工矿自动化

Research and development of 5G communication system standards for coal mines

SUN Jiping

2023, 49(8): 1-8. doi: 10.13272/j.issn.1671-251x.18147

<Abstract>(1189) <HTML> (63) <PDF>(126)

Abstract:
In order to meet the needs of remote monitoring, video monitoring, data acquisition, and voice communication in coal mines, the 5G communication system used in coal mines should have the following functions. ① The system has different service-bearing functions such as remote control, monitoring, positioning, surveillance, and voice. ② The system has remote control functions such as coal mining machines, roadheaders, electric shovels, excavators, trackless rubber wheeled vehicles, and electric locomotives. ③ The system has an emergency remote takeover function for mining transportation vehicles. ④ The system has a remote real-time transmission function of camera audio and video. ⑤ The system has data collection functions such as monitoring equipment, sensors, and vehicle-assisted driving. ⑥ The system has a voice call function. ⑦ The system has an end-to-end slicing function that meets the differentiated business performance requirements of remote control, monitoring, video, and voice. ⑧ The system supports SA networking and 5G NR communication system. ⑨ The system supports 5G LAN Ethernet communication. ⑩ The system has an emergency inertia operation function. In case of disconnection between the mining area's private network and the communication operator's public network, local businesses can continue to operate online. ⑪ The system has a device level redundancy protection function that ensures uninterrupted data service in the event of a single physical port failure. ⑫ The system has a dual device redundancy protection function of the core network that allows for the switching of backup devices to continue providing services when the main device fails. ⑬ The system has the core network control surface transmits confidentiality and integrity protection functions to ensure the security of the core network control surface. ⑭ The system has terminal authentication, checking, and restricting access to unauthorized terminals in the system, supporting the authentication of terminals by coal mining enterprise security servers. ⑮ The system has functions that prevent terminal attacks on the system and legitimate terminal. ⑯ The system has the integrated management function of the core network, transmission equipment, base station controller, base station, and terminal. ⑰ The system has a centralized monitoring function for network performance and business service performance. ⑱ The system has an abnormal visual alarm and fault location function. ⑲ The system has the evaluation function of mining 5G network resources. The system can evaluate the utilization rate of 5G network resources and provide a report on whether new services can be accessed when the coal mine adds new services or more terminals are connected to the 5G network. ⑳ The system has backup power supply. The main technical indicators of the 5G communication system used in coal mines should meet the following requirements. ① When the uplink rate is 20 Mbit/s and the wireless working frequency band is 700-900 MHz, the wireless coverage radius (unobstructed) of the base station in the underground coal mine should be ≥ 500 meters. When the wireless working frequency band is other working frequency bands, the wireless coverage radius (unobstructed) of the base station in the underground coal mine is ≥ 150 m. When the uplink rate is 30 Mbit/s, the wireless coverage radius (unobstructed) of the base station in the open-pit coal mine is ≥ 400 m. ② The wired transmission distance from the base station to the base station controller is ≥ 10 km. ③ The maximum number of access terminals in the system is ≥ 20000. ④ The wireless transmission power of the base station and terminal of the underground coal mine is ≤ 6 W. The transmission power of the base station in the open-pit coal mine is ≤ 320 W. The wireless transmission power of the terminal in the open-pit coal mine is ≤ 6 W. ⑤ The base station wireless reception sensitivity is ≤ −95 dBm. The terminal wireless reception sensitivity is ≤ −85 dBm. ⑥ The wireless working frequency should be selected from the frequency bands of 700 MHz, 800 MHz, 900 MHz, 1.9/2.1 GHz, 2.6 GHz, 3.3 GHz, 3.5 GHz, 4.9 GHz, 6 GHz, etc. (preferably 700 to 900 MHz for underground coal mine). ⑦ When the format is TDD and the frame structure is 1D3U1S, the average uplink throughput rate of multiple users accessed by the base station is ≥ 600 Mbit/s, and the average downlink throughput rate is ≥ 250 Mbit/s. ⑧ For underground coal mines, when operating upstream services at 1 Mbit/s and 20 Mbit/s, the average system delay should be less than 20 ms, and the probability of end-to-end delay stability being less than 100 ms should not be less than 99.99%. For open-pit coal mines, when operating upstream services at 1 Mbit/s and 30 Mbit/s, the average system delay should be less than 20 ms, and the probability of end-to-end delay stability being less than 100 ms should not be less than 99.9%. ⑨ The packet loss rate of a single user is ≤ 0.01%. ⑩ The handover delay for a single user from cell A of the base station to cell B of the base station is ≤ 100 ms. ⑪ The continuous working time of the mobile station battery should not be less than 11 hours, among which the call time should not be less than 2 hours. ⑫ After a power outage in the power grid, the backup power supply continuously provides power to the base station, base station controller, and transmission equipment for ≥ 4 hours.

