工矿自动化

Research on mine electric spark recognition and alarm method based on the sum of adjacent frame pixel grayscale of images

SUN Jiping, LI Xiaowei, WANG Jianye

2023, 49(7): 1-5. doi: 10.13272/j.issn.1671-251x.18141

<Abstract>(196) <HTML> (49) <PDF>(31)

Abstract:
Early detection of mine electric sparks and alarm can prevent or reduce gas and coal dust explosions and mine fire accidents. There are no natural light sources such as sunlight, moonlight, and starlight underground. The main factor affecting the recognition of mine electric sparks is the mine light source. By adjusting the installation position and angle of the camera, the impact of fixed mine light sources on electric spark recognition can be avoided or reduced. But it cannot solve the impact of mobile mine light sources on electric spark recognition. The discharge cycle of electric sparks generated by different forms of circuits is different, but the discharge time of electric sparks is less than 4 ms. The minimum bright duration of the flash light source is 240 ms. Therefore, the features of the short emission time of electric sparks and longer exposure time of mine moving light sources to cameras can be utilized to eliminate the impact of mine moving light sources on camera exposure. The camera shoots at a high frame rate, and the electric spark image has a feature of 1 frame dark -1 frame bright -1 frame dark, that is, a "dark light dark" frame feature. The "bright" frames with sparks have a large sum of pixel grayscales in a single frame. The "dark" frames without sparks have a small sum of pixel grayscales in a single frame. The illumination of a moving light source on the camera is variable, going through a process of no light, light, and no light. In the absence of electric sparks, the camera also shoots at a high frame rate. The images of both moving constant bright light sources and moving flashing light sources do not exhibit the "dark bright dark" frame feature. Based on the unique "dark bright dark" frame feature of electric spark images, a mine electric spark recognition and alarm method based on the sum of adjacent frame pixel grayscale is proposed. The method collects monitoring area video images in real time. According to the set frame rate, the method preprocesses the video image into frames and calculates the pixel grayscale of a single frame image separately. If the difference between the current frame image pixel grayscale and the previous frame image pixel grayscale is less than the pre-set threshold, the method continues to collect the video image. Otherwise, the method calculates the difference between the current frame image pixel grayscale and the subsequent frame image pixel grayscale. If the difference is less than the pre-set threshold, the method continues to collect video images. Otherwise, the method issues a mine electric spark alarm signal. After the mine electric spark alarm, if the emergency response is not activated manually, the mine electric spark alarm will continue. Otherwise, the method exits the current alarm state and continues to collect video images. This method can effectively eliminate the interference of moving constant light sources and flashing light sources.

Research progress and challenges faced by unmanned aerial vehicles in complex underground spaces

WANG Baobing, WANG Kai, WANG Dandan, GAO Haiyue, WANG Chunxi

2023, 49(7): 6-13, 48. doi: 10.13272/j.issn.1671-251x.2022100078

<Abstract>(205) <HTML> (37) <PDF>(49)

Abstract:
The technological development and application status of underground complex space UAVs are analyzed. It is pointed out that underground complex space UAVs face problems such as insufficient individual performance, limited environmental situational awareness and autonomous navigation capabilities, and limited formation collaboration capabilities. In order to solve the above problems, the development trends of key technologies for underground UAVs are prospected. ① Small and lightweight integrated UAV design technology is proposed. By improving the mechanical structure of the UAV, improving the integration of information perception sensors such as LiDAR and depth camera with control systems, and optimizing power management systems, the ultimate goal is to improve the cruise speed, endurance time, and other performance of individual UAV. ② Situation awareness and autonomous navigation technology in GPS rejection environment is proposed. The key technical challenges such as simultaneous localization and mapping (SLAM) navigation and real-time path planning should be overcome. The limitations of algorithms around specific scenarios should be gradually broken through. The perception capability, environmental adaptability, and robustness of unmanned systems should be improved. ③ Formation collaboration control technology under limited information is proposed. The technical problems such as heterogeneous/isomorphic UAV cluster collaboration, and wireless communication in complex channel environments should be overcome. By optimizing UAV swarm intelligence control strategies, information interaction mechanisms, and task decision-making collaboration mechanisms, the robustness of clustered unmanned systems should be enhanced. The adaptability of unmanned systems in complex underground environments should be improved. Furthermore, the task execution efficiency and success rate of unmanned systems should be improved.

Wireless communication technology evolution in underground coal mines

BAI Xuefeng, SHU Xiaojun

2023, 49(7): 14-18. doi: 10.13272/j.issn.1671-251x.2023020012

<Abstract>(154) <HTML> (439) <PDF>(74)

Abstract:
Mine communication is an indispensable part of the intelligent development of coal mines. This paper analyzes the advantages and disadvantages of different communication systems and technologies in different development stages of wireless communication in underground coal mines. It is concluded that 5G and WiFi6 currently have significant technological advantages and good applicability in terms of transmission rate, delay, reliability, and user capacity. They are the mainstream solutions for mine wireless communication technology in recent years. This paper discusses the problems in underground wireless communication system construction and the corresponding development directions. ① Technological convergence issues. Different scenarios have different requirements for bandwidth, delay, power consumption and reliability. A single communication technology can not meet the requirements of all application scenarios in coal mines. It is necessary to select appropriate communication technology according to the actual application situation to meet the application requirements. Different technologies are integrated through protocol converters, gateways and other devices to form an integrated communication network platform for technological convergence above the ground and underground. ② System integration issues. Since each system is provided by different manufacturers, and the communication protocol and hardware interface are inconsistent, it is difficult to truly realize the data sharing of each system in the mine. It is necessary to provide standard interfaces, open network architecture, and unified user data to provide interconnection and inter-working functions for monitoring, emergency, production, and other systems. It is further required to provide unified data messages to enable data transmission with low delay, high concurrency, real-time, and synchronization within each system. ③ Device power consumption issues. The existing mine 5G base station and other equipment can only be made into a single flameproof product due to their high power consumption. The equipment is bulky, large, and difficult to install and maintain, which restricts the popularization of 5G technology in coal mine. Low-power design can be started from the main chip, RF module, and power amplifier module.

Research on coal flow measurement based on binocular structured light vision

ZHANG Junsheng, WANG Honglei, LI Jiacheng

2023, 49(7): 19-26. doi: 10.13272/j.issn.1671-251x.2022100050

<Abstract>(165) <HTML> (84) <PDF>(27)

Abstract:
In the conventional binocular vision system, the commonly used speeded up robust features and scale-invariant feature transform matching algorithms have high requirements for image quality. When applied to scenes with relatively single color and texture such as coal, it is prone to failure. It needs to consume a lot of computing resources, which is difficult to ensure real-time performance. When using LiDAR for coal quantity measurement, the effective field of view is relatively small. The corresponding measurement points are few and the scanning frequency is low. When the belt conveyor runs at a faster speed, the precision will be significantly reduced. In order to solve the above problems, a coal flow measurement method based on binocular structured light vision is proposed. The linear structured light is introduced into the binocular vision system. By using the constraint of linear structured light, the image feature point matching is simplified into matching between left and right image lines. On the basis of ensuring the parallelism of the optical axis of the binocular system camera, corresponding row matching is used to calculate three-dimensional coordinate points. The sampling frequency and resolution is improved. The precision of coal flow measurement is improved. The dependence of the measurement system on lighting and environment is reduced. Point cloud acquisition: It uses the line structured light to highlight the coal material section curve, and extracts the image coordinates of the coal material section center line. It uses the binocular camera to obtain the left and right coal material section line structured light images. It establishes binocular structured light 3D reconstruction model. The left and right image center line coordinates form a matching point pair to participate in the calculation of the coal material section 3D coordinates, so as to achieve real-time acquisition of point clouds. Coal flow calculation: The point cloud of coal material is obtained by combining the point cloud of no-load belt section and the point cloud of loaded belt section. The infinitesimal method is used to sample the 3D point cloud of coal material. The volume of coal material in unit time is calculated by the uniform meshing method and the triangle meshing method, respectively. The coal flow measurement of belt conveyor is realized. The experimental results show that the average relative error of coal volume measured by the uniform meshing method is 6.758%. The average relative error of coal volume measured by the triangle meshing method is 2.791%. The measurement precision of the triangle meshing method is higher than that of the uniform meshing method. The industrial test results show that compared with the electronic belt weigher, the maximum absolute error of the coal flow measurement method based on binocular structured light vision is 87.855 t/h. The average absolute error is 25.902 t/h, the maximum relative error is 2.876%, and the average relative error is 0.847%. The results meet the requirements of non-contact coal flow measurement in coal mines.

Coal flow detection system for belt conveyor based on dual lidar

YU Haili, SUN Lichao, ZUO Sheng, CHEN Dawei, ZENG Xiangyu, DU Yuanjiang

2023, 49(7): 27-34, 59. doi: 10.13272/j.issn.1671-251x.2022120004

<Abstract>(282) <HTML> (45) <PDF>(44)

Abstract:
Due to the presence of stacking angles during the transportation of coal flow by belt conveyors, the shape of the coal flow is approximately triangular. It can easily lead to blind spots in detection. In order to solve this problem, a coal flow detection system for belt conveyors based on dual lidar is proposed. The method places two single-line lidars on the left and right sides above the belt conveyor, and measures the outer contour feature points of the coal flow in each half of the area. The method uses the fusion algorithm to fuse the outer contour feature points of the coal flow in the left and right areas. Then the method uses the least squares polynomial fitting algorithm to calculate the outer contour of the coal flow in the entire area, thus achieving blind spot-free measurement of the coal flow contour. The method uses the photoelectric encoder to achieve the real-time detection of the conveyor belt running speed. The method uses the trapezoidal area accumulation method to calculate the coal flow cross-sectional area. The method uses the panel integration method to calculate of the coal flow rate of the belt conveyor. The on-site test results show that when there is no coal bias, the scanning results of single/dual lidar are basically consistent, and the system measurement error is 2%~3%. The results meet the requirements of coal flow detection. When there is coal bias, the system error based on a single lidar is relatively large. The result cannot meet the requirements of coal flow detection. The measurement error based on a dual lidar system can still be maintained at 2% to 3%. The paper proposes a selection criterion for single/dual lidar. It is concluded that the coal flow detection system based on dual lidar is more suitable for belt conveyors in the presence of coal bias or large blocks of coal.

Coal flow volume measurement based on linear model partitioning

CHEN Xiangyuan, XUE Xusheng

2023, 49(7): 35-40, 106. doi: 10.13272/j.issn.1671-251x.2022090085

<Abstract>(96) <HTML> (27) <PDF>(20)

Abstract:
The precision and computational efficiency of coal quantity measurement for belt conveyors based on linear laser stripes are low. There is a trailing phenomenon during belt operation, as well as data misalignment caused by deviation and drifting. In order to solve the above problems, a coal flow volume measurement method based on linear model partitioning is proposed. Firstly, the method uses a high-speed line laser camera to collect coal flow data. Secondly, a point cloud registration algorithm based on linear model partitioning is used to fuse the point cloud data at the bottom of the belt with the surface data of the coal flow, forming a complete three-dimensional coal flow data. Finally, the coal flow volume measurement is achieved through a coal flow volume measurement model. The experimental results show that the coal flow volume measurement method based on linear model partitioning has a precision of over 95% when measuring rough surface iron blocks, smooth surface iron blocks, and physical objects (coal and gangue) in high dust environment, coal flow surface watering environment, dim environment, and normal lighting environment. Moreover, the measurement precision of smooth-surface iron blocks is higher than that of rough-surface iron blocks in four simulated environments. It indicates that the better the flatness of the object surface, the higher the measurement precision. The environment has little impact on measurement precision. The actual test results show that the coal flow volume measurement method based on linear model partitioning has a measurement precision of over 97%. The corresponding average time is within 80 ms. Compared with the measurement method based on the KD tree, the overall precision has been improved by more than 6% and the processing timeliness has been doubled.

Research on fault detection of belt conveyor drum based on improved YOLOv5s

MIAO Changyun, SUN Dandan

2023, 49(7): 41-48. doi: 10.13272/j.issn.1671-251x.2022100039

<Abstract>(193) <HTML> (54) <PDF>(39)

Abstract:
At present, the detection efficiency of belt conveyor drum fault detection methods is low, the recognition accuracy is not high, and the feature extraction capability is poor. In order to solve the above problems, a belt conveyor drum fault detection method based on improved YOLOv5s is proposed. A small-sized detection layer has been added to the YOLOv5s network model, making it easier to detect smaller drum faults. The method introduces the convolutional block attention module (CBAM) between the Backbone and Neck to improve the accuracy of target detection. The method introduces efficient channel attention mechanism (ECA) in Neck to enhance feature extraction capabilities for drum faults. The experimental results show the following points. ① On the premise of meeting the real-time detection requirements, the average recognition accuracy of the improved YOLOv5s network model reaches 94.46%, which is 1.65% higher than before the improvement. ② The average accuracy of the improved YOLOv5s network model for detecting drum opening, rubber coating wear, and rubber coating detachment are 95.29%, 96.43%, and 91.65%, respectively, which are 1.56%, 0.89%, and 2.50% higher than before the improvement. A belt conveyor drum fault detection system based on improved YOLOv5s is designed and validated. ① The experimental platform test results show that the average accuracy of the belt conveyor drum fault detection system based on improved YOLOv5s for drum welding, rubber coating wear, and rubber coating detachment detection reach 95.29%, 96.43%, and 91.65%, respectively. The average accuracy of the three types of faults reaches 94.46%, and the detection speed is about 14 frames/s. ② The on-site test results show that the confidence levels for rubber coating wear and rubber coating detachment are 0.92 and 0.97, respectively. The fault type and location of the drum can be accurately identified. This indicates that the improved YOLOv5s-based belt conveyor drum fault detection system is feasible.

Fault diagnosis method for mine hoisting motor based on VMD and CNN-BiLSTM

LI Jingzhao, HE Na, ZHANG Jinwei, WANG Qing, LI Huashun

2023, 49(7): 49-59. doi: 10.13272/j.issn.1671-251x.2022120065

<Abstract>(88) <HTML> (15) <PDF>(30)

Abstract:
The traditional motor fault diagnosis method based on the audio signal is insufficient to obtain the feature information of the motor audio signal and the fault diagnosis precision is not high. In order to solve the above problems, a mine motor fault diagnosis method based on optimized variational mode decomposition (VMD) and convolutional neural network CNN bidirectional long short-term memory (BiLSTM) is proposed. The whale algorithm (WOA) optimized VMD is used to decompose the motor audio signal to address the issues of modal aliasing and endpoint effects. The motor audio signal is decomposed into K intrinsic mode functions (IMF). After Pearson correlation coefficient screening, the 13-dimensional static MFCC feature parameters of the main IMF component are extracted. In order to obtain the dynamic features of the signal, the first and second-order difference coefficients of the 13-dimensional static MFCC are extracted to form a 39-dimensional feature vector. By combining dynamic and static features, the performance of fault diagnosis can be improved. In order to improve the precision of fault diagnosis, a BiLSTM layer is introduced into the CNN. The CNN extracts local features of the audio signal in the spatial dimension. The BiLSTM preserves bidirectional time series information of the audio signal in the temporal dimension. It captures long-distance dependencies of the audio signal, thereby maximizing the preservation of global and local features. The experimental results show the following points. ① Each IMF component of VMD decomposition has an independent center frequency and uniform distribution, and exhibits sparsity in the frequency domain. It can effectively avoid modal aliasing problems. In IMF solving, VMD decomposition avoids endpoint effects in empirical mode decomposition (EMD) and ensemble empirical mode decomposition (EEMD) through mirror extension. ② The fault diagnosis accuracy based on 13-dimensional static MFCC features is 97.5%. The fault diagnosis accuracy based on 39-dimensional dynamic and static MFCC features is 1.11% higher than that based on 13-dimensional static MFCC features. ③ The accuracy of the diagnostic model based on CNN-BiLSTM reaches 98.61%, which is 5.83%, 4.17%, and 3.89% higher than the current universal diagnostic models CNN, BiLSTM, and CNN-LSTM, respectively.

Global dynamic collaborative management and control of diversified business in coal mines driven by digital twins

XING Zhen

2023, 49(7): 60-66, 82. doi: 10.13272/j.issn.1671-251x.2023060003

<Abstract>(132) <HTML> (27) <PDF>(32)

Abstract:
As a typical complex production system with multi business collaboration, coal mines face problems such as poor collaboration and linkage in safety, production, and operation under the dynamic sales demand and uncertain production environment. Digital twin technology can provide technical support for achieving data fusion, collaborative control, and intelligent linkage in the global business system of coal mines. This article analyzes the integration and collaborative control of diversified businesses in coal mines from two aspects: internal integration of professional businesses and integration between professional businesses. The internal integration of professional businesses includes three parts: safety monitoring business integration, production collaboration business integration, and operational business integration. We have constructed a business dynamic collaborative management and control architecture based on digital twins. It includes a physical object perception layer, a virtual space simulation layer, and a collaborative management and control decision-making layer. A triple-driven dynamic collaborative control model for the coal mine global business has been proposed. It includes "proactive static planning of decision algorithms, predictive collaborative control of twin models, and real-time data dynamics collaborative control". In the twin world, the multi-business virtual model for coal mine "safety - production - operation" is constructed. The proactive static planning for safety assurance, production linkage, and business management is carried out to formulate the optimal initial work plan for business collaborative control. The numerical simulation method is used to achieve the predictive operation of virtual models in the information world. After the effectiveness of control measures is verified, decision instructions are promptly issued to the physical world. Therefore, the disturbances from passive processing after the event are transformed into active control before the event, improving the effectiveness of control decisions. When there are unforeseen disturbances in the twin world during the operation of coal mining enterprises' business, based on real-time monitoring data, decision algorithms are used to determine the level of disturbance events. The corresponding emergency plans are triggered. The dynamic collaborative management and control of the diversified business of the enterprise are achieved.

Adaptive control of temporary support force based on PSO-BP neural network

TIAN Jie, LI Yang, ZHANG Lei, LIU Zhen

2023, 49(7): 67-74. doi: 10.13272/j.issn.1671-251x.2022100017

<Abstract>(137) <HTML> (31) <PDF>(20)

Abstract:
In order to make the temporary support force better adapt to the mine pressure and improve the support capacity of the support, taking the dual self-moving temporary support as the research object, an adaptive control method of temporary support force based on particle swarm optimization (PSO) - BP neural network is proposed. The initial weights of the BP neural network are optimized by using the global search capability and fast convergence features of the PSO algorithm to improve the rate of convergence of the BP neural network. Then, the optimized BP neural network is used to achieve online self-adjustment of PID parameters. The PSO-BP neural network is constructed to optimize the PID controller. This enables the temporary support force to reach the predetermined value more quickly and accurately, achieving adaptive control of the temporary support force. It avoids damage to the roof due to the mismatch between support force and roof pressure. The expected initial support force of the temporary support is simulated using unit step signals for experimental verification. The results show that compared with the BP neural network optimized PID controller and traditional PID controller, the PSO-BP neural network optimized PID controller can achieve the expected initial support force faster and more accurately. The adjustment time is only 0.5 s and there is almost no overshoot. Based on actual geological conditions, the roof pressure on the support during excavation support is simulated. The adaptive control effect of three controllers for support force is studied. The results show that under the control of the PSO-BP neural network optimized PID controller, the system error is only 0.02 MPa, with the smallest error and the best control effect.

Research on positioning method of mine track locomotives based on not line of sight error suppression

LI Zongwei

2023, 49(7): 75-82. doi: 10.13272/j.issn.1671-251x.2022120030

<Abstract>(61) <HTML> (10) <PDF>(15)

Abstract:
The positioning method for mine track locomotives is mainly based on ultra wide band (UWB) positioning. But the complex environment of underground transportation roadways and frequent not line of sight（NLOS） propagation seriously impact the precision of UWB positioning. At present, research on positioning errors caused by NLOS has problems such as complex algorithms and poor real-time positioning. Based on the analysis of the positioning characteristics of railway locomotives and the UWB positioning technology based on the time of arrival (TOA), a positioning method of mine track locomotive based on NLOS error suppression is proposed. Two positioning cards are installed at different positions on the track locomotive. Radio frequency identification technology is used to accurately divide the relative position relationship between the positioning card and the positioning base station. The distance measurement value of the UWB positioning signal is greater than the actual value under NLOS propagation conditions. Based on this situation, the empirical range of the difference between the distance measurement value and the actual value between the two positioning cards is used to calculate the discrimination threshold for NLOS propagation conditions under different position nelationship between positioning cards and positioning base station. By using this discrimination threshold, the ranging value of the positioning signal propagated by NLOS path is eliminated. The ranging value of the positioning signal propagated by the line of sight path is used for positioning calculation, so that to suppress NLOS errors and improve the average positioning precision of mine track locomotives. The test results show that using the positioning method of mine track locomotives based on NLOS error suppression, the average positioning error of locomotives is within 1 m when the positioning signal is in line of sight propagation conditions. When the positioning signal is in NLOS propagation conditions, most of the NLOS errors are effectively suppressed. The average positioning precision is about 1 meter. The positioning precision of track locomotives has been greatly improved compared with normal UWB positioning method based on TOA.

CatBoost mine pressure appearance prediction based on Bayesian algorithm optimization

CHAI Jing, ZHANG Ruixin, OUYANG Yibo, ZHANG Dingding, WANG Runpei, TIAN Zhicheng, LIU Hongrui, HAN Zhicheng

2023, 49(7): 83-91. doi: 10.13272/j.issn.1671-251x.2022110065

<Abstract>(74) <HTML> (26) <PDF>(10)

Abstract:
Obtaining mine pressure data through traditional monitoring methods and using statistical or machine learning algorithms to predict mine pressure can no longer meet the requirements of intelligent development in mines. It is necessary to seek new methods to improve the accuracy and real-time performance of mine pressure data monitoring and prediction. Based on three-dimensional similar physical model experiments, a distributed fiber optic monitoring system is constructed. The distributed fiber optic cables are pre-embedded along the model's direction and height. Pressure data is collected during the simulated mining process of the working face, and the optical fiber Brillouin frequency shift mean variation degree is introduced as an indicator to determine whether the pressure is coming. By preprocessing the optical fiber monitoring data such as noise removal, normalization and phase space reconstruction, the one-dimensional initial monitoring data is converted into three-dimensional data. The method uses Bayesian algorithm to iteratively optimize the parameters of the CatBoost algorithm. After reaching the maximum number of iterations, the optimal parameter combination is loaded into the CatBoost algorithm. The prediction model for mine pressure appearance is obtained by training. The results show that the Bayesian algorithm has fewer iterations and smaller errors than traditional grid search methods. Compared with random forest (RF), gradient boosting decision tree (GBDT) and extreme gradient boosting (XGBoost), the CatBoost algorithm has higher prediction accuracy and stronger generalization capability. The CatBoost mine pressure appearance prediction model optimized by the Bayesian algorithm can accurately predict the three weighting in the test set. The overall prediction trend is in line with the measured value, with mean absolute error of 0.0091, root-mean-square error of 0.0077, and determination coefficient of 0.933 9.

Ensemble learning mine pressure prediction method based on variable time series shift Transformer-LSTM

LI Zexi

2023, 49(7): 92-98. doi: 10.13272/j.issn.1671-251x.18142

<Abstract>(73) <HTML> (91) <PDF>(25)

Abstract:
The existing mine pressure prediction models are mostly single prediction models that rely on fixed length time series features. It is difficult to accurately capture the composite features of mine pressure time series data, which affects the accuracy of mine pressure prediction. To solve this problem, an ensemble learning mine pressure prediction method based on variable time series shift Transformer-long short-time memory (LSTM) is proposed. Based on the Laida criterion and Lagrange polynomial method, the outlier values in the mine pressure monitoring data are eliminated, and the missing values are inserted. Then normalized preprocessing is performed. The paper proposes a variable time series shift strategy to divide mine pressure time series data at different scales. It avoids potential data shift issues that may exist in fixed length time series. On this basis, the ensemble learning mine pressure prediction model based on Transformer-LSTM is constructed. By combining the attention mechanism and the accurate time series feature representation capability, the dynamic features of the mine pressure change law are captured at multiple levels. The voting algorithm of ensemble learning is used to jointly predict the mine pressure data to overcome the limitations of a single prediction model. The experimental results show that the voting algorithm of ensemble learning can reduce the volatility of mean absolute error (MAE) and root mean square error (RMSE) of mine pressure prediction. It effectively reduces the sensitivity impact of different scale feature series to the mine pressure prediction results. Compared with the Transformer model, the MAE of the Transformer-LSTM model's prediction results on two roof mine pressure datasets of fully mechanized working faces improves by 8.9% and 9.5% respectively, and the RMSE has increased by 12.7% and 16.5% respectively. The above indexes are also higher than those of back propagation (BP) neural network model and LSTM model. The method proposed in the paper effectively improves the accuracy of mine pressure prediction.

Study on the features of coal rock failure potential signal based on multiscale multifractal analysis method

WANG Heng, LI Zhonghui, ZHANG Xin, LEI Yueyu

2023, 49(7): 99-106. doi: 10.13272/j.issn.1671-251x.2022120003

<Abstract>(71) <HTML> (15) <PDF>(7)

Abstract:
The surface potential signals induced by the deformation and failure of coal and rock contain key information on damage evolution. It has been widely studied in the field of coal and rock dynamic disaster monitoring. However, most of these studies focus on the fluctuation features of potential time series signals in a single time dimension. There is a lack of in-depth research on the nonlinear and multiscale feature changes of the time series signals. To solve this problem, a monitoring system for the potential of coal and rock failure is built, and the potential time series signals of raw coal and gabbro samples are synchronously tested. Through the multiscale multifractal analysis (MMA) method, the nonlinear features of potential signals at multiple scales are studied in depth. The singularity index, singularity dimension, local Hurst index and other parameters of the potential time series signals are obtained. The Hurst surface is quantified by the L2 norm. The experimental results show that the overall potential signals of raw coal and gabbro show multiscale multifractal features, and the potential multifractal maps before and after crack initiation show some differences. Compared with gabbro, the positive and negative trends of the singularity index difference Δα of the potential signals of the coal samples at different positions in the pre-loading and post-loading phases show different features. It indicates a stronger non-linear evolution of the coal samples. The L2 norm of the local Hurst index at multiple scales better reflects the long-range correlation between different channel potential signals of the sample. It can quantify the nonlinear evolution features of the sample time series signals, thereby achieving the prediction of coal rock instability and failure.

Experimental platform for coal gangue sorting robot based on image detection

LI Sanxi, LI Ya'nan, WANG Zijie, HOU Peng, XUE Guanghui

2023, 49(7): 107-113. doi: 10.13272/j.issn.1671-251x.2022120028

<Abstract>(116) <HTML> (12) <PDF>(29)

Abstract:
Currently, coal gangue pre-sorting is still mostly done manually, with high labor intensity, low sorting efficiency, and safety hazards. Using coal gangue sorting robots to replace manual coal gangue pre-sorting is an effective way to ensure the health and safety of workers and improve work efficiency. However, the existing coal gangue sorting robots have poor performance in situations such as low light intensity and coal gangue surface covered with coal powder. To solve the above problems, an experimental platform for coal gangue sorting robot based on image detection is proposed. This experimental platform collects coal gangue images through industrial cameras. The platform uses ResNet18-YOLOv3 deep learning algorithm to identify the coal gangue in the images. The platform uses TCP communication to provide the position information of the gangue to the coal gangue sorting module for trajectory planning, then controls the manipulator to clamp the gangue and completes the gangue sorting operation. The platform uses the Halcon calibration method for hand-eye calibration of the experimental platform, in order to achieve the conversion of camera pixel coordinates and manipulator spatial coordinates. The positioning error of the experimental platform is calibrated. For coal gangue samples with sizes above 50 mm, the positioning error should not exceed 9 mm. The experimental results show that the recognition accuracy of the experimental platform for coal gangue under strong lighting conditions is 99%. The recognition accuracy of coal gangue under weak lighting conditions is 95%. The recognition accuracy of coal gangue under pulverized coal adhesion conditions is not less than 82%. The accuracy of coal gangue sorting is 82%.

Excitation device for mining steel wire rope based on magnetic flux leakage detection

LI Jianhui, SUN Xianbin, LIU Lunming, CHEN Rongxin, ZHAN Weixia

2023, 49(7): 114-119. doi: 10.13272/j.issn.1671-251x.2022090059

<Abstract>(177) <HTML> (52) <PDF>(21)

Abstract:
The current research does not consider the influence of wire rope swing on the excitation device in the context of engineering applications, resulting in unsatisfactory detection results. In order to solve this problem, a set of mining wire rope excitation devices has been designed. Through the establishment of the wire rope simulation model, the influence of different air gap and lift off value on the magnetic leakage field of wire rope is simulated and studied. It is found that increasing the air gap or lift off value will reduce the magnetic induction intensity of the magnetic leakage field of the wire rope and affect the magnetic leakage detection results of the wire rope. However, in practical applications, the swing amplitude of mining wire ropes is large and the ropes are easy to be polluted. Therefore, the air gap and lift off value of the wire rope excitation device should not be too small. Under the conditions of considering engineering applicability, the air gap is set to 6 mm and the lift off value is set to 5 mm. Further simulation analysis is conducted on the effects of permanent magnet thickness and length, magnetic pole spacing, and armature thickness on the leakage magnetic field of steel wire ropes. It is found that the thickness and length of permanent magnets have the greatest influence on the leakage magnetic field of steel wire ropes. The magnetic pole spacing has a small influence on the leakage magnetic field of steel wire ropes. The influence of armature thickness on the leakage magnetic field of steel wire ropes can be ignored. Based on the simulation results and considering economy and portability, the parameters of the wire rope excitation device are set. The permanent magnet thickness is set to 10 mm, the permanent magnet length is set to 30 mm, the magnetic pole spacing is set to 180 mm, and the armature thickness is set to 10 mm. The dynamic simulation results show that the peak-to-peak value of the magnetic flux density of the wire rope magnetic leakage field reaches 0.9 mT. It indicates that the excitation device can ensure high magnetic leakage at the damage. The experimental results show that the magnetic flux leakage signal shows significant fluctuations at different broken wire locations of the steel wire rope. It indicates that the excitation device has a good excitation effect and can accurately detect broken wire damage of the steel wire rope.

Short circuit transient power analysis of hybrid intrinsically safe circuit

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

2023, 49(7): 120-125. doi: 10.13272/j.issn.1671-251x.2023030085

<Abstract>(102) <HTML> (28) <PDF>(16)

Abstract:
Currently, research on the intrinsically safe features of intrinsically safe circuits mostly relies on the IEC spark experimental device as the experimental platform. The research only analyzes the discharge features of a single capacitor or inductance circuit. There are problems such as poor applicability and high requirements for experimental conditions. There is a lack of research on the intrinsically safe features of hybrid intrinsically safe circuits. To solve this problem, based on GB/T 3836.4-2010 Explosive Atmospheres - Part 4: Equipment Protected by Intrinsic safety Type "i", a short circuit transient energy experiment is carried out with the hybrid circuit under the cutoff type protection mode as the experimental object. By analyzing the release process of short circuit transient energy, a mathematical model of short circuit transient energy is established. The paper analyzes the effects of capacitance, inductance, power supply voltage, and protection time on short circuit transient energy in the equivalent mathematical model. The Matlab simulation results show that as the capacitance and inductance increase, the transient energy of the short circuit will gradually increase and eventually approach a stable value. Increasing the power supply voltage will significantly increase the short circuit transient energy. Shortening the action protection time can effectively reduce transient energy. But its effect is only significant when the protection time is less than the critical time. An intrinsically safe power supply is developed based on a mathematical model of short circuit transient energy. The short circuit experiments are conducted. The experimental results show that the waveform of short circuit current and voltage is basically consistent with theoretical analysis. The transient energy of short circuit is 33.22 μJ, which meets intrinsic safety requirements and can provide a reference for the design of intrinsically safe power supplies.

Transient interference analysis and suppression in the secondary circuit of electric control system of kilometer drilling rig

GUAN Zhengqi, LI Haiying, SONG Jiancheng

2023, 49(7): 126-133, 140. doi: 10.13272/j.issn.1671-251x.2022100053

<Abstract>(79) <HTML> (20) <PDF>(7)

Abstract:
The electric control system of kilometer drilling rig has complex operating conditions, variable loads, and integrates multiple primary circuits. This results in high randomness of transient interference spectrum distribution and easy occurrence of modal aliasing. In order to improve the precision of intelligent perception, the secondary circuit of the electric control system of kilometer drilling rig often uses a high bandwidth gain operational amplifier. The existing models applicable to the secondary port equipment are no longer suitable for the stability analysis of small signal detection circuits. The transient interference frequency domain distribution range of the electric control system of kilometer drilling rig is wide, requiring the circuit to have strong anti-interference capability in a wide frequency range. Traditional anti-interference measures have the disadvantages of narrow frequency bands and poor high-frequency suppression effects. Multi-level RC and LC filtering circuits have problems of impedance mismatch and large volume. In order to solve the above problems, the signal acquisition circuit of the secondary circuit of the electric control system of 15000-kilometer drilling rig is taken as the research object to analyze the transient interference in the secondary circuit. The empirical wavelet transform (EWT) algorithm based on nonparametric scale space is used to divide spectral segmentation points using scale space transformation. The modal components with tightly supported frames are extracted. The kurtosis index characteristics of modal components are introduced to divide transient interference signals and white noise signals. The frequency domain distribution of transient interference is determined. By constructing an equivalent model of the small signal circuit with parasitic parameters in the secondary circuit of the electronic control system, the law between the parasitic capacitance of the feedback circuit pin and the frequency threshold of the interference signal of triggering ringing or self-excited oscillation is explored. The influence of the frequency domain characteristics of the transient interference on the circuit stability is analyzed. The results indicate that when there is a pins parasitic capacitance of 30 pF in the input and output, conducting transient interference signals leads to a decrease in stability. The triggering frequency that causes instability decreases as the pin parasitic capacitance increases. Using the high resistance characteristics of ferrite bead similar to parallel resonance, a second-order filtering circuit is designed. The experimental verification results in laboratry show that when the interference passes through the second-order filtering circuit containing ferrite bead, the interference amplitude is suppressed below −35 dBV in the frequency band above 0.2 MHz sensitive to the signal sampling circuit. The signal sampling circuit has no abnormal output. The interference amplitude of sensitive frequency band in the operation data of the industrial prototype is suppressed below −35 dBV. The actual operation test results are basically consistent with the laboratory test results, meeting the anti-interference requirements.

Intelligent detection method of working personnel wearing safety helmets in underground mine

DU Qing, YANG Shijiao, GUO Qinpeng, ZHANG Huanbao, WANG Yuchen, YIN Yu

2023, 49(7): 134-140. doi: 10.13272/j.issn.1671-251x.2022090033

<Abstract>(231) <HTML> (76) <PDF>(41)

Abstract:
Visual image-based methods are currently a hot topic in intelligent detection of mine personnel wearing safety helmets. However, existing methods use limited underground mining data and the classification of safety helmet features is not accurate enough. By collecting images of actual production scenes such as underground mining sites and roadways, a mining helmet wearing dataset (MHWD) is constructed. The helmet wearing situation is further divided into three categories: correct wearing, non-standard wearing, and non wearing. YOLOX algorithm is used to detect personnel wearing helmets. In order to enhance YOLOX's capability to extract global features, the attention mechanism is introduced. The effective channel attention module is embedded in the spatial pyramid pooling bottleneck layer of YOLOX's backbone network. The convolutional block attention module is added after each upsampling and downsampling of the path aggregation feature pyramid network, thus the YOLOX-A model is built. By using MHWD, the YOLOX-A model is trained and validated. The results show that the YOLOX-A model can accurately identify the wearing of safety helmets by personnel in mine images with low illumination, blurriness, and personnel obstruction. The F1 scores for the classification results of non-standard wearing, correct wearing, and non wearing safety helmets are 0.86, 0.92, and 0.89, respectively. The average precision is 93.16%, 95.76%, and 91.69%. The average precision mean is 93.54%. The overall F1 score is 4% higher than the YOLOX model. The detection precision is higher than the mainstream target detection models EfficientDet, YOLOv3, YOLOv4, YOLOv5 and YOLOX.

A filtering method for underground manipulator communication signal

LIU Yanjia

2023, 49(7): 141-146. doi: 10.13272/j.issn.1671-251x.18057

<Abstract>(80) <HTML> (27) <PDF>(14)

Abstract:
When a multi-degree-of-freedom manipulator operates in underground coal mines, its communication signal is easily affected by external factors such as pulses and electromagnetic interference. It reduces the reliability of signal transmission and affects the flexibility and control precision of the multi-degree-of-freedom manipulator. To improve the communication signal quality of underground multi-degree-of-freedom manipulator, a filtering method for underground manipulator communication signal based on improved time-frequency peak filtering is proposed. The method extracts carrier features based on the carrier nodes of the manipulator communication signal, and extracts transmission period features based on Hilbert transform theory. The method uses wavelet decomposition to solve the wavelet transform coefficient of each level, obtains the disturbance frequency of communication signals, and completes the classification of communication signal features. The method introduces phase compensation factors to improve time-frequency peak filtering. The reduction threshold is calculated according to the scaling factor and reduction factor, so as so adjust the fixed window length. The improved time-frequency peak filtering is used to filter the noise in communication signals. Thus the noise suppression of the multi-degree-of-freedom manipulator communication signal is achieved. A 3-degree-of-freedom manipulator is taken as the experimental object. The pulse width is set to be 8 μs. The Baud rate is 120 bit/s, the communication signal transmission interval is 2 ms, and the signal amplitude is 1. 30 dB Gaussian white noise is added to the communication signal. The experimental results show that the output signal-to-noise ratio of the communication signal processed by the filtering method for underground manipulator communication signal based on improved time-frequency peak filtering is 94 dB. The bit error rate is basically stable at 0.6%, and the average frame loss rate is reduced from 2.9% before filtering to 0.8%. The method effectively improves the communication signal quality and transmission reliability of the multi-degree-of-freedom manipulator.

2023 Vol. 49, No. 7

Submission System

Latest AcceptedMore>