工矿自动化

Explosion proof requirements and detecting methods for radio wave transmission power

SUN Jiping, PENG Ming

2024, 50(6): 1-5, 22. doi: 10.13272/j.issn.1671-251x.18203

<Abstract>(1105) <HTML> (64) <PDF>(247)

Abstract:
The current national standard GB/T 3836.1-2021 Explosive atmospheres-Part 1: Equipment-General requirements and the international standard IEC 60079-0:2017 Explosive atmospheres-Part 0: Equipment-General requirements stipulate that the threshold power of a radio transmitter is the product of the effective output power of the radio transmitter and the antenna gain. Under the condition of a certain threshold for the safe transmission power of radio wave explosion-proof, the larger the antenna gain, the smaller the effective output power of the radio transmitter. This will limit the improvement of wireless transmission distance by increasing the antenna gain. Therefore, it is necessary to study the correctness of the threshold power specified in the national standard GB/T 3836.1-2021 and the international standard IEC 60079-0:2017, and propose reasonable explosion-proof requirements and detection methods for radio wave transmission power. It has been proposed that the safe transmission power of radio waves is independent of antenna gain, and the threshold power of radio transmitters specified in the national standard GB/T 3836.1-2021 and the international standard IEC 60079-0:2017 is incorrect. It is proposed that the threshold for the safe transmission power of underground wireless radio waves in coal mines should be greater than 16 W and independent of antenna gain. The national standard GB/T 3836.1-2021 and the international standard IEC 60079-0:2017 stipulate that the threshold power shall not exceed 6 W, which is incorrect. A method for detecting the explosion-proof safety performance of wireless radio waves has been proposed. The method detects the output power of wireless transmitters. This not only ensures the explosion-proof safety of the detected wireless equipment, but also simplifies the detection method. The method improves the wireless radio wave transmission power of the wireless equipment, removes the limitation on antenna gain, and greatly improves the wireless transmission distance of wireless explosion-proof equipment in coal mines.

Current status and prospects of research on landslide disasters in mine slopes based on multi-source information fusion

LI Hui, HAN Xiaofei, ZHU Wancheng, SONG Qingwei, ZHOU Wenlong

2024, 50(6): 6-15. doi: 10.13272/j.issn.1671-251x.2024040064

<Abstract>(741) <HTML> (80) <PDF>(246)

Abstract:
In order to overcome the problem that a single information source cannot accurately characterize the evolution features of mining landslide disasters, based on multi-source information fusion technology, this paper summarizes the research progress of mine slope landslide disasters from three aspects: multi-source information acquisition of mine slopes, multi-source information fusion of mine slopes, and mine slope displacement prediction and landslide risk assessment. The study summarizes typical slope monitoring methods of "sky", "air", and "ground" , as well as integrated collaborative monitoring method of "sky-air-ground". The study sorts out the slope multi-source information fusion process that includes data level, feature level, and decision level