
<Abstract>(142) <HTML> (15) <PDF>(26)
Abstract:
In order to solve the problem of slow detection speed of existing deep learning based belt conveyor foreign object detection methods, an improved YOLOv3 model is proposed and applied to coal mine belt conveyor foreign object detection. The model uses the lightweight network DarkNet22-DS as the backbone feature extraction network. DarkNet22-DS replaces the standard convolution with depthwise separable convolution, which reduces the network parameters significantly and improves the feature utilization efficiency by composite residual blocks. By introducing weighted bi-directional feature pyramid networks and dual-scale output, the model improves the feature fusion network and enhances the model's detection efficiency of large foreign objects. The complete intersection ratio loss function is used as the target box regression loss function, and the correlation between the target box information is fully utilized to improve the convergence speed and detection accuracy of the model. The improved YOLOv3 model is deployed on the embedded platform Jetson Xavier NX for coal mine belt conveyor foreign object detection experiments. The results show that compared with the YOLOv3 model, the weight file size of the improved YOLOv3 model is reduced by 91.4%, and the amount of model parameters is reduced significantly. The detection speed is increased by 16 times, reaching 30.7 frames/s. The performance meets the real-time detection requirements of foreign objects in coal mine belt conveyors.
Fault diagnosis of rolling bearings based on GAF and DenseNet
JIANG Jiaguo, GUO Manli, YANG Siguo
2021, 47(8): 84-89. doi: 10.13272/j.issn.1671-251x.2021040095
<Abstract>(147) <HTML> (11) <PDF>(14)
Abstract:
The model-based and signal-based rolling bearing fault diagnosis methods have problems such as difficult modeling and cumbersome signal analysis. The data-driven rolling bearing fault diagnosis methods mostly use convolutional neural networks, but as the number of network layers increases during network training, gradient disappearance occurs. Moreover, taking the vibration signal of the rolling bearing directly as the network input will cause incomplete feature extraction. In order to solve the above problems, a rolling bearing fault diagnosis method based on Gramian angular field(GAF) and densely connected convolutional network(DenseNet) is proposed. The one-dimensional time series of rolling bearing vibration signals are converted into two-dimensional images by GAF, which preserves the correlation information between the time series data. The two-dimensional images are used as the input of the densely connected convolutional network, and the feature extraction of the two-dimensional images is carried out by the DenseNet, which improves the feature information utilization and realizes the fault classification. Experiments are carried out by using the data from the Case Western Reserve University bearing dataset. The results show that the method can identify rolling bearing fault types effectively with a fault diagnosis accuracy rate of 99.75%. In order to further prove the superiority of this method, the fault diagnosis methods of gray-scale image+DenseNet, GAF+residual network(ResNet), gray-scale image + ResNet are selected for comparison. The results show that the GAF+DenseNet method has the highest accuracy rate, and the gray-scale image+ResNet method has the lowest accuracy rate. Compared with the gray-scale image, the GAF converted two-dimensional image retains the relevant information between the original time series data. Compared with ResNet, DenseNet is able to extract the fault features more adequately due to denser connection method.