Research on fault warning technology for cutting part of cantilever roadheader based on virtual and real fusion data

ZHANG Xuhui, BAI Linna, YANG Hongqiang

2023, 49(8): 9-19. doi: 10.13272/j.issn.1671-251x.2023050063

<Abstract>(1103) <HTML> (57) <PDF>(43)

Abstract:
Currently, the fault warning technology for the cutting part of cantilever roadheader relies on traditional data collection methods. In the operation process of the cutting part of the roadheader, problems such as difficulty in obtaining signals and high noise limit the capability to predict and warn faults in the cutting part of the roadheader. In order to solve the above problems, a fault warning method for the cutting part of cantilever roadheader based on virtual and real fusion data is proposed. The method performs three-dimensional solid modeling of the cutting section of a cantilever roadheader. It uses the automatic dynamic analysis of mechanical systems (ADAMS) to obtain virtual data of the cutting section's mechanical system, constructs its dynamic simulation model to obtain virtual data. The method uses the cosine similarity function to characterize its similarity with real data to verify the credibility of the virtual data. The method uses Bayesian estimation and adaptive complementary weighted fusion methods to perform similarity association and complementary association fusion on virtual and real data, respectively, to obtain virtual and real fusion data. In response to the problem that the learning efficiency of traditional self-organizing mapping (SOM) neural networks is easily affected by the learning rate, a fault warning model based on an improved SOM neural network is established. A monotonic decreasing function about time is introduced to train the SOM neural network, ensuring both the learning rate and the stability of the model. The method inputs the fused data into the fault warning model based on SOM neural network to determine the winning neuron and adjust its weight. The method calculates the distance between the real data and the winning neuron and adjusts its weight to achieve fault warning. The experimental results show that the average operating efficiency of the improved SOM neural network can be improved by 35.84%. The fault warning method for the cutting part of a cantilever roadheader based on virtual and real fusion data can successfully predict the types of single and composite faults, with a prediction accuracy of 83.33%.

Status and prospect of the application of mine DC electrical method technology

YANG Shaowen, ZHANG Pingsong, XU Shi'ang, WU Haibo, QIU Shi, JIAO Wenjie

2023, 49(8): 20-29. doi: 10.13272/j.issn.1671-251x.2023050099

<Abstract>(344) <HTML> (63) <PDF>(42)

Abstract:
As an efficient means of geophysical exploration, mine DC electrical method plays an important role in accurately circling various types of anomalous zones. The development of underground detection is limited by many factors due to small space, interference and high technological requirements. Therefore, it is significant to establish a mine DC electrical method technology system that matches the intelligent mine production under the rapid extraction mode. The paper gives an overview of mine DC electrical method from three aspects, namely, basic principle, technology development and classification. It summarizes the latest progress of mine DC electrical method used in roof and floor exploration, roadway advance detection, and anomaly area exploration in the working face, etc. It analyzes the progress of the research and development of mine DC electrical method instruments and equipment. It enumerates several common types of mine DC electrical method instruments. It analyzes the key problems of mine DC electrical method in solving the engineering problems. ① The current mine DC electrical method advance detection technology has low positioning precision in the circled space of water-bearing/conducting anomalies. At the same time, there is the problem of insufficient effective detection distance. ② The construction space of mine DC electrical method is narrow, and the electrical response of multi-directional geological anomalies is superimposed in the limited testing space. It increases the difficulty of data processing and interpretation. ③ The mine DC electrical method is susceptible to interference from metal sources at the site when applied underground, especially by large metal parts such as roadheaders, hydraulic supports, anchor locks (nets) supports, rails, and conveying pipelines. The future development direction of the mine DC electrical method is prospected. ① It is suggested to construct a multi-source geoelectric field data response feature library. ② It is suggested to obtain interpretation of multi-source data fusion. ③ It is suggested to establish the intelligent monitoring system of mine DC electrical method.

Design of end controller for the electrohydraulic control system of intelligent working face hydraulic support

ZHANG Xiaohai, TIAN Muqin, ZHANG Minlong, SONG Jiancheng, XU Chunyu, NIE Honglin, YANG Yongkai

2023, 49(8): 30-36. doi: 10.13272/j.issn.1671-251x.2023060031

<Abstract>(278) <HTML> (60) <PDF>(30)

Abstract:
With the continuous promotion of the construction of unmanned automated fully intelligent mechanized working faces, higher technical requirements have been put forward for the automation control function of the hydraulic support electrohydraulic control system. The electrohydraulic control technology developed in China has problems such as low communication speed, delayed response, and poor reliability in meeting the requirements of intelligent production technology. An end controller for the electrohydraulic control system of hydraulic support based on a 32-bit processor has been developed. A communication architecture of the end controller based on industrial Ethernet and CAN bus has been designed. According to the technical requirements of intelligent perception, intelligent decision-making, and automatic control for unmanned intelligent mechanized working faces, parameter inspection, parameter modification, online upgrade, and control functions of automatic follow-up have been designed in the end controller. In order to meet the requirements of standardization and normalization of data in the hydraulic support electrohydraulic control system in intelligent fully mechanized working faces, the end controller can encode the data generated by the hydraulic support electrohydraulic control system according to the data encoding standard based on tag numbers. Through the experiment on the "three machines" experimental platform of fully mechanized working faces, the results show the following points. The entire process from issuing inspection instructions to receiving data from 27 support controllers on the experimental platform takes 1.8 s for the end controller. It is 1.5 s faster than using RS485 communication to achieve parameter inspection. The size of the upgrade program sent by the end controller is 38 KiB and the transmission time is 1.2 s. After testing, it takes 4-6 s for all support controllers in the fully mechanized working face to receive the upgrade command and successfully upgrade together, achieving the expected goal. The end controller can control the corresponding hydraulic support to make correct actions based on the position of the shearer. It can meet real-time requirements.

Shape monitoring of scraper conveyor based on inertial measurement unit

WEI Dong, LI Zuxu, SI Lei, TAN Chao, WANG Zhongbin, LIANG Bin, XIAO Junpeng

2023, 49(8): 37-52, 80. doi: 10.13272/j.issn.1671-251x.2023010003

<Abstract>(227) <HTML> (61) <PDF>(37)

Abstract:
Scraper conveyor is the core transportation equipment of the fully mechanized working face. Accurately perceiving its form is an important prerequisite to enhance its carrying capacity, alleviate the transmission impact, and improve the straightness of fully mechanized working face. The commonly used indirect measurement methods for the shape of scraper conveyors are difficult to accurately characterize their shape, resulting in significant measurement model errors. To address this issue, an inertial measurement unit is used to directly measure the original pose information of the middle trough of scraper conveyor, achieving accurate acquisition of the shape data of scraper conveyor. A wavelet thresholding denoising method that combines Heursure threshold rules and a new threshold function is used to filter out noise interference in the acceleration signal of the middle trough. Based on this, the motion features of the middle trough are analyzed, and a middle trough motion state recognition model based on random forest algorithm is designed. Based on the motion state recognition results, different strategies are used to update the position of the middle trough. It reduces the accumulated IMU data error over time and improves the precision of IMU position calculation. The improved Harris hawk optimization (HHO) algorithm unscented Kalman filter (UKF) is designed for middle trough attitude calculation. It is verified through experiments that the attitude angle calculated by this method meets the requirements of middle trough attitude measurement. The experimental platform for shape monitoring of scraper conveyors is constructed. It conducts experimental verification on the shape calculation method of scraper conveyors based on motion state recognition and improved HHO optimized UKF. The results show that when the scraper conveyor performs a single sliding with a step distance of 250 mm, the maximum cumulative errors of displacement in the X and Y directions of the scraper conveyor composed of 10 middle troughs are 6.4 mm and 8.4 mm respectively under the horizontal working condition of bottom plate. It remains unchanged in the Z direction. The maximum cumulative errors of pitch angle, roll angle, and heading angle are −0.148°, −0.035°, and 0.457° respectively. Under the working condition of floor undulation, the maximum cumulative errors of displacement in the X, Y, and Z directions are 6.6 mm, 11.5 mm, and 6.9 mm respectively. The maximum cumulative errors of pitch angle, roll angle, and heading angle are −0.540°, −0.157°, and 0.817° respectively. This method can effectively suppress cumulative errors, reduce measurement errors, and achieve accurate perception of the shape of the scraper conveyor.

Path planning of coal mine rescue robot based on improved A^* algorithm

JIANG Yuanyuan, FENG Xueyan

2023, 49(8): 53-59. doi: 10.13272/j.issn.1671-251x.2022120027

<Abstract>(820) <HTML> (26) <PDF>(55)

Abstract:
Path planning is one of the important contents of research on coal mine rescue robots. A path planning method for coal mine rescue robots based on improved A^* algorithm is proposed to address the unstructured features of post disaster coal mine environments and the problems of non-shortest path length, multiple turns, and poor smoothness of path planned by traditional A^* algorithm. The method constructs raster maps by binarizing map information in real environments, determines the relative position between the current point and the target point, and uses the improved A^* algorithm for path planning. Then a path from the current point to the target point is obtained. Douglas-Pucker (D-P) algorithm is used to extract key nodes on the path, and cubic spline interpolation function is used to fit the key nodes, thereby completing the smooth processing of the path. The improved A^* algorithm expands the traditional A^* algorithm's 8 neighborhood search to a purposeful 13 neighborhood search. When conducting path search, the position relationship between the current point and the target point is first determined, thereby reducing path nodes and length, and improving path smoothness. The Matlab simulation results show that compared with the 8 neighborhood A^* algorithm, 24 neighborhood A^* algorithm, and 48 neighborhood A^* algorithm, the improved A^* algorithm has certain optimizations in path length, number of turns and smoothness. It is more suitable for path planning of coal mine rescue robots. Compared with the Fuzzy algorithm, the improved A^* algorithm achieve shorter path planning time, shorter planned path length, and fewer turns.

Efficient task assignment algorithm for coal mine underground group robots

WU Wenzhen

2023, 49(8): 60-69. doi: 10.13272/j.issn.1671-251x.2022120067

<Abstract>(143) <HTML> (24) <PDF>(22)

Abstract:
The loose cooperative group robot system has broad application prospects in the current coal mine auxiliary robot operation. However, the task assignment process of the loose cooperative group robot system did not provide feedback to the division process, resulting in insufficient efficiency and rationality of the task division and assignment process. To address this issue, an efficient task assignment algorithm for coal mine underground group robots based on an improved Rubinstein negotiation strategy is proposed. Based on the multi-party game features of task division and assignment in group robot systems, the Rubinstein negotiation strategy is extended from a bipartite game to a multi-party joint game. A "bid-bargain-counteroffer" rule for multi-party negotiation games is proposed. From the perspective of the difference between the execution capability and task execution status of individual robots, a discount factor calculation method based on the task completion quantity per unit time of robot individuals is proposed. A task completion status feedback parameter model based on the task execution status of each assignment cycle is also proposed to achieve dynamic task division and assignment. By collaborating with three groups of robots to perform overall monitoring tasks in coal mining areas, experimental verification is conducted on the performance of the algorithm. The results show the following points. ① Algorithm 3 uses an improved Rubinstein negotiation strategy. Algorithm 1 directly uses the ratio of the number of unmanned aerial vehicles in each group multiplied by their running speed as the standard for task division and assignment in three groups of unmanned aerial vehicles. Algorithm 2 uses the Rubinstein negotiation strategy of multi-party negotiation without considering the feedback parameters of task completion status. Algorithm 3 has a higher efficiency in task division and assignment than Algorithm 1 and Algorithm 2 by 30.10% and 18.29% respectively. ② The average maximum time difference for the three groups of unmanned aerial vehicles based on Algorithm 3 to execute tasks is 42 seconds. It is 77.66% and 65.29% optimized compared to Algorithm 1 and Algorithm 2, respectively. This is because Algorithm 3 introduces task completion status feedback parameters to timely evaluate the task execution process of the task participants. Algorithm 3 provides feedback on the task assignment and execution process to the task division stages, making the task division and assignment more accurate.

A localization and navigation method for underground mine autonomous driving based on local geometric topology map

LIU Shijie, ZOU Yuan, ZHANG Xudong

2023, 49(8): 70-80. doi: 10.13272/j.issn.1671-251x.2023020010

<Abstract>(120) <HTML> (72) <PDF>(21)

Abstract:
Unmanned driving technology has enormous advantages in improving efficiency, saving costs and reducing safety hazards. In the current implementation of localization and navigation solutions in underground environments, there are problems of implementation difficulties, high costs, and time-consuming construction of maps. In order to solve the above problems, a localization and navigation method for underground mine autonomous driving based on local geometric topology map is proposed. A local geometric topology map has been designed. The main structure of the underground environment road network is represented by a topology map. The map defines roadways (sides) and intersections (nodes), and stores a local geometric map built around the node in each node to achieve precise positioning at the node. A localization method based on local geometric topology map is proposed, which uses a LiDAR-based intersection detection algorithm and intersection localization algorithm for global vehicle localization. A trajectory-following algorithm based on adaptive model predictive control (MPC) has been designed to ensure the path-tracking precision of vehicles turning at high curvature intersections. A simulation environment and vehicle simulation model for underground mines are constructed by using a 3D physical simulation platform. The simulation results show that this method can achieve underground mine autonomous driving localization and navigation functions. The positioning errors are within 0.2 m at various types of intersections, meeting the positioning localization precision requirements of autonomous driving. Throughout the entire driving process, the vehicle maintains a relatively stable driving state and a small tracking error. Compared with the current localization and navigation methods that rely on technologies such as 5G and UWB, this method only relies on two types of vehicle sensors: LiDAR and inertial measurement unit. It has great advantages in controlling equipment costs.

Positioning method for underground unmanned aerial vehicles in coal mines based on global point cloud map

GAO Haiyue, WANG Kai, WANG Baobing, WANG Dandan

2023, 49(8): 81-87, 133. doi: 10.13272/j.issn.1671-251x.2022110024

<Abstract>(643) <HTML> (53) <PDF>(34)

Abstract:
When simultaneous localization and mapping (SLAM) technology is applied to autonomous positioning of unmanned aerial vehicles in coal mines, the use of feature points to construct maps can easily lead to degradation issues, resulting in inaccurate positioning. Moreover, due to its use of the body as a reference coordinate system, global positioning cannot be achieved. In order to solve the problems, a positioning method for underground unmanned aerial vehicles (UAV) in coal mines based on global point cloud map is proposed. The method uses Fast-LIO2 algorithm as the lidar SLAM algorithm to obtain UAV position and attitude estimation. An iterative nearest-neighbor algorithm is used for two-step matching of the acquired real-time lidar point cloud and the global point cloud map to achieve UAV position and attitude correction. To address the issue of point cloud matching speed not ensuring real-time positioning due to the excessive number of point clouds, a time-based position and attitude output strategy is designed to increase the frequency of outputting UAV position and attitude data. The SLAM precision and position and attitude correction effect of the UAV positioning method are tested in a 1 000 m underground coal mine roadway. The results show that in long-distance roadway environments, the cumulative positioning error of the Fast-LIO2 algorithm is less than 1 m, and is less than 0.3 m in the range of 600 m or more, which is significantly smaller than the cumulative positioning errors of LOAM-Livox algorithm and LIO-Livox algorithm. The position and attitude estimation output by the Fast-LIO2 algorithm has been corrected by the correction algorithm, and all flight paths are located in the global point cloud map, verifying the effectiveness of the position and attitude correction algorithm. The time consumption of single SLAM algorithm operation is 14.83 ms, the one of single position and attitude correction is 883 ms, and the output frequency of position and attitude data is 10 Hz, meeting the real-time requirements of UAV positioning.

Research on high sampling frequency mine electric spark image recognition and anti-interference methods

LI Xiaowei, WANG Jianye

2023, 49(8): 88-93, 147. doi: 10.13272/j.issn.1671-251x.18145

<Abstract>(579) <HTML> (58) <PDF>(19)

Abstract:
Leakage of electricity from cables and electrical equipment outside the explosion-proof enclosure, and mine sparks generated by high-power radio transmissions on metal supports and metal of electromechanical equipment due to induced electromotive discharges, can cause gas and coal dust explosions and mine fires. Therefore, it is necessary to detect mine electrical sparks as soon as possible. The main factor affecting the recognition of mine electric sparks is the mine light source. In order to reduce the interference of mine light sources on mine electric spark image recognition, a high sampling frequency mine electric spark image recognition and anti-interference method has been proposed. Based on the longest continuous emission time of the electric spark and the shortest continuous emission time of the flash light source, the sampling frequency of the camera is calculated to ensure that the electric spark image only appears in one frame of the image each time the electric spark appears. When the mine light source exists, the interference light source image appears on at least 2 consecutive frames of image. The method calculates the pixel grayscale sum of each image frame. If the difference between the pixel grayscale of the current frame image and the pixel grayscale sum of adjacent frames is greater than the set threshold, a mine electric spark alarm signal will be issued. The experimental results show that under the condition of no interference light source, this method can accurately recognize mine electric spark images with an accuracy rate of 100%. Under the interference of constant light sources such as fluorescent lamps and incandescent lamps, the recognition accuracy of electric sparks in mixed images of electric sparks and fluorescent lamps reaches 99.40%. The recognition accuracy of electric sparks in mixed images of electric sparks and incandescent lamps reaches 99.67%. Under the interference of a flashing light source, the accuracy of electric spark recognition in the mixed image of electric spark and flash lamp reaches 100%.

Mine image enhancement method based on multi-scale local histogram equalization

TU Yihan, WANG Puqing

2023, 49(8): 94-99. doi: 10.13272/j.issn.1671-251x.2023010015

<Abstract>(645) <HTML> (49) <PDF>(13)

Abstract:
There are problems of under-enhancement and over-enhancement in commonly mine image enhancement methods such as histogram equalization, Retinex theory, homomorphic filtering, wavelet analysis, etc. In order to solve the above problems, a mine image enhancement method based on multi-scale local histogram equalization is proposed. According to the independent features of color components (hue component and saturation component) and brightness component of image in HSI color space, the low-light RGB mine image is converted into the HSI color space. The method uses bilateral filtering to decompose the brightness component into lighted images and reflected images. The method divides the lighting image into small, medium, and large blocks, and performs local histogram equalization on each image block to improve image brightness and contrast. The method performs 8-direction gradient enhancement on the reflected image to enrich the texture edges of the image. The method performs Retinex inverse transformation on the light image after multi-scale local histogram equalization and reflection image after directional gradient enhancement to obtain the enhanced brightness component. Then the brightness, hue and saturation components are transformed into RGB color space to obtain an enhanced mine image. Experimental verification of the mine image enhancement method based on multi-scale local histogram equalization is conducted by using actual monitoring images of coal mines. The enhancement effect is evaluated subjectively and objectively. The results show that compared with existing image enhancement methods, this method has a greater improvement in image brightness and contrast with richer detail information. The information entropy has increased by over 7.23%, and the mean average gradient has increased by over 31.6%. It has better image enhancement effects.

Lightweight multi-scale cross channel attention coal flow detection network

ZHU Fuwen, HOU Zhihui, LI Mingzhen

2023, 49(8): 100-105. doi: 10.13272/j.issn.1671-251x.2023030045

<Abstract>(166) <HTML> (49) <PDF>(21)

Abstract:
In order to improve the operating efficiency of belt conveyors through variable frequency speed regulation, it is necessary to detect the coal flow of belt conveyor. The existing deep learning-based coal flow detection methods for belt conveyors are difficult to achieve a balance between model lightweight and classification accuracy. There are few researches on the impact of imbalanced channel weight distribution on detection accuracy in the feature extraction process. In order to solve the above problems, a lightweight multi-scale cross channel attention coal flow detection network is proposed. The network consists of a feature extraction network and a classification network. The lightweight residual network ResNet18 is used as the feature extraction network, and on this basis, the coal flow channel attention (CFCA) subnetwork is introduced. The CFCA subnetwork uses multiple one-dimensional convolutions with different kernel sizes, and stacks the output of one-dimensional convolution to capture cross channel interaction relationships at different scales in the feature map. It achieves the reassignment of feature map weights, thereby improving semantic expression capability of the feature extraction network. The classification network consists of three fully connected layers, which take the output of the vectorized feature extraction network as input and perform nonlinear mapping on it. It ultimately obtains the probability distribution of three types of results: "little coal", "moderate coal", and "much coal". By transforming the coal flow detection problem into an image classification problem, the problem of frequent frequency conversion and speed regulation of belt conveyors caused by excessive fluctuations in instantaneous coal flow is avoided. It improves stability of belt conveyor operation. The experimental results show that the ResNet18+CFCA network improves classification accuracy by 1.6% compared to the ResNet18 network, with almost no increase in network parameters and computational complexity. It can distinguish foreground information in images more effectively and accurately extract coal flow features.

A fault diagnosis method for roller based on small sample sound signals

HAO Hongtao, QIU Yuanyuan, DING Wenjie

2023, 49(8): 106-113. doi: 10.13272/j.issn.1671-251x.2022120007

<Abstract>(678) <HTML> (54) <PDF>(32)

Abstract:
Fault diagnosis methods based on deep learning have high requirements for the quality of the dataset, requiring a large amount of data for good model training to achieve accurate fault diagnosis. However, the fault signals that can be collected in practical applications are usually limited. A method for diagnosing roller faults based on small sample sound signals is proposed to address the problem of limited performance of intelligent fault diagnosis methods due to the difficulty in obtaining sound signals for roller faults and the small sample size. The feature transformation method is used to convert one-dimensional sound signals into two-dimensional time-frequency images, incorporating features from the frequency domain to improve the dataset's capability to express fault features. A dataset expansion method combining multiple types of time-frequency maps has been proposed. The method combines time-frequency maps drawn by three time-frequency analysis methods: short time fourier transform (STFT), continuous wavelet transform (CWT), and Hilbert Huang transform (HHT) to expand the dataset and increase data styles. The concept of deep transfer learning is introduced, using bearing datasets to pre-train the model, and then using roller data to fine-tune the pre-trained model to further improve the recognition accuracy of the model. The experimental results show that the dataset expansion method combining multiple types of time-frequency maps can effectively solve the problem of overfitting when training models with small sample data. After using transfer learning, the testing accuracy of the model reaches 98.81%, an improvement of 7% compared to not using transfer learning. There was no overfitting phenomenon, indicating that the model is well-trained. Compared to the method of generating adversarial networks to expand the STFT time-frequency map dataset and transfer learning, the method of dataset expansion by combining multiple types of time frequency maps and transfer learning has an accuracy improvement of 4%. It is easier to implement, and has stronger interpretability.

Research on the application of time shifting aeromagnetic method in detecting coal mine burning areas

YU Yongning, LI Xiongwei, SHI Lei, LIU Kaiyuan, GUO Jianlei, MA Guoqing

2023, 49(8): 114-120. doi: 10.13272/j.issn.1671-251x.2022110027

<Abstract>(169) <HTML> (66) <PDF>(16)

Abstract:
The spontaneous combustion of coal seams leads to the formation of magnetic minerals in the overlying strata, exhibiting high magnetic anomaly features, providing a physical prerequisite for the magnetic method to detect the burning area. The aeromagnetic method has achieved good results in detecting coal mine burning areas, but it cannot effectively detect the development trend of coal mine burning areas. In order to solve the above problems, based on the aeromagnetic method method, a time-shifting aeromagnetic method is proposed. It involves conducting two aeromagnetic detections within a certain time interval. Based on the difference between the two aeromagnetic inversion results, the features of the coal mine burning area over time are determined. It achieves the goal of effectively detecting the distribution range and development trend of the coal mine burning area. In order to balance the terrain fitting effect and inversion calculation efficiency in undulating areas, a composite mesh generation method of regular and irregular grids is adopted. The tetrahedral irregular grid generation is used in undulating areas on the surface, and hexahedral regular grid generation is used in areas below the surface. The results show that the regular and irregular composite mesh generation method not only meets the requirements for inversion precision under undulating terrain conditions, but also improves the inversion calculation efficiency by nearly 6 times compared to the tetrahedral irregular mesh generation method. A numerical model is established based on actual geological conditions. The actual testing is conducted using unmanned aerial vehicles and aviation optical pump magnetometers. The numerical simulation and actual measurement results indicate that the time-shifting aeromagnetic method can accurately detect the distribution range of burning areas and the development trend of burning areas over time. It provides a basis for carrying out fire prevention and extinguishing work in coal mines.

Method for detecting the status of retaining walls in intelligent mines

XU Lianhang, LI Xi, GUO Xusen, LI Jing

2023, 49(8): 121-126. doi: 10.13272/j.issn.1671-251x.2023040036

<Abstract>(201) <HTML> (74) <PDF>(31)

Abstract:
During the driving process of unmanned vehicles in mines, if the retaining wall in the mining area is damaged and not detected and repaired in a timely manner, the vehicle may exceed the safety range of the retaining wall during driving or unloading. It can easily cause safety accidents. The existing methods for detecting the status of retaining walls are mostly based on point cloud data collected by vehicle and drone sensing devices. The methods have limited field of view, high sparsity and poor stability. There is a lack of detection methods for the integrity status of retaining walls. In order to solve the above problems, a method for detecting the integrity of retaining wall status based on roadside LiDAR sensors is proposed. A high-resolution roadside LiDAR sensor is used to collect point cloud data of the retaining wall in the driving area of the vehicle. Polygonal area filtering and voxel rasterization are used to obtain complete point cloud data of the retaining wall. A sliding trace search technique is used to divide the retaining wall into sub units along its extension direction to accommodate the different shaped retaining walls. In response to the problem of false detection caused by uneven mining sites and sparse remote point cloud data, a dual threshold method of height difference threshold and density threshold is adopted. It detects the integrity of the entire retaining wall status by detecting the defects of sub units. The method collects point cloud data of "L" and "S" type retaining walls in a mining area in Inner Mongolia. The on-site experiments are conducted in both occluded and unobstructed scenarios. The results show that this detection method has strong detection capability for defects in different shapes of retaining walls. The method can identify and mark the damaged parts of point cloud data in real-time.

Intelligent measurement and control system of mine water level based on Raspberry Pi

CHEN Haijian, WANG Weiyi, FAN Jinge, PAN Yidong, YAN Ziji, WU Baolei

2023, 49(8): 127-133. doi: 10.13272/j.issn.1671-251x.2022110072

<Abstract>(227) <HTML> (70) <PDF>(29)

Abstract:
The current water level monitoring methods have the problems of low precision, susceptibility to environmental impact, weak real-time performance, high requirements for machine computing power, and high hardware costs. In order to solve the above problems, a Raspberry Pi-based intelligent water level measurement and control system for underground water storage is proposed. The system collects water level images around the water tank scale through explosion-proof monitoring cameras, and uses raspberry pie as the image processing platform. Firstly, the method converts the collected color images into grayscale images, and uses the Otsu method to perform threshold segmentation on the images. The method removes noise and enhances image edge information through morphological operations, and then separates the ruler contour from the background. Secondly, the Canny operator is used to detect the edge of the scale, and the Hough transform method is used to extract the intersection line between the water level line and the vertical edge of the scale, obtaining the coordinates of the water level line in the image space. Thirdly, threshold segmentation and filtering enhancement processing are performed on the digital image of the scale within a certain range of the area near the water level line. Then, the template matching method is used to achieve the recognition of the scale number, thereby obtaining the water level line value. Finally, the method converts the numerical value of the water level line in the water tank into a current analog quantity, and uses Raspberry Pi to send the water pump controller to control the start and stop of the water pump based on the current magnitude. The method achieves intelligent control of the mine water level. This system has the advantages of low cost, convenient deployment, high precision, and good real-time performance. It can achieve rapid and accurate recognition and control of mine water level.

Optimization and transformation of ventilation system in Jining No.2 Coal Mine

ZHANG Yiran, TAO Weiguo, GUO Chuanqing, CHEN Xiujie, MIAO Dejun

2023, 49(8): 134-141, 155. doi: 10.13272/j.issn.1671-251x.2023020061

<Abstract>(232) <HTML> (42) <PDF>(16)

Abstract:
Currently, there's a lack of research on air volume regulation and mine resistance reduction of ventilation system in mine working face. In order to solve the above problem, taking 10303 working face and 33_low 02 working face of Jining No.2 Coal Mine as the engineering background, the original ventilation systems in these two areas are optimized and transformed in terms of air volume regulation and mine resistance reduction. The ventilation system diagram of the working face is imported into Ventism software, generating a solid roadway and iterating the calculation to construct a mine ventilation network solution model. The main parameters measured on-site are input into the model for airflow calculation. The errors between calculated relevant data such as flow velocity, temperature, and air volume in the roadway obtained and the on-site measurement data are within the standard range. From the measurement results of mine ventilation resistance, it can be seen that the original ventilation system has the following problems. The setting of the regulating air wall at the south wing stone gate is unreasonable. The actual air supply volume of 33_low02 working face is less than the ideal air volume. The ventilation route of the south wing -740 horizontal track main roadway is long. It is affected by the parallel intake of auxiliary transportation roadways, resulting in high resistance in the south wing return air main roadway. In order to solve the above problems, three renovation measures are proposed. ① A closed air door is installed at the intersection of the south wing return air stone gate and the north wing belt conveyor roadway. The original air window area of the south wing belt conveyor roadway and return air stone gate is adjusted to 2.9 m². ② A 0.1 m² adjustable wind window is installed at the intersection of the extension of the third mining area's track downhill and the 33_low02 connecting roadway. ③ The 0.9 m² adjustable air window at the interface between the pipe duct in the 11th mining area and the south wing -740 horizontal track roadwayhas been changed to 2.4 m²,so as to reduce the air volume of the south wing -740 horizontal track roadway and increase the parallel air volume of the auxiliary transportation roadway. The simulation results of the modified ventilation system show that the resistance of the southern wing -740 horizontal track main roadway has been reduced by 32.7%. The air volume of the 33_low02 working face has been increased by 19.8%. The total resistance of the mine ventilation route has been reduced by 6.4%. The on-site measurement results of the modified ventilation system show that the average relative error between the measured air volume and numerical simulation results is 1.28%. The average relative error between the measured resistance and numerical simulation results is 2.52%. The optimized simulation results are basically consistent with the on-site test results. The range of changes in air volume and resistance of the intake shaft before and after the entilation system adjustment is not significant. The air volume at the measuring point of the return air shaft decreases, and resistance decreased. The optimized measured air volume at the 33_low02 track connecting roadway and the measuring points of the working face increase by 25.3% and 21.4%, respectively, and the resistances increase by 57.4% and 41.1%. The optimized measured air volume at the south wing -740 horizontal track roadway decreases by 20.3%, and resistance decreases by 36.6%. After the renovation, the air volume of the working face and the total resistance of the mine have achieved the expected results.

Study on gas desorption dynamic features of mixed coal samples with different particle sizes

MA Xingying, GONG Xuanping, CHENG Xiaoyu, CHENG Cheng, LI Debo

2023, 49(8): 142-147. doi: 10.13272/j.issn.1671-251x.2022110069

<Abstract>(179) <HTML> (64) <PDF>(21)

Abstract:
Currently, research on the dynamic features of gas desorption mainly focuses on single particle size coal samples. There is less research on the dynamic features of gas desorption of mixed coal samples with different particle sizes. To solve this problem, a multi field coupled seepage desorption experimental system is used to mix coal samples with three different particle sizes (0,0.25) mm, [0.25, 0.5) mm, and [0.5, 1] mm in different proportions. Gas desorption experiments are conducted on mixed coal samples with different particle sizes. The changes in gas desorption kinetic parameters such as gas desorption amount, diffusion coefficient, and desorption attenuation coefficient are analyzed under different particle size coal sample proportions. The results indicate the following points. ① During the gas desorption process of mixed coal samples with different particle sizes, the main factor affecting the gas desorption amount in the early stage is particle size. In the later stage, the main factor affecting the gas desorption amount is the proportion of coal samples with different particle sizes. The larger the proportion of small coal particles, the greater the amount of gas desorption in the coal sample. ② The gas diffusion coefficient of mixed coal samples with different particle sizes exhibits temporal variability. As the gas desorption time increases, the gas diffusion coefficient decreases and eventually approaches 0. The initial gas diffusion coefficient decreases with the increase of the proportion of small particle coal. ③ The larger the proportion of small particle coal, the greater the attenuation coefficient of gas desorption. Therefore, in the process of underground gas content measurement, the proportion of large particle coal in the coal samples obtained should be increased as much as possible to reduce gas loss during the sampling process and improve the accuracy of gas content measurement.

Estimation of coal vitrinite reflectance based on random forest and dendritic network

YUAN Yilin, ZHAO Ronghuan, HE Kun, HUANG Xiu, WANG Hongdong, ZOU Liang

2023, 49(8): 148-155. doi: 10.13272/j.issn.1671-251x.18082

<Abstract>(172) <HTML> (56) <PDF>(17)

Abstract:
The mean maximum vitrinite reflectance is an important indicator of the degree of coalification, and plays a key role in determining coal grade, identifying mixed coal, and guiding coking coal blending. The traditional reflectance measurement methods are time-consuming and labor-intensive. The subjectivity of measurement results is strong, resulting in poor comparability of identification results between laboratories. To address this issue, a method for estimating coal vitrinite reflectance based on random forests(RF) and dendritic networks(DDNet) is proposed. It mainly includes three parts: coal rock microscopic image segmentation, vitrinite recognition, and mean maximum vitrinite reflectance prediction. The elbow method and K-Means algorithm are used to achieve segmentation of different maceral regions of the clustering microscopic images. The artificial minority oversampling method (SMOTE) is used to oversample minority samples to improve the imbalance between vitrinite and nonvitrinite regional samples in coal and rock. The DDNet-based regression algorithm is used to estimate the mean maximum vitrinite reflectance. When building a regression model, multiple 41×41 pixel square windows are selected from the vitrinite regions to extract their grey scale features. It improves the robustness of the algorithm, with a determination coefficient of 0.990. The experimental results show that using elbow method to automatically determine the parameter K of the K-Means algorithm, which has good adaptive capability. It can automatically distinguish different types of microscopic components. The SMOTE method can effectively avoid the problem of over-learning sample prior information, which leads to good recognition of the majority class and poor recognition of the minority class. It improves classification accuracy. Among them, the recognition model based on RF has an accuracy rate of 97.0%. Seven regression estimation models have been established, among which the DDNet regression model has the best performance, with a determination coefficient of 0.990. The predicted results are highly consistent with the actual values, verifying the feasibility of the proposed method.

Experimental study on the influence of cutting distance on the rock-breaking features of pick-shaped cutter

LIU Bin, LI Xuefeng

2023, 49(8): 156-164. doi: 10.13272/j.issn.1671-251x.2022120046

<Abstract>(147) <HTML> (26) <PDF>(27)

Abstract:
The pick-shaped cutter is the most widely used cutter type in mining machinery such as roadheader and coal mining machine. In the actual cutting process, the pick-shaped cutter mainly works under the multi-tooth coupling cutting condition. The cutting distance is an important parameter under this working condition. At present, research on the influence of cutting spacing on the rock-breaking process has not considered the weakening effect of interference cutting. A calculation method for cutting force during multi-tooth coupling cutting is proposed to solve the above problem. Full-size single-tooth cutting tests are carried out on limestone, red sandstone and two simulated rock samples, comparing and analyzing the rock-breaking processes of free cutting and interference cutting. The experiment collects cutting force data and conducts noise reduction processing and collects cutting debris to analyze the impact law of cutting spacing on cutting load, cutting debris coarseness, cutting energy consumption, and cutting grooves. The experimental results show the following points. ① The cutting force of the cutter increases with the increase of cutting distance and gradually approaches the free cutting state. Moreover, there is a good linear relationship between the cutting force ratio under interference cutting and free cutting conditions and the cutting distance/cutting depth. The correlation coefficients are greater than 0.95. The cutting load of the cutter under interference cutting conditions can be estimated using the free cutting load. The cutting force estimation equation under interference cutting conditions based on the existing peak cutting force model is obtained. ② The coarseness index (CI) and specific energy (SE) are used respectively to evaluate the particle size distribution and cutting energy consumption of the cutting experiment. As the cutting distance increases, CI shows a trend of first increasing and then decreasing. SE shows a trend of first decreasing and then increasing. ③ When the cutting distance is small, the interference between the cutting grooves is significant, the residual rock ridges between the cutting grooves are small, and the cutting load is small. However, due to the interference between the cutting grooves, more small debris is generated. It consumes more energy and increases energy consumption. As the cutting distance increases, the residual rock ridge increases, and the cutting force increases. However, due to the weakening effect of existing cutting grooves on the rock and less interference between cutting grooves, the proportion of large debris formed increases, and the cutting energy consumption decreases. As the cutting distance further increases, there is no interference between the cutting grooves. The weakening effect of the existing cutting grooves on the rock decreases. The cutting force increases, the coarseness decreases, and the cutting energy consumption increases. The cutting state gradually approaches free cutting.

2023 Vol. 49, No. 8

Submission System

Latest AcceptedMore>