工矿自动化

Research on denoising of uneven lighting images in coal mine underground

ZHANG Xuhui, MA Bing, YANG Wenjuan, DONG Zheng, LI Yuyang

2024, 50(2): 1-8. doi: 10.13272/j.issn.1671-251x.2023110090

<Abstract>(835) <HTML> (66) <PDF>(71)

Abstract:
The space of the fully mechanized working face is small, and the lighting environment is complex and variable. During the coal mining process, there is a large amount of dust and fog, which leads to problems such as exposure and weakened detail features in the collected images. It is difficult to effectively extract features from images with excessive lighting intensity in the underground lighting area. In order to solve the above problems, a denoising algorithm for uneven lighting images in coal mines is proposed. Firstly, the video is captured as an image to determine whether lighting suppression is necessary. The RGB image that requires lighting suppression is split into channels, and the lighting adjustment factor for each channel is calculated to achieve overall lighting adjustment of the image. Secondly, the images that have not undergone overall lighting suppression and those that have undergone overall lighting suppression are subjected to reflection component extraction. The input image is converted into an HSV spatial image, and the single scale Retinex (SSR) algorithm is used to separately process the illumination component in the V channel image. The incident component in the V component is removed, while the reflection component is retained. The histogram equalization algorithm is used to achieve illumination equalization for the reflection component. Finally, a dark channel prior algorithm with guided filtering is used to defog the light-processed image. The gamma correction function is used to readjust the image with uneven brightness. The subjective evaluation results indicate that the proposed denoising algorithm for uneven lighting images in coal mines effectively suppresses the problem of high overall brightness caused by lighting. The blurry parts of the original image are clearer due to factors such as fog and dust, and the detailed features of the image are more prominent. The effectiveness of the proposed algorithm is objectively evaluated using four evaluation indicators: information entropy, mean, standard deviation, and spatial frequency. The results showed that the proposed algorithm has achieved an average improvement of 21.87%, −56.06%, 153.43%, and 294.45% in terms of information entropy, mean, standard deviation, and spatial frequency compared to the multi-scale Retinex (MSR) algorithm. The proposed algorithm has achieved an average improvement of 1.18%, −39.56%, 33.29%, and −4.71% compared to the multi-scale Retinex with color preservation (MSRCR) algorithm. The proposed algorithm has achieved an average improvement of 38.06%, −55.27%, 462.10%, and 300.96% compared to the multi-scale Retinex with color restoration (MSRCR) algorithm. The results indicate that the proposed algorithm can more effectively increase image information, suppress lighting intensity, improve edge information, and image clarity.

Quantitative analysis of coal particle size based on bi-level routing attention mechanism

CHENG Deqiang, ZHENG Lijuan, LIU Jingjing, KOU Qiqi, JIANG He

2024, 50(2): 9-17. doi: 10.13272/j.issn.1671-251x.2023100002

<Abstract>(177) <HTML> (65) <PDF>(29)

Abstract:
The distribution features of coal particle size are closely related to the analysis of methane gas propagation in coal. At present, the coal particle size analysis method based on image segmentation has become one of the mainstream solutions to obtain coal particle size. But there are problems such as loss of contextual information, improper fusion of coal particle features resulting in missed segmentation and over-segmentation of coal particles. In order to solve the above problems, a coal particle size analysis model based on bi-level routing attention (BRA) is designed. The BRA module is embedded in the residual U-shaped network ResNet-UNet to obtain the B-ResUNet network model. To reduce the problem of missed segmentation in coal particle segmentation, a BRA module is added before upsampling in the ResNet-UNet network. It allows the network to adjust the importance of the current feature layer based on the features of the previous layer, enhance the expression capability of features, and improve the transmission capability of long-distance information. To reduce the problem of over segmentation in coal particle segmentation, a BRA module is added after the feature concatenation module of the ResNet-UNet network. By dynamically selecting and aggregating important features, more effective feature fusion is achieved. The feature information from the segmented coal particles is extracted. The coal particle size of the coal particle dataset used in the experimental analysis is equivalent to the cell size. In order to accurately characterize the coal particle size, equivalent circular particle size is used to obtain the coal particle size and size distribution. The experimental results show the following points. ① The accuracy, average intersection to union ratio, and recall of the B-ResUNet network model have been improved by 06.%, 14.3%, and 35.9% compared to the ResNet-UNet basic network, with an accuracy of 99.6%, an average intersection to union ratio of 92.6%, and a recall of 94.4%. The B-ResUNet network model has good segmentation performance in coal samples and can detect relatively complete particle structures. ② When the BRA module is introduced before upsampling and after feature concatenation, the network pays sufficient attention to the edge areas of coal particles and reduces attention to some less important areas, thereby improving the computational efficiency of the network. ③ The particle size of coal particles shows a relatively balanced distribution trend within 1-2 mm, with the maximum proportion of coal particles within 1-2 mm being 99.04% and the minimum being 90.59%. It indicates that the image processing method based on BRA has high accuracy in particle size analysis.

Path planning algorithm for tracked directional drilling rigs in coal mines

MAO Qinghua, YAO Lijie, XUE Xusheng

2024, 50(2): 18-27. doi: 10.13272/j.issn.1671-251x.2023080085

<Abstract>(204) <HTML> (56) <PDF>(36)

Abstract:
In the process of path planning for tracked directional drilling rigs in coal mines, there are constraints on the body volume and the demand for driving efficiency in actual scenarios. However, the commonly used A^* algorithm has slow search speed, multiple redundant nodes, and the planned path is close to obstacles and has poor smoothness. This study proposes a path planning algorithm for coal mine tracked directional drilling rigs, which uses the improved A^* algorithm to plan global paths and integrates the dynamic window approach (DWA) to plan local paths. Considering the influence of directional drilling rig size, a safety extension strategy is introduced in the traditional A^* algorithm. The safety distance constraints are added between the directional drilling rig, roadway walls, and obstacles to improve the safety of the planned path. Adaptive weighting is applied to the heuristic function of the traditional A^* algorithm, while incorporating the influence of the parent node into the heuristic function to improve the efficiency of global path search. The principle of obstacle detection is used to eliminate redundant nodes in the path planning of the improved A^* algorithm. The segmented cubic Hermite interpolation is used for quadratic smoothing to obtain the global optimal path. The improved A^* algorithm is integrated with DWA for path planning of directional drilling rigs in coal mines. Matlab is used to simulate and do comparative analysis of directional drilling rig path planning algorithms under different working conditions.The results show that compared with Dijkstra algorithm and traditional A^* algorithm, the improved A^* algorithm accelerates the search speed while ensuring a safe distance. It reduces search time by 88.5% and 63.2% respectively, and to some extent shortens the length of the planned path, making the path smoother. The improved A^* algorithm and DWA fusion algorithm can effectively avoid unknown obstacles on the path planned by the improved A^* algorithm. The path length is reduced by 5.5% and 2.9% compared to the paths planned by the PRM algorithm and RRT * algorithm, respectively.

Design of coal mine gas intelligent extraction control system based on industrial Internet architecture

YIN Jianhui

2024, 50(2): 28-34. doi: 10.13272/j.issn.1671-251x.2023080030

<Abstract>(252) <HTML> (75) <PDF>(53)

Abstract:
There are currently problems with the intelligent gas extraction system in coal mines. ① The system's functions are limited to a certain process control, resulting in incomplete coverage of gas extraction business management and inadequate implementation of measures. ② Based on the traditional "chimney style" IT architecture, subsystems are scattered, data utilization is low, collaborative capabilities are poor. The cost of later subsystem integration is high, making system expansion inconvenient. ③ There are still many manual links in the gas extraction process, and the system's intelligence and automation capabilities still need to be further improved. In order to solve the above problems, a coal mine gas intelligent extraction control system based on industrial Internet architecture is designed. A multi-source heterogeneous data collection process for gas extraction has been developed based on the publish/subscribe model. It promotes data decoupling and sharing, reduces system complexity, and achieves unified collection of multi-source heterogeneous data such as gas extraction pipeline network data, drilling operation and trajectory data, equipment working condition data, and meeting standard evaluation data. Based on digital twin technology, a three-dimensional extraction system model is constructed, achieving a three-dimensional display of the up and down extraction system. Based on rule engine technology, sensor data processed by the message center can be judged based on subscribed Topic. The processed sensor data is stored in the database to realize the automation and process operation of gas extraction standard evaluation. By using machine vision video analysis technology to identify the number of drill pipes, automatic counting of drill pipe numbers (drilling depth) during the drilling process and information management of drilling engineering are achieved. Combined with drilling measurement instruments, analysis and visualization of drilling trajectory left and right, upper and lower deviations are achieved. The on-site application results show that gas extraction management personnel can quickly and in real time understand the extraction situation, extraction evaluation, drilling engineering construction, and system failure of each extraction face by viewing the intelligent control system software for gas extraction. It improves the level of gas extraction information and intelligent management.

A method for extracting axis and constructing section in long roadway deformation monitoring

CHEN Xiaowei, CHEN Lei, LI Meng, HU Chengjun, SONG Lei, YUAN Pengzhe

2024, 50(2): 35-41. doi: 10.13272/j.issn.1671-251x.2023090077

<Abstract>(152) <HTML> (48) <PDF>(9)

Abstract:
The 3D laser scanning technology is widely used in the research of deformation monitoring technology for long roadways. But there is a phenomenon of benchmark point displacement in the point cloud data collected through multiple scans in current research. The common features of adjacent point cloud data collected are not obvious, and the splicing of multi site clouds will lead to an increase in cumulative errors. The deformation of advanced roadways is affected by advanced supports. In order to solve the above problems, based on the traditional cross point method, which involves the intersection of the midpoint of the roof and floor and the two key points of the two sides, a method for extracting the axis of the roadway based on the least squares method is proposed. The origin of the rectangular coordinate system defined by the roadway is located at the laser beam emission point. The z-axis is located within the vertical scanning plane of the laser scanner. The x-axis and y-axis are located within the horizontal scanning plane of the scanner. The central axis reflects the overall direction and position and posture of the roadway. When the roadway excavation is completed without being affected by mining, the entire roadway is scanned for the first time, and the center point of the entire roadway is determined by the least squares method. Each center point is connected and fitted to form a complete central axis. In the subsequent monitoring of roadway deformation, point cloud data is superimposed based on the midpoint position of the first monitoring to accurately obtain the changes in various point clouds within the roadway cross-section, and thus obtain the deformation of the roadway. And the roadway section is constructed based on the fitted central axis. A 3D laser scanning system is used to test the deformation of the 30507 working face return air roadway in Tashan Coal Mine. The results showed the following points. ① The deformation of the roadway decreases with the increase of the distance from the measuring point to the working face, and the leading influence range of the 30507 working face return air roadway is 150 meters. The maximum point of roadway deformation is located on the side near the goaf of the floor. ② The advanced range determined by the 3D laser scanning and microseismic monitoring system is close, indicating that the supporting coal body has started to be under stress when entering 150 meters. The maximum point of roadway deformation is located on the side near the goaf of the floor, rather than the floor observed by the cross point observation method. This proves that the 3D laser scanning results are more accurate and greatly reduces the intensity of the operation.

Research on flow compensation technology for hydraulic system in working face

ZHAO Shuji

2024, 50(2): 42-48. doi: 10.13272/j.issn.1671-251x.2023080060

<Abstract>(128) <HTML> (51) <PDF>(20)

Abstract:
Currently, there is a lack of analysis on the pressure and flow characteristics of hydraulic systems during continuous propulsion in the optimization research of working face hydraulic systems. There is a lack of simple and effective solutions to the problem of pressure and flow fluctuations in hydraulic systems. In response to the demand for rapid support movement of the working face, with the construction of a new large mining height working face in the 2⁻² coal seam of Zhangjiamao Coal Mine as the engineering background, a single hydraulic support and a group hydraulic support simulation model are established using AMEsim software. Based on the action timing of the jack in the automatic follow-up and support movement, the hydraulic support movement and sliding process during coal mining are simulated. The study analyzes the pressure and flow changes of the hydraulic system in the working face when different numbers of hydraulic supports act simultaneously. It is pointed out that the reason for the slow movement of the supports is that the instantaneous liquid demand of the hydraulic supports exceeds the maximum flow rate of the pump station. At the same time, there is a contradiction between insufficient instantaneous liquid demand and excess liquid supply capacity of the pump station at some times during the movement of the hydraulic supports. Aiming at the intermittent high flow demand of hydraulic systems, a flow compensation technology based on accumulators is proposed. Through simulation verification, the pressure fluctuation of the hydraulic system is significantly suppressed after the installation of accumulators, and the movement speed of each jack is significantly improved. On site experiments are conducted on the flow compensation technology based on accumulator in the new construction working face of Zhangjiamao Coal Mine. The results show that after the accumulator is connected, the average pressure drop of the hydraulic system decreases by 74.1%, and the pressure fluctuation is significantly suppressed. This verifies that the flow compensation technology can meet the intermittent high flow demand of the hydraulic system and provide guarantees for rapid support movement.

Research on coal gangue recognition method based on CED-YOLOv5s model

HE Kai, CHENG Gang, WANG Xi, GE Qingnan, ZHANG Hui, ZHAO Dongyang

2024, 50(2): 49-56, 82. doi: 10.13272/j.issn.1671-251x.2023090065

<Abstract>(619) <HTML> (61) <PDF>(55)

Abstract:
Due to the complex working conditions of high noise, low illumination, and blurred movement in coal mines underground, as well as the phenomenon of coal gangue easily gathering, it is difficult to extract features from coal gangue object detection models. The classification and positioning of coal gangue are inaccurate. In order to solve the above problems, a coal gangue recognition method based on the CED-YOLOv5s model is proposed. Firstly, the coordinate attention (CA) mechanism is introduced into the YOLOv5s backbone network, which encodes feature maps by embedding coordinate information into channel relationships and long-range dependencies. The method fully utilizes channel attention information and spatial attention information to make the model focus more on important features and suppress irrelevant information. Secondly, the EIoU regression loss function is introduced in the detection head of YOLOv5s to minimize the width and height difference between the object box and anchor box. It enhances the position and boundary information of the object, improves the positioning precision and convergence speed of the model in dense objects. Finally, a lightweight decoupling head is introduced in the detection head of YOLOv5s, decoupling separate feature channels for classification and regression tasks. It solves the interference problem between the coupling head part of the class task and the regression task in the original model, further improving the parallel operation efficiency and detection precision of the model. The experimental results show that the CED-YOLOv5s model has the best overall performance compared to other YOLO series object detection models. It has an average detection precision of 94.8%, an improvement of 3.1% compared to the YOLOv5s model, and a detection speed of 84.8 frames/s. The results can fully meet the real-time detection requirements of coal gangue in coal mines.

Large coal detection for belt conveyors based on improved YOLOv5

QIN Yulong, CHENG Jiming, REN Yige, WANG Xiaoqing, ZHAO Qing, AN Cuijuan

2024, 50(2): 57-62, 71. doi: 10.13272/j.issn.1671-251x.2023080096

<Abstract>(680) <HTML> (54) <PDF>(58)

Abstract:
Oversized coal blocks can easily cause poor coal flow, blockage, and coal stacking when transported on a belt conveyor. However, the differences in appearance and color between large coal blocks and ordinary coal blocks are small, and there are obstructions and stacking between coal blocks. Existing coal block detection methods are not precise enough to distinguish between large coal blocks and ordinary coal blocks, which can easily lead to missed or false detections. In order to solve the above problems, a modified YOLOv5 model is proposed for detecting large coal blocks in belt conveyors. The model uses parallel dilated convolution modules to replace some ordinary convolution modules in the YOLOv5 backbone network. It expands the receptive field, improves multi-scale feature learning capability, and better distinguishes large coal blocks from ordinary coal blocks. The joint attention module is added to the neck network to better integrate contextual information and improve the positioning capability for large coal blocks. The model uses the trained improved YOLOv5 model to detect real-time coal transportation videos captured by the camera, and links PLC alarms in real-time based on the quantity information of large coal blocks. The experimental results show that compared to the original YOLOv5 model, the improved YOLOv5 model has improved recall and average precision by 3.4% and 2.0%, respectively. PLC can operate corresponding indicator lights and buzzers to alert based on the quantity of large coal blocks detected by the improved YOLOv5 model. The improved YOLOv5 model is applied to actual coal transportation videos in coal mines, with a detection precision of 97.0% for large coal blocks, effectively avoiding missed and false detections.

Super resolution reconstruction of noisy images based on dense residual connected U-shaped networks

LIU Pengnan, LI Long, ZHANG Zihao, ZHU Xingguang, CHENG Deqiang

2024, 50(2): 63-71. doi: 10.13272/j.issn.1671-251x.2023080098

<Abstract>(127) <HTML> (43) <PDF>(8)

Abstract:
The existing image super-resolution reconstruction networks are difficult to apply to noise intensive application scenarios in coal mines. Most networks improve performance by increasing depth, which leads to problems such as ineffective extraction of key features and loss of high-frequency information. In order to solve the above problems, a dense residual connected U-shaped network is proposed for super-resolution reconstruction of low resolution noisy images. The denoising module based on dense residual connections is introduced in the feature extraction path, fully extracting image features through dense connections. The features of residual learning are used to effectively denoise low resolution noisy images. The residual feature attention distillation module is introduced in the reconstruction path, by incorporating enhanced feature attention blocks into the residual blocks, different weights are assigned to features in different spaces to enhance the network's capability to extract key image features. The loss of image detail features is reduced in the residual blocks, thus better restoring image detail information. Comparative experiments are conducted on coal mine underground image datasets and public datasets, and the results show that in terms of objective evaluation index, structure similarity and image perception similarity of the proposed network are superior to the comparison network. It has a good balance in complexity and running speed. In terms of subjective visual effects, the image reconstructed by the proposed network basically eliminates the original image noise and effectively restores the detailed features of the image.

A line feature matching algorithm for mine images based on line segment detection and LT descriptors

ZHU Daixian, QIU Qiang, KONG Haoran, HU Qisheng, LIU Shulin

2024, 50(2): 72-82. doi: 10.13272/j.issn.1671-251x.2023090045

<Abstract>(114) <HTML> (33) <PDF>(12)

Abstract:
Image matching is an extremely important part of simultaneous localization and mapping (SLAM) technology. It is used to determine camera position and posture based on the transformation relationship between images. The image matching method based on line features has strong robustness and noise resistance, making it more suitable for underground image matching. The line descriptors based on deep learning have high robustness to scenes such as line segment occlusion, and their performance is better than traditional descriptors. However, the descriptors of convolutional neural network architecture abstract variable length line segments into fixed dimensions for description, which is not conducive to matching images with large changes in line segment length and parallax. In order to solve the above problems, a line feature matching algorithm for mine images based on line segment detection and line transformers (LT) is proposed. The algorithm uses single parameter homomorphic filtering in the frequency domain to reduce the lighting component of the image, enhance the reflection component, and improve brightness and contrast. The algorithm uses contrast limited adaptive histogram equalization (CLAHE) algorithm in YUV space to balance brightness components and make brightness distribution more even. The algorithm transforms to RGB space to extract line segment detection (LSD) lines. A LT descriptor based on Transformer architecture is introduced to construct the feature vector of LSD lines, and finally complete line feature matching. The experimental results show that the algorithm combines the advantages of homomorphic filtering and CLAHE algorithm. After image enhancement, the brightness of the image is moderate, the contrast is good, the grayscale distribution is even. The enhancement effect is better than the single parameter homomorphic filtering algorithm and EnlightenGAN algorithm. The number of line features extracted by this algorithm has increased by an average of 32.92% compared to the original image. It has good robustness in matching underground images with different proportions of similar textures, varying degrees of rotation and translation changes. The average correct matching number is 61.75 pairs, with an average precision of 86.83%. It is superior to the line binary descriptor (LBD) algorithm, LBD_NNDR algorithm, and LT algorithm. It can meet the requirements of robust matching of mine images.

Local feature-guided label smoothing and optimization for re-identification of underground personnel with weak features

ZHANG Jie, MIAO Xiaoran, ZHAO Zuopeng, HU Jianfeng, MIN Bingbing, GAO Yumeng

2024, 50(2): 83-89. doi: 10.13272/j.issn.1671-251x.2023080092

<Abstract>(112) <HTML> (37) <PDF>(7)

Abstract:
The low light, strong light disturbance, high dust and other environmental conditions underground in coal mines, as well as the similarity of clothing and coal falling on the face of underground personnel, make it difficult to re identify underground personnel with weak features. The existing personnel re identification methods only extract global features and do not fully consider local features, resulting in low accuracy of underground personnel re identification. In order to solve the above problems, a local feature guided label smoothing and optimization method for re-identification of underground personnel with weak features is proposed. This method first extracts global and local features of underground personnel images through convolutional neural networks. Secondly, the k-nearest neighbor similarity is used to calculate the complementarity score between global and local features, in order to measure the degree of similarity between global and local features. Finally, based on the score of feature complementarity, label smoothing is performed on local features and label optimization is performed on global features. The weight of each local feature is dynamically adjusted to improve the label of each local feature. The prediction results of local features are summarized. The more reliable information is used to improve the label as a global feature label, thereby reducing image noise and enhancing feature identification capability. The experimental results show that the method outperforms mainstream personnel re identification methods in terms of mean average precision (mAP), rank-1 accuracy (Rank-1), and mean inverse negative penalty (mINP) on both publicly available datasets and self built datasets containing images of underground personnel. It has good generalization and robustness, and can effectively achieve underground weak feature personnel re identification.

Gas concentration prediction model based on SSA-LSTM

LAN Yongqing, QIAO Yuandong, CHENG Hongming, LEI Lixing, LUO Huafeng

2024, 50(2): 90-97. doi: 10.13272/j.issn.1671-251x.2023090067

<Abstract>(494) <HTML> (41) <PDF>(33)

Abstract:
In order to better capture the time-varying patterns and effective information of gas concentration, and achieve precise prediction of gas concentration in coal working faces, a gas concentration prediction model based on SSA-LSTM is proposed by optimizing the long short term memory (LSTM) network using sparrow search algorithm (SSA). The model uses the mean replacement method to process missing and abnormal data in the original gas concentration time series data, followed by normalization and wavelet threshold denoising. The performance differences between SSA and grey wolf optimization (GWO) and particle swarm optimization (PSO) algorithms are compared and tested. The result verifies the advantages of SSA in terms of optimization precision, convergence speed, and adaptability. By utilizing the adaptability of SSA, the hyperparameters of LSTM, such as learning rate, number of hidden layer nodes, and regularization parameters, are sequentially optimized to improve the global optimization capability and avoid the prediction model falling into local optimum. The obtained optimal hyperparameter combination is substituted into the LSTM network model and the prediction results are output. Comparing SSA-LSTM with LSTM, GWO-LSTM, and PSO-LSTM gas concentration prediction models, the experimental results show that the root mean square error (RMSE) of the gas concentration prediction model based on SSA-LSTM is reduced by 77.8%, 58.9%, and 69.7% compared to LSTM, PSO-LSTM, and GWO-LSTM, respectively. The mean absolute error (MAE) decreases by 83.9%, 37.8%, and 70%, respectively. The LSTM prediction model optimized by SSA has higher prediction precision and robustness compared to traditional LSTM models.

Research on the recognition model of mine water inrush source based on improved SSA-BP neural network

LIU Weitao, LI Beibei, DU Yanhui, HAN Mengke, ZHAO Jiyuan

2024, 50(2): 98-105, 115. doi: 10.13272/j.issn.1671-251x.2023070101

<Abstract>(564) <HTML> (60) <PDF>(26)

Abstract:
The combination of machine learning and optimization algorithms has been widely applied in the recognition of mine water inrush sources. However, the data of water inrush samples is stochastic and the optimization algorithm is prone to getting stuck in local optima. Further research is needed to improve the model's generalization capability and jump out of local optima. In order to solve the above problems, an improved sparrow search algorithm (SSA) is proposed to optimize the BP neural network model for quantitative recognition of mine water inrush sources. Taking Yangcheng Coal Mine of Luneng Coal and Electricity Co., Ltd. as the research object, the hydrochemical characteristics of the coal mine water sample are analyzed through conventional ion concentration analysis and Piper three line diagram. It is preliminarily determined that the mine water comes from the Ordovician limestone aquifer and the three limestone aquifers. The Na⁺+K⁺ concentration, Ca²⁺ concentration, Mg²⁺ concentration, ${\mathrm{HCO}}_3^- $ concentration, ${\mathrm{SO}}^{2-}_4 $ concentration, Cl⁻ concentration, mineralization degree, total hardness, and pH value are determined as the recognition indicators for water inrush source. The mine water inrush source recognition model is established based on an improved SSA-BP neural network. Firstly, the SSA parameters are set. Sine chaotic mapping is introduced to evenly distribute the sparrow population. Secondly, the sparrow population is updated by calculating fitness values, and a random walk strategy is introduced to perturb the current optimal individual. If the termination condition is met, the optimal BP neural network weight and threshold are obtained. Finally, based on the constructed BP neural network, the recognition results are output. The research results indicate the following points. ① The improved SSA-BP model has an recognition accuracy of 95.6% in the training set and 100% in the testing set. ② The comparison results of the improved SSA-BP neural network model with the BP neural network model and SSA-BP neural network model show that the BP neural network model has a misjudgment rate of 5/18, the SSA-BP neural network model has a misjudgment rate of 2/18, and the improved SSA-BP neural network model has a misjudgment rate of 0. After 10 iterations, it tends to stabilize and has the smallest error difference from the set target. The initial fitness value is the best, and the recognition results have high credibility. ③ Five sets of mine water samples from Yangcheng Coal Mine are inputted into the trained model as input layer data. The main sources of mine water samples are the Ordovician limestone aquifer, the three limestone aquifers, and the Shanxi formation aquifer. The results of model recognition are mutually confirmed with the conclusions of hydrochemical characteristic analysis, and precise segmentation is achieved.

Research on dust settlement under mixed air flow control in fully mechanized excavation face

GONG Xiaoyan, WANG Tianshu, CHEN Long, PEI Xiaoze, LI Xiangbin, ZHU Qianli, NIU Huming

2024, 50(2): 106-115. doi: 10.13272/j.issn.1671-251x.2023090022

<Abstract>(108) <HTML> (40) <PDF>(11)

Abstract:
Dust accumulation is severe during coal mine excavation. Currently, research on the dust settlement law and optimization under mixed air flow control in fully mechanized excavation faces is not in-depth enough. Based on a hybrid air flow control system and relying on the fully mechanized excavation face of Shaanxi Coal Group Shenmu Ningtiaota Mining Co., Ltd., the influence of mixed air flow control parameters such as the distance from the pressure air outlet to the working face, the right angle of the pressure air outlet, the pressure air outlet diameter, the distance from the extraction air outlet to the working face, and the pressure extraction ratio on the dust settlement law is analyzed. As the distance between the pressure air outlet and the working face increases, the proportion of large particle dust in the cross-section of the personnel breathing zone on the return air sides and the driver's location first increases, then decreases, and then increases again. The proportion of small particle dust increases. As the right deviation angle of the air inlet increases, the proportion of large particle dust in the personnel breathing zone section on the return air sides and the driver's location changes significantly. As the diameter of the air inlet increases, the proportion of small particle dust in the driver's location section first increases, then decreases, and then increases again. The proportion of large particle dust in the personnel breathing zone section on the return air side first increases and then decreases. As the distance between the extraction air outlet and the working face increases, the proportion of large particle dust at the driver's location section first increases and then decreases. The proportion of small particle dust first increases and then decreases and then increases again. The particle size distribution of dust at the personnel breathing zone section on the return air side does not change much. As the pressure-pumping ratio increases, the proportion of small particle dust in the cross-section of the personnel breathing zone on return air sides and the driver's location decreases. Taking the above air flow control parameters as independent variables, the average concentration of total dust in the personnel breathing zone on the return air side and the average concentration of exhaled dust at the driver's location are the optimization objectives. A dust settlement optimization regression model is established, and the particle swarm optimization algorithm is used to solve the model. The optimal air flow control scheme is obtained. The distance between the pressure air outlet and the working face is 8.9 meters, the right angle of the compressed air outlet is 14.8°, the diameter of the compressed air outlet is 0.9 meters, the distance between the extraction air outlet and the working face is 4.3 meters, and the pressure-pumping ratio is 1.1. The experimental platform for dust settlement under wind flow control is built. The experimental results show that the error between the test values and the simulated values of the dust settlement optimization regression model is within 13%, which verifies the accuracy of the model. The optimized dust with particle sizes of 71-100 μm is significantly affected by the wind flow regulation parameters and settles in front of the roadheader. After optimization, the average dust concentration of total dust in the personnel breathing zone on the return air side and the average dust concentration at the driver's location decrease by 47.4% and 42.4%, respectively, indicating a significant dust reduction effect.

Prediction of height adjustment of shearer drum based on improved gated recurrent neural network

QI Ailing, WANG Yu, MA Hongwei

2024, 50(2): 116-123. doi: 10.13272/j.issn.1671-251x.2023110039

<Abstract>(130) <HTML> (38) <PDF>(15)

Abstract:
The adaptive cutting technology of the shearer is a key technology for achieving intelligent mining in fully mechanized working faces. In order to solve the problem of low automatic cutting precision of the shearer in complex coal seams, a prediction method for the height adjustment of shearer drum based on improved gated recurrent neural network (GRU) is proposed. Considering the correlation between adjacent data in the longitudinal and transverse directions of the cutting trajectory, the fixed length sliding time window method is used to obtain the height data of the shearer drum. The input data is divided into continuous and adjustable subsequences, while processing the feature information in the transverse and longitudinal directions. To improve the prediction efficiency of the model and meet the real-time requirements of cyclic cutting, causal convolution gated recurrent unit(CC-GRU) is proposed to perform dual feature extraction and dual data filtering on input data. CC-GRU utilizes causal convolution to focus on the local temporal features in the longitudinal direction of the sequence in advance, in order to reduce computational costs and improve computational speed. CC-GRU uses gating mechanism to serialize and model the features obtained from convolution to capture long-term dependencies between elements. The experimental results show that using the CC-GRU model to predict the height adjustment of the shearer drum, the MAE is 43.80 mm, MAPE is 1.90%, RMSE is 50.35 mm, the determination coefficient is 0.65, and the prediction time is only 0.17 seconds. Compared to long short term memory (LSTM) neural networks, GRU, and temporal convolutional network (TCN), the CC-GRU model has a faster prediction speed and higher prediction precision. It can more accurately predict the height adjustment trajectory of the shearer in real time. This provides a basis for the establishment of coal seam models in the working face and the prediction of the height adjustment trajectory of the shearer.

Optimization strategy for multi-level relay drainage system in mines under time of use electricity price

ZHAO Yinghua, QIAO Zilong, WANG Yanbo, WU Qiang, HAN Yu, WANG Lei, WANG Lian

2024, 50(2): 124-129. doi: 10.13272/j.issn.1671-251x.2023080064

<Abstract>(96) <HTML> (29) <PDF>(15)

Abstract:
The efficiency of the underground drainage system in coal mines directly affects the production safety and economic benefits of coal mines. The existing multi-level relay drainage system in mines does not fully consider the peak and valley features of electricity bills and the safety constraints of the drainage system required by the Coal Mine Water Prevention and Control Regulations. It is difficult to achieve integrated safe and economic operation of the entire system. In order to solve the above problems, based on the avoiding peaks and filling valley strategy and dynamic programming method, an optimization strategy for multi-level relay drainage system in mines under the time of use electricity price is proposed. By considering the multi-level series structure, water inflow, and drainage capacity of water pumps, a mathematical model of a multi-level relay drainage system in coal mines is established. Based on the strategy of avoiding peaks and filling valleys, with the lowest electricity cost as the objective function and constraints such as water level in water tanks, drainage capacity of water pumps, and coal mine safety requirements, a multi-level relay drainage system optimization problem based on time of use electricity price is constructed. The solution algorithm based on dynamic programming method is provided. Taking the 4-level drainage system of a certain mine as an example for simulation analysis, the results show that this strategy can effectively control the underground water level and ensure that the water level is at a reasonable height. When the electricity price is high, the number of drainage pumps opened is very small or zero, and the water tank is in a high water level state. When the electricity price is low, the number of drainage pumps opened is larger, and the water tank is in a low water level state. This strategy can improve economic benefits while ensuring the production efficiency and safety of coal mines.

Research on fault positioning of underground power cable

SHANG Liqun, ZHANG Shaoqiang, RONG Xiang, LIU Jiangshan, WANG Yue

2024, 50(2): 130-137. doi: 10.13272/j.issn.1671-251x.2023080014

<Abstract>(127) <HTML> (22) <PDF>(16)

Abstract:
The traditional underground power cable fault positioning method relies on subjective parameter selection and noise resistance is poor. It cannot meet the accurate fault positioning requirements of underground power cable under strong noise background. In order to solve the above problems, a fault positioning method of underground power cable based on salp swarm algorithm (SSA) optimizing variational mode decomposition (VMD) combined with novel Teager energy operator (NTEO) is proposed. In response to the problem of modal aliasing, over decomposition, and under decomposition in signal decomposition of VMD, SSA is used to optimize the modal number K and penalty factor α of VMD parameters using fuzzy entropy as the fitness function. The intrinsic modal function that better reflects the fault feature information is obtained. NTEO is used to calibrate the first wave head to obtain the arrival time of the wave heads at both ends. The fault position is determined based on the dual end distance measurement method. PSCAD/EMTDC is used for underground power cable fault simulation. It simulates underground fault signals with strong background noise. The results show the following points. ① After adding 9 dB and 12 dB noise to the ideal current signal, the signal-to-noise ratio of SSA-VMD is the lowest, and the Pearson correlation coefficient is the highest. It indicates that SSA-VMD can effectively preserve the characteristic information of the signal while minimizing noise. ② Under different transition resistances, the positioning precision of SSA-VMD-NTEO is relatively high. ③ Under different fault phase angles, although SSA-VMD-NTEO may have different sampling points, the positioning position remains unchanged and still maintains high precision. ④ SSA−VMD-NTEO can ensure high positioning precision at different fault distances. ⑤ Under high underground noise and a sampling frequency of 10 MHz, SSA-VMD-NTEO has significant advantages in positioning precision compared to wavelet modulus maximum and VMD+NTEO methods.

Health status evaluation of CNN-GRU mine motor based on adaptive multi-scale attention mechanism

TAN Donggui, YUAN Yiping, FAN Panpan

2024, 50(2): 138-146. doi: 10.13272/j.issn.1671-251x.2023110024

<Abstract>(478) <HTML> (32) <PDF>(33)

Abstract:
When using multi-sensor information fusion technology to evaluate the health status of motors, there are outliers and missing values in the monitoring data of mine motors. However, deep learning models such as convolutional neural networks and recurrent neural networks find it difficult to effectively extract data features and update network weights when the data quality is severely degraded, resulting in problems such as vanishing or exploding gradients. In order to solve the above problems, A CNN-GRU (CNN-GRU-AMSA) model based on adaptive multi-scale attention mechanism is proposed to evaluate the health status of mine motors. Firstly, the model fills in, removes, and standardizes the motor operation data collected by sensors, and classifies the operating conditions of mine motors based on environmental temperature changes. Secondly, based on the Mahalanobis distance, the health index (HI) of health evaluation indicators such as motor current, three-phase temperature of motor winding, front bearing temperature of motor, and rear bearing temperature of motor are calculated. The Savitzky Golay filter is used to denoise, smooth, and normalize the HI indicator. Combining the contribution of different indicators calculated by principal component analysis method to mine motors, the weighted fusion of indicator HI is used to obtain the mine motor HI. Finally, the mine motor HI is input into the CNN-GRU-AMSA model, which dynamically adjusts attention weights to achieve information fusion of features at different scales, thereby accurately outputting the health status evaluation results of the motor. The experimental results show that compared with other common deep learning models such as CNN, CNN-GRU, CNN-LSTM, and CNN-LSTM Attention, the CNN-GRU-AMSA model performs better in evaluation metrics such as root mean square error, mean absolute error, accuracy, Macro F1, and MicroF1. The model has a smaller fluctuation range and better stability in predicting residuals.

Design of coal mine emergency rescue auxiliary decision system based on emergency plan

GAO Hongbo

2024, 50(2): 147-152, 160. doi: 10.13272/j.issn.1671-251x.2023090033

<Abstract>(186) <HTML> (47) <PDF>(28)

Abstract:
In the coal mine emergency rescue auxiliary decision system, there are problems such as insufficient application of emergency plans, low application efficiency, and poor execution of rescue plans generated by the system. In order to solve the above problems, a design method for a coal mine emergency rescue auxiliary decision system based on emergency plans is proposed. This method uses information extraction technology based on large language models to extract key task elements from emergency plans, such as task names, triggering conditions, executing departments, and task content. This method forms meta tasks, and constructs a meta task library that classifies and stores meta tasks based on accident types and levels. When a coal mine safety accident occurs, this method uses semantic matching technology based on the SBERT model to classify and grade the accident based on the information collected on site. The method selects the meta task set that matches the current emergency needs from the meta task library. To improve the feasibility of tasks, this method combines meta tasks with real-time collected on-site data, constructs specific action instructions through instruction templates. The method uses task planning techniques to optimize and adjust the priority of instructions, and generate practical and feasible on-site rescue plans. The coal mine emergency rescue auxiliary decision system based on emergency plans fully utilizes the standardized content of emergency plans, forming a rescue plan closely integrated with on-site information and resource optimization. The system further improves the accuracy, scientificity, and intelligence level of rescue decision-making.

Research on weighting strategies for safety status evaluation indicators in coal mine working faces

WANG Meng, LIU Shulin

2024, 50(2): 153-160. doi: 10.13272/j.issn.1671-251x.18148

<Abstract>(143) <HTML> (60) <PDF>(15)

Abstract:
Accurate evaluation of the safety status of the working face can promote the improvement of mine safety management level and disaster prevention and resilience. Using CH₄ concentration, CO₂ concentration, CO concentration, O₂ concentration, temperature, and wind speed as evaluation indicators, the safety status of the working face is evaluated and analyzed. To reasonably determine the weight of evaluation indicators and improve the accuracy of safety evaluation results, the fuzzy analytical hierarchy process(FAHP) is used to calculate the subjective weight of evaluation indicators, and the G-GRITIC method is used to calculate the objective weight of indicators. The combination weighting method based on improved game theory (IGT) combines subjective weight with objective weight to obtain the combination weight of evaluation indicators, solving the problem of inconsistent subjective and objective information in the decision-making process. Based on the data collected by the safety monitoring system of the 209 fully mechanized working face of Shaanxi Huangling No.2 Coal Mine Co., Ltd., experimental verification is conducted on the IGT based combination weighting method. The results show that this method effectively avoids the subjective judgment of linear weighting method and average weighting method, optimizes the deviation results of the game theory (GT) combination weighting method. It obtains more reasonable evaluation indicators, which can obtain more accurate evaluation results of the safety status of the coal mine working face.

2024 Vol. 50, No. 2

Submission System

Latest AcceptedMore>