Research on fault prediction of working face equipment based on time series data
ZHENG Lei
2021, 47(8): 90-95. doi: 10.13272/j.issn.1671-251x.17694
<Abstract>(75) <HTML> (8) <PDF>(11)
Abstract:
Coal mine working face equipment are usually consists of several complex system modules that have strong coupling among each other. Moreover, the equipment fault mechanism is complex. Therefore, when the equipment fault prediction is carried out, it is necessary to conduct real-time monitoring of equipment operation status, environmental data and operation data so as to obtain time series data of electrical, mechanical, thermal and other parameters. A method for fault prediction of working face equipment based on time series data is proposed. Firstly, the time series alignment algorithm is used to align the collected equipment monitoring data. The time columns of monitoring data are reordered, and the time columns are the key values. Each monitoring data is filled in as the label value, and the previous value is filled in the vacant value. Secondly, the fault-related factors are selected according to the fault characterization phenomenon and the occurrence mechanism. And the correlation between the relevant factors is calculated by Pearson correlation coefficient analysis method, thereby determining the fault prediction factor set. Finally, the long short-term memory(LSTM) network is used to establish a fault prediction model for working face equipment. The normalized set of fault prediction factor set is used as the input and the fault is used as the output of the LSTM prediction model. The delay time period is introduced into the LSTM prediction model to realize advanced prediction of delay faults. The test is carried out by taking the shearer overheating trip fault as an example. Through analysis, it is found that the fault prediction factor set is {drum temperature, drum current, drum start and stop, traction temperature, transformer temperature, rocker arm temperature}. When the number of LSTM network cell layers is 10, the number of hidden cells is 10, the learning rate is 0.001, the number of iterations is 1 500, and the number of samples read per time is 120, the delay time of shearer overheating trip fault is 30 min. When the test set is used for fault prediction, the advanced prediction time is 26 min , which is 4 min shorter than the delay time, indicating that the LSTM network can effectively achieve advanced fault prediction of working face equipment based on time series data.
A parameter offline identification method of asynchronous motor of mine inverter
JIANG Dezhi, RONG Xiang, CHEN Wenya, WANG Yue, LIAN Chao
2021, 47(8): 96-101. doi: 10.13272/j.issn.1671-251x.17815
<Abstract>(125) <HTML> (13) <PDF>(8)
Abstract:
Under the condition of no speed sensor, the speed regulation performance of inverter is closely related to the parameters of asynchronous motor. Most of the existing offline identification methods of asynchronous motor parameters use fast Fourier transform to calculate the current amplitude and phase. There are spectrum leakage and fence effect and it is likely to cause measurement errors. In order to solve the above problems, a parameter offline identification method of asynchronous motor of mine inverter with the asynchronous motor equivalent circuit as the load is proposed. The steady-state response method is used to identify the stator resistance, and the transient response method is used to identify the rotor resistance and leakage inductance. On the basis of this method, the motor model in synchronous rotating coordinate system is established. The current instantaneous values are transformed in rotating coordinate to calculate the current amplitude and phase under the no-load operation condition of constant voltage frequency ratio. Then, the stator-rotor mutual inductance and no-load excitation current are identified, avoiding the spectrum leakage and fence effect. Matlab/Simulink simulation results show that the rotor resistance and leakage inductance parameter identification results are consistent with the given motor model parameters, which verifies the accuracy of the parameter identification method. 380 V/2.2 kW motor parameter identification experimental results show that the voltage signal applied to the motor by the inverter gets the correct current response, which is consistent with the simulation results. Moreover, and the motor parameter identification process has good repeatability and high identification accuracy. The 660 V/90 kW mine inverter core is built, and the identified motor parameters are used in the mine inverter vector control system. The actual test results show that the inverter can start with load and continue to run stably under rated torque. The motor parameter identification results can meet the vector control requirements of the mine inverter.
Coal mine gas emission prediction method based on random forest regressio
WU Fengliang, HUO Yuan, GAO Jianan
2021, 47(8): 102-107. doi: 10.13272/j.issn.1671-251x.2021010024
<Abstract>(82) <HTML> (8) <PDF>(11)
Abstract:
In order to improve the prediction accuracy and efficiency of coal mine gas emission, a coal mine gas emission prediction method based on random forest regression is proposed. The bootstrap self-service resampling technology is used to collect training sample data and construct a random forest regression model. The mean value of the decision tree output value is taken as the prediction result of coal mine gas emission and the out-of-bag data is used to evaluate the prediction performance of the regression model. The optimal hyperparameters of the random forest regression model are determined by calculating the mean of squared residuals and goodness of fit of the out-of-bag data. The increase in the mean of squared residuals of the out-of-bag data is used to characterize the importance of the characteristic variables. All the characteristic variables of coal mine gas emission are replaced by some characteristic variables with cumulative influence weight of 90%. And eight characteristic variables with high importance are selected as input variables of the model, including coal mining height, coal thickness, coal seam gas content, recovery rate, burial depth, daily progress, mining intensity and adjacent layers spacing. The test results show that the random forest regression model with all characteristic variables and some characteristic variables has good prediction performance. After selecting characteristic variables, the average absolute error of the model decreases from 022 m3/min to 021 m3/min, and the average relative error decreases from 355% to 347%. The random forest regression model based on characteristic variable selection reduces the dimensionality of the characteristic variables of the prediction model, reduces the original data acquisition work, and improves the prediction efficiency under the premise of ensuring better prediction performance.
Study on the distribution of O2 concentration field of coal spontaneous combustion in high ground temperature goaf
LIU Yikang, NIU Huiyong, NIE Qimiao, LU Yi, LI Shilin
2021, 47(8): 108-114. doi: 10.13272/j.issn.1671-251x.2020120021
<Abstract>(26) <HTML> (8) <PDF>(9)
Abstract:
It is difficult to fully reflect the distribution of O2 concentration field of coal spontaneous combustion in goaf by using theoretical analysis and experimental research methods to study the impact of high ground temperature on coal spontaneous combustion in goaf. Fluent numerical simulation software is used to analyze the distribution law of O2 concentration field in the inlet air side, return air side and the middle section of the high ground temperature goaf. The results are listed as follows. ① When the temperature increases from 24.8 ℃ to 40 ℃ with the same ventilation volume, O2 flows into the whole goaf with the wind, and the O2 concentration decreases with the increase of goaf depth. When the air volume increases from 1 800 m3/min to 2 700 m3/min with the same temperature, the air leakage range in goaf increases significantly, the O2 concentration field in goaf changes obviously, and O2 almost fills the whole goaf. Moreover, the existence range of high concentration O2 increases, then the temperature of goaf increases due to heat accumulation, and the temperature of the residual coal inside goaf also continues to increase, the coal-oxygen reaction accelerates, and the possibility of spontaneous combustion of the residual coal increases. ② As the distance between goaf and the working face increases, the O2 concentration decreases. The O2 concentration on the inlet air side is greater than the O2 concentration on the return air side, indicating that the risk of coal spontaneous combustion on the inlet air side is greater than that on the return air side. ③ As the depth of goaf increases, the volume fraction of O2 on the inlet air side and the middle section of goaf continues to decrease, and the slope of the curve increases first and then decreases. The volume fraction of O2 on the return air side decreases with the increase of goaf depth. A large amount of high concentration O2 exists before 150 m of goaf, and the risks of coal spontaneous combustion in the inlet air side and the middle section of the whole goaf are greater than in the return air side. ④ When the temperature is 40 ℃ and the ventilation volume is 2 700 m3/min, the maximum width of the oxidation zone is 126 m. This maximum width is regarded as the maximum theoretical width of mining. Further calculation of the safe advancing speed can provide a theoretical basis for coal mining.
Design of networking platform for coal mine safety monitoring system based on localizatio
CHEN Qing
2021, 47(8): 115-120. doi: 10.13272/j.issn.1671-251x.2021030093
<Abstract>(162) <HTML> (17) <PDF>(14)
Abstract:
The existing coal mine safety monitoring system networking platforms are mainly based on the software and hardware equipment of foreign companies. Therefore, the proposed platform is domestically adapted to respond to the call of the strategic goal of independent and controllable national information security. A networking platform for coal mine safety monitoring system based on domestic chips, domestic operating system, domestic database and domestic middleware is designed. The software and hardware of the platform are adapted to the domestic software and hardware environment, and the platform is equipped with a small number of open source frameworks in the Java ecosystem. Both the server and the terminal CPU adopt Feiteng processors, the printer adopts Lanxum printer. The server operating system adopts Galaxy Kylin, and the terminal operating system adopts Union Tech. The database adopts the open source memory database Redis, the domestic relational database Dameng and the open source columnar storage database HBase. The middleware adopts TongLINK/Q and TongWeb from Tong Tech. The office suite adopts Kingsoft WPS, and the browser adopts Firefox. The memory database Redis is used to build real-time data cache region to achieve efficient access to hot data. Based on the performance index of Dameng database, the storage structure is designed through database table partitioning and other modes. Based on columnar storage database Hbase, a historical data storage system is designed to realize retrospective analysis of historical data. The field test results show that the data transmission delay of the platform is less than 10 s, the real-time data query time is in the range of 240-300 ms, and the platform supports the second-level backtracking of historical sampling data. These characteristics can meet the application requirements of localization of coal mine supervision and monitoring.
Design of gas electric locking detection device for safety monitoring system
DAI Wanbo
2021, 47(8): 121-127. doi: 10.13272/j.issn.1671-251x.2020090027
<Abstract>(89) <HTML> (9) <PDF>(11)
Abstract:
Coal mine safety monitoring systems mostly use CAN bus-based multi-master communication mechanism or RS485 bus-based master-slave communication mechanism, which calculates the power-off time by intercepting the time of methane overrun messages and the time of power-off messages of the breaker. Since the communication mechanism and communication protocol of the monitoring system of different manufacturers are different, it is impossible to design a unified gas locking detection method and device. At present, there are many researches for shortening the power-off time of gas electric locking, but there are few researches on gas electric locking detection. In order to solve the above problems, a gas electric locking detection device for safety monitoring system is designed based on the detection method of sensor communication switching and analog sensor message sending. The sensor communication switching method is not limited by the communication protocol and bus form, and the same manufacturer's sensors, substations, circuit breakers and safety monitoring software are used for detection, without the need to know the communication protocol. This method is simple to operate and does not require the cooperation of the manufacturer. It is suitable for gas electric locking detection using the master-slave communication mechanism. The analog sensor message sending method uses the detection device to simulate a methane sensor to send messages to communicate with the substation. There is no need to know the communication protocol or consider the communication mechanism. The messages are provided by the manufacturer and configured according to the order and times of message sending. This method is more cumbersome to operate and requires the manufacturer to provide communication messages. However, the test accuracy is higher and the method is suitable for gas electric locking detection of various communication protocols. The verification results show that the sensor communication switching method is not suitable for CAN bus-based multi-master communication mechanism. The analog message sending method based on RS485 or CAN bus can achieve local power-off time less than 2 s and remote power-off time less than 20 s. The timing accuracy of the two detection methods can both reach 0.01 s.
.Precise personnel positioning method in underground mine based on grey prediction model
TANG Lijun, WU Wei, LIU Shisen
2021, 47(8): 128-132. doi: 10.13272/j.issn.1671-251x.2021060027
<Abstract>(88) <HTML> (9) <PDF>(12)
Abstract:
The positioning accuracy of the precise personnel positioning system in underground mine is affected by the non-line-of-sight error and clock error. At present, the system mostly uses Kalman filter-based positioning method to reduce the error, but the positioning accuracy is not high when there is gross error in the measured data. In order to solve this problem, a precise personnel positioning method in underground mine based on grey prediction model is proposed. When a person carrying a marker card enters the coverage area of the positioning reader, the positioning reader calculates the measured distance between the marker card and the reader through wireless positioning technology and stores the measured distance into the data cache area. According to the measured distance in the data cache area, the GM (1, 1) model is used to calculate the predicted distance between the marker card and the reader at the next moment. When the prediction accuracy level of this predicted distance is excellent and the difference with the measured distance exceeds the error judgment threshold, the predicted distance is used to replace the measured distance to achieve the optimal compensation of the distance measurement error. The test results show that the method is not affected by the distance measurement error. When there is a gross error in the measured distance, the positioning accuracy of this method is significantly better than that of the Kalman filter-based positioning method.

苏ICP备 05016349号-2

Supported by: Beijing Renhe Information Technology Co. Ltd.  Visits：