Online First have been peer-reviewed and accepted, which are not yet assigned to volumes /issues, but are citable by Digital Object Identifier (DOI).
Bauxite Load Classification Based on Wavelet Packet Decomposition of Vibration Characteristic and SSA-BPNN
Abstract:
The physical and chemical properties of bauxite and rock are different. The mining machinery will produce noise pollution and mechanical loss when cutting rock. Due to the influence of ore-body occurrence conditions and underground environment, the artificial judgment of cutting load type has lag and uncertainty. To solve the problem, put forward a kind of based on wavelet packet decomposition and sparrows search algorithm to optimize BP load classification method, with the two types of load under the conditions of the actual working condition of vibration data for the data samples, based on the wavelet packet decomposition, get the vibration data of feature vector, using principal component analysis (pca), reduces the dimensions of feature vector. The topology structure of BP neural network and the parameters of sparrow search algorithm are determined, and the weights and thresholds of BP neural network are optimized by sparrow search algorithm, which speeds up the convergence of the network and avoids falling into the local minimum value. Experimental results show that this method improves the efficiency and accuracy of load classification.
<Abstract>(0)
Research on autonomous Positioning technology of inspection robot in Fully mechanized Mining face
Abstract:
Because the work space is small and the environment is changing with the advancing of the working face, a kind of inspection robot system is designed which can complete the inspection task independently. The robot is equipped with 3d scanner to complete the recon-struction of underground space working face scene, and the rigid-flexible track design is adopted to adapt to the environment of the fully mechanized working face. Aiming at the design characteristics of rigid-flexible integrated track, in order to reduce the influence of the track connectors on the inertial navigation and positioning of the inspection robots. Inertial navigation/odometer incremental navigation is used to complete the positioning of the inspection robot,the experiment shows that the positioning accuracy of this scheme can reach 10-3 magnitude on 40km road. At the same time, through the dynamic analysis of the gyroscope data when the inspection robot passes through the track connection point, the identification of the track connection piece and the detection of the jitter point are completed, and the segmentation filtering is realized through the jitter detection to improve the positioning accuracy of the inspection robot. The experi-mental results show that the judgment of track connection is accurate, and the piecewise filtering based on jitter detection can effectively improve the positioning accuracy of inspection robot.
<Abstract>(0)
Research progress and key technologies of gangue sorting robot
Abstract:
According to our country coal mine intelligent development strategy, in-depth analysis of the current situation of coal waste sorting at home and abroad, combined with the present problems existing in the coal waste sorting has carried on the analysis and discussion, summarizes the common key technologies of coal waste sorting robot, pointed out the coal waste sorting robots technical characteristics, and future development trend, and the feasible solution is given. First of all, from coal gangue identification, mechanical arm manipulator trajectory planning and coordination of gangue intelligent sorting robot research progress were analyzed, and points out that the existing recognition methods are achieved from the principle of coal gangue identification, also taking into account the complex working condition of coal waste sorting, able to adapt to the study in underground coal waste identification and sorting feature extraction method; It is pointed out that most of the existing trajectory planning methods have the problems of weak prediction ability and low adaptability to moving targets with variable trajectories. It is necessary to further study the precise positioning of dynamic targets, synchronous tracking and online trajectory planning of manipulator. It is pointed out that the existing multi-manipulator sorting method has the problems of low autonomy of single manipulator and low cooperation degree of multi-manipulator. It is necessary to further study the intelligent cooperative control method of multi-manipulator coal gangue sorting system. Secondly, by sorting out the current related technologies of intelligent coal and gangue sorting robots, it is pointed out that coal gangue identification technology, robot trajectory planning method, and multi-dynamic target multi-robot collaboration technology have become common key technologies that need to be studied. Finally, it is concluded that: Coal gangue data set construction and amplification, coal gangue identification and sorting feature extraction are the key technologies to achieve efficient coal gangue identification. Accurate tracking of dynamic gangue, synchronous tracking of dynamic target trajectory planning by manipulator and stable grasping of fast and large mass target are the key technologies to achieve stable grasping of gangue by manipulator. Efficient multi-task allocation, anti-collision path planning and intelligent cooperative control are the key technologies to realize efficient and intelligent cooperative sorting of multi-manipulator. Finally, in view of the common problems existing in the current intelligent sorting robot, a solution is proposed: in the aspect of recognition, the coal gangue recognition and grasping feature extraction method based on multi-modal deep learning is studied to realize the rapid recognition of coal gangue suitable for underground; In the aspect of trajectory planning, the precise positioning and real-time tracking methods of dynamic gangue are studied to realize the adaptive and stable grasping of dynamic gangue by robot. In terms of collaborative sorting, a multi-layer and multi-manipulator cooperative control model can be constructed to achieve efficient and intelligent collaborative sorting under complex environment of multi-manipulator.
<Abstract>(0)
Visual SLAM algorithm for underground coal mine considering image enhancement
Abstract:
In order to improve the applicability of the visual SLAM algorithm in underground coal mines, a visual SLAM algorithm considering image enhancement is proposed, which improves the overall performance of visual SLAM through image enhancement processing. Firstly, a Retinex algorithm based on improved bilateral filtering is designed to enhance the coal mine image. Convert the original image to the HIS (Hue, Saturation, Intensity) color space, and use the improved bilateral filter function to replace the Gaussian kernel function in the Retinex algorithm to estimate the reflection component of the intensity component , and then convert back to the RGB color space to obtain an enhanced image with improved contrast and unaffected by lighting. Compared with the Single-Scale Retinex (SSR) and Muti-Scale Retinex (MSR) algorithms, the images processed by this algorithm do not appear obvious whitening and halo phenomena, and the image quality has been significantly improved. Secondly, the algorithm is introduced into the classic ORB-SLAM2 algorithm framework for subsequent pose estimation and mapping. Finally, in order to verify the feasibility and applicability of the algorithm in this paper, experiments are carried out in the coal mine roadway environment. The results show that, compared with the ORB-SLAM2 algorithm, this algorithm has better positioning accuracy and mapping effect in underground coal mine, which provides an important technical support for the visual perception and positioning of the mine robot.
<Abstract>(0)
More
[Academic Column of Editorial Board Member]
Coal mine rock burst and coal and gas outburst perception alarm method based on color image
SUN Jiping, CHENG Jijie, WANG Yunquan
2022, 48(11): 1-5.   doi: 10.13272/j.issn.1671-251x.18042
Abstract: The paper analyzes the image characteristics of the thrown coal and rock when rock burst and coal and gas outburst occur. ① The coal and rock thrown out during rock burst and coal and gas outburst are mainly black, but the underground equipment of the coal mine is generally not black. Therefore, nonblack mining equipment can be used as the background and color cameras can be used to identify coal and rock. ② The normal coal falling speed, the moving speed of shearer, roadheader, and the moving speed of underground personnel and vehicles are far less than the speed of coal and rock thrown out in the event of rock burst and coal and gas outburst. Therefore, according to the speed characteristics, the interference from normal coal falling, movement of equipment such as shearer and roadheader, and movement of underground personnel and vehicles can be eliminated. ③ The explosion of gas and coal dust will also cause the objects in the roadway to have a high speed in a short time, accompanied by high brightness. But the rock burst and coal and gas outburst will not produce high brightness. Therefore, according to the average image brightness, the interference of gas and coal dust explosion can be eliminated. The paper proposes a color camera set method. The camera of the heading face should be set at the roof of the heading roadway or near the roof on both sides of the heading roadway. The camera of the working face should be set on the top of the hydraulic support. The paper puts forward a coal mine rock burst and coal and gas outburst perception alarm method based on color image. ① The color camera with fill light shall be set at the roof of the heading roadway or near the roof on both sides of the heading roadway, and at the top of the hydraulic support of the working face. The nonblack mining equipment is used as the background. ② The method monitors and identifies whether the color of the color image has changed greatly. ③ If the image color changes significantly, the average brightness of the image is identified, otherwise the monitoring of the identified image color change continues. ④ If the average brightness of the image is less than the set brightness threshold, the movement speed of the object causing a large change in the image color is identified, otherwise the monitoring of the identified image color change continues. ⑤ If the movement speed of the object is greater than the set speed threshold value, the methane concentration in the monitoring area is identified, otherwise the monitoring of the identified image color change continues. ⑥ If the methane concentration rises rapidly or reaches the alarm value, the coal and gas outburst alarm will be given. Otherwise, the rock burst alarm will be given. The method has the advantages of non-contact, wide monitoring range, low cost, convenient use and maintenance.
<HTML> <PDF>(806KB)
[Academic Column of Editorial Board Member]
Analysis of 5G private network technology and coal mine 5G private network scheme
HUO Zhenlong
2022, 48(11): 6-10, 19.   doi: 10.13272/j.issn.1671-251x.18040
Abstract: Most of the existing communication networks except 5G exist in the form of independent private networks. The performance of 5G wireless communication in bandwidth, delay, number of terminal connections, and reliability has been greatly improved. Accordingly, there are new changes in network architecture and networking mode. The private network scheme is no longer a single independent private network scheme. There are also hybrid private networks and virtual private networks. This paper introduces the key technologies of 5G private network, such as network slicing, mobile edge computing, 5G LAN, time-sensitive network and so on. This paper proposes three kinds of 5G private network schemes, namely virtual private network, hybrid private network and independent private network. The virtual private network has the characteristics of wide service scope, high flexibility, low cost, and short construction cycle. It is used for various applications which have wide coverage, have access terminals not fixed in time and space, and have certain service quality requirements and certain degree of data isolation requirements. The hybrid private network has short transmission path, high security, and low end-to-end delay. It can carry out a variety of flexible independent services, but privacy is weak. The independent private network provides a physical exclusive 5G private network to meet the needs of industrial users for high bandwidth, low delay, high security and high-reliability data transmission. This paper puts forward the general principles of 5G private network scheme selection in terms of safety, availability and reliability. This paper also proposes the special requirements of 5G private network in coal mine in terms of dispatching function, integration demand, independent operation and maintenance and intrinsic safety. The selection of 5G special network scheme for coal mine is proposed. In the early stage of intelligent construction of coal mine or there is no strict requirements on data confidentiality, system use convenience, system function expansion in the coal mine, the virtual private network or hybrid private network scheme can be selected. Otherwise, the independent private network scheme can be selected. It is pointed out that relatively more hybrid private network and virtual private network schemes are adopeted in coal mines at present. The hybrid private network and virtual private network will have some advantages in the future. With the gradual establishment of small core network ecology, the independent private network scheme will be recognized by more users. In a certain period in the future, the independent private network, virtual private network and hybrid private network schemes will give full play to their respective advantages. They serve the intelligent construction of coal mines in different periods and with different requirements.
<HTML> <PDF>(587KB)
[Academic Column of Editorial Board Member]
Research on multi monitoring information fusion and linkage of intelligent mine
HE Yaoyi, GAO Wen, YANG Yao, JING Cheng, ZHU Shasha, CHEN Xing
2022, 48(11): 11-19.   doi: 10.13272/j.issn.1671-251x.17962
Abstract: There are many types of coal mine automatic monitoring systems, and the technical routes are not unified. The system software is relatively independent, and the data is lack of correlation. At present, field data fusion and linkage control are mostly realized by the underground fusion substation or the ground fusion platform. It is difficult to realize the unified integration and linkage control of the whole mine from the bottom perception to the ground fusion. Based on the requirements of multi-system fusion of coal mine safety monitoring system and intelligent construction of coal mine, the key problems to be solved in multi monitoring information fusion of mine are analyzed. The problems include integrated acquisition and fusion of monitoring data of personnel, machine and environment, efficient and consistent sharing of safety monitoring and control data, low code and rapid secondary development of automatic monitoring system, and integrated supervision of whole life cycle of mine equipment objects. The scheme of multi monitoring information fusion and linkage for intelligent mine is proposed. The overall framework including underground data fusion and linkage control and ground multi monitoring information fusion is constructed. This paper introduces the implementation scheme of underground data fusion and linkage control based on edge fusion substation, and expounds on the key technologies of ground multi monitoring information fusion from three aspects, which include unified technology system, unified technology architecture and data processing mechanism, and deep information fusion based on the mine object information model. Therefore, an open integrated management and control platform of multi monitoring information is developed. Based on the coal industry communication driving protocol set embedded in the scheme and the basic supporting technologies such as coal mine monitoring, control, position service, 2D and 3D GIS, and workflow engine, the following platforms can be rapidly developed: the independent software platforms for automatic monitoring system of environmental safety monitoring, mobile target positioning and coal flow transportation control, the integrated safety production monitoring and control platform and integrated management and control platform of intelligent mine. The scheme forms industry-level real-time industrial configuration software.
<HTML> <PDF>(8128KB)
[Special of Intelligent Coal Separation Technology and Application]
Development and exploration of intelligent dense medium separation technology for coal
DAI Wei, WANG Yudong, DONG Liang, ZHAO Yuemin
2022, 48(11): 20-26, 44.   doi: 10.13272/j.issn.1671-251x.2022060106
Abstract: Dense medium separation, the most widely used coal preparation process, is moving from automation and informatization to intelligence. At present, the intelligent construction of dense medium coal preparation plant only realizes partial intelligent construction. It is deficient in the whole intelligent construction. The intelligent development of the core production equipment (dense medium cyclone and shallow groove) is insufficient. In order to solve the above problems, the research status of intelligent dense medium separation is described from three aspects of intelligent perception, intelligent control and intelligent optimization decision. The challenges faced by dense medium separation in the process of developing from automation to intelligence are analyzed. The challenges include the unstable operation caused by the fluctuation of raw coal quality, the high complexity of dense medium separation, and the limitations of intelligent construction of dense medium coal preparation plant. In order to promote the intelligence and greening of the dense medium separation industry, realize the autonomous control of the whole equipment, reduce the number of operators and even realize unmanned, a system is proposed. It is pointed out that the dense medium coal preparation plant should build a set of intelligent optimization production system with the integration of "intelligent perception, intelligent control and intelligent optimization decision". Intelligent perception, the basis of intelligence, is used to realize the perceptual acquisition of coal preparation process data. Intelligent optimization decision analyzes the operation state of the preparation process in the intelligent control module and adjusts the set value of the process index. Intelligent optimization decision analysis intelligent control module is used to sort process operating state, adjust the process indicators set value, so as to achieve dynamic optimization of the process indicators set value. The mutual coordination of perception, control and decision promotes the improvement of the intelligence level and production efficiency of the coal preparation plant. The coordination provides a new idea for realizing intelligent collaborative optimization control of the whole dense medium separation production process in the future.
<HTML> <PDF>(754KB)
[Special of Intelligent Coal Separation Technology and Application]
X-ray transmission intelligent coal-gangue recognition method
WANG Wenxin, HUANG Jie, WANG Xiuyu, SHI Yulin, WU Gaochang
2022, 48(11): 27-32, 62.   doi: 10.13272/j.issn.1671-251x.18037
Abstract: The coal-gangue image recognition is an important part of coal-gangue separation technology based on pseudo dual energy X-ray transmission (XRT). However, it is difficult to segment the coal-gangue image due to the close proximity or occlusion of coal-gangue, and it is easy to cause classification and recognition errors of coal-gangue based on artificial threshold discrimination. Due to the above influence, existing coal-gangue recognition methods have low precision. In this paper, an X-ray transmission intelligent coal-gangue recognition method is proposed. A U-Net model combined with the receptive field block (RFB) is used to realize the effective segmentation of the pseudo dual energy X-ray coal-gangue image, which is termed as RFB + U-Net model. The problem that the recognition precision is affected by the close proximity or shielding of coal-gangue is solved. The recognition features of coal-gangue are the minimum gray value of the low-energy image in the gray level features of coal-gangue image, and the minimum value and the average difference of sharpened low-energy image in the texture features. A multi layer perceptron (MLP) model is used to realize coal-gangue recognition. Experimental results show that the RFB+U-Net model is superior to the active contour model, U-Net model and SegNet model in terms of coal-gangue segmentation accuracy, coal-gangue particle size precision, coal-gangue pixel mean intersection ratio and image segmentation effect. The reasoning time of the model is short, meeting the real-time requirements of coal-gangue image segmentation. When the number of hidden layers in the MLP model is 8, the average coal-gangue recognition accuracy under two test sets is more than 87%. Under the same data set and experimental conditions, the average recognition accuracy and gangue removal rate of the MLP model are higher than those based on Bayesian classifier, support vector machine, logic regression, decision tree, gradient boosting decision tree and K-nearest neighbor algorithm. The coal carrying rate of gangue shall not exceed 3%, meeting the requirements of actual dry coal-gangue separation.
<HTML> <PDF>(1579KB)
[Special of Intelligent Coal Separation Technology and Application]
Accurate recognition of coal-gangue image based on lightweight HPG-YOLOX-S model
CHEN Biao, LU Zhaolin, DAI Wei, SHAO Ming, YU Dawei, DONG Liang
2022, 48(11): 33-38.   doi: 10.13272/j.issn.1671-251x.18035
Abstract: The existing coal-gangue separation methods based on vision technology have problems of large model parameter amount, poor feature extraction capability and low recognition precision. In order to solve the above problems, a coal-gangue recognition method based on YOLOX-S model combined lightweight Ghost-S network and hybrid parallel attention module (HPAM) named HPG-YOLOS-S model is proposed. Firstly, HPAM is added to the backbone network of YOLOX-S model. Thus the important information in an image is enhanced, and the secondary information is inhibited. The feature extraction capability of the backbone network is enhanced. Secondly, the backbone network of YOLOX-S model is replaced by Ghost-S network with smaller parameter quantity. The utilization rate and feature fusion capability are improved. Finally, in the predection layer, the SIOU loss function is used to replace the loss function of YOLOX-S model to impsrove the detection and positioning precision and enhance the extraction capability of the target. In order to verify the detection effect of the proposed method on large coal-gangue, the HPG-YOLOX-S model is compared with YOLOX-S model. The results show that the identification accuracy of the HPG-YOLOX-S model for coal and gangue is 99.53% and 99.60% respectively, which is 2.51% and 1.27% higher than those of YOLOX-S model. The results of validation show that the precision rate, recall rate and F1 value of the HPG-YOLOX-S model are all above 94%, which are 5.68%, 3.51% and 2.91% higher than those of YOLOX-S model respectively. The parameters amount of the HPG-YOLOX-S model is 7.8 MB, which is 1.2 MB lower than that of YOLOX-S model. The ablation experiment results show that the mean average precision of the HPG-YOLOX-S model is 9.17% higher than that of YOLOX-S model. The experiment result of visualization of the thermodynamic diagram shows that the HPG-YOLOX-S model focuses on the surface differences between coal and gangue, such as texture and contour. The model pays more attention to the overall target of coal-gangue.
<HTML> <PDF>(9170KB)
[Special of Intelligent Coal Separation Technology and Application]
Coal and gangue recognition research based on improved YOLOv5
ZHANG Shiru, HUANG Zongliu, ZHANG Yuanhao, ZHANG Ao, JI Liang
2022, 48(11): 39-44.   doi: 10.13272/j.issn.1671-251x.2022060052
Abstract: The existing deep learning-based coal and gangue recognition methods are prone to false detection and missed detection when applied to underground complex environments. The recognition precision of small target coal and gangue is low. In order to solve this problem, an improved YOLOv5 model is proposed, and coal and gangue recognition is realized based on that model. Data enhancement is carried out on the collected coal and gangue data to enrich the data set and improve the data utilization rate. The atrous convolution and residual block are introduced into the spatial pyramid pooling (SPP) module to obtain the residual ASPP module. On the premise of not losing image information, the convolution output receptive field can be increased to enhance the extraction of deep features from the model. The AdaBelief optimization algorithm is used to replace the original Adam optimization algorithm of YOLOv5 to improve the convergence speed and recognition precision of the model. The experimental results show that the AdaBelief optimization algorithm and residual ASPP module can effectively improve the precision, recall rate and mean average precision (mAP) of the YOLOv5 model. The mAP of the improved YOLOv5 model reaches 94.43%, which is 2.27% higher than that of original YOLOv5 model. The frame rate is reduced by 0.03 frames/s. The performance of the improved YOLOv5 model is superior to SSD, Faster R-CNN, YOLOv3, YOLOv4 and other mainstream target detection models. In extremely dark environments, the improved YOLOv5 model can also accurately delineate the target boundary, and the recognition effect is better than other improved YOLOv5 models.
<HTML> <PDF>(2563KB)
[Special of Intelligent Coal Separation Technology and Application]
Application status and prospect of AI video image analysis in intelligent coal preparation plant
SHE Xiaojiang, LIU Jiang, WANG Lanhao
2022, 48(11): 45-53, 109.   doi: 10.13272/j.issn.1671-251x.2022060092
Abstract: Artificial intelligence (AI) video image analysis is an important part of intelligent coal preparation plant. It can realize the intelligent monitoring of important parameters of the equipment, environment, personnel and the whole process of coal preparation. The basic structure of the intelligent coal preparation plant is proposed. It is pointed out that the existing research mostly uses AI video image analysis technology to construct the safety monitoring system of coal preparation plant for personnel, machine, environment and management. The construction process of the intelligent video image monitoring system is proposed. In view of the two goals of safe and environment-friendly production and improving product quality in the intelligent construction of coal preparation plant, the application status of AI video image analysis technology in the intelligent coal preparation plant is introduced from six aspects. The aspects include foreign object detection, intelligent separation, equipment running state monitoring, coal particle size detection, personnel behavior monitoring and environmental safety detection. The intelligent application of AI video image analysis in coal preparation plant is proposed. It is pointed out that it is necessary to build a multi-level video monitoring system based on 5G communication, the Internet of Things, AI, intelligent control theory and coal preparation industry technology from the macro architecture. It is also necessary to optimize existing general intelligent video monitoring methods or algorithms from a micro perspective, and develop intelligent video image analysis technology suitable for the coal preparation plant environment. Machine vision and computer vision should be highly integrated with deep learning. The different advantages of machine vision and computer vision should be reasonably applied in different working conditions. It is suggested to establish a multi-level integrated monitoring system framework, and deploy and optimize the algorithm model within the framework. It is suggested to establish a diversified video image database, make full use of data characteristics of different image types, and develop targeted analysis algorithms. It is suggested to deeply study the distributed data stream and real-time AI video image analysis, build a real-time AI distributed system, reasonably schedule the video image analysis model, and improve the calculation efficiency and accuracy of the real-time model.
<HTML> <PDF>(1084KB)
More

Video RecommendationMore>

Industry NewsMore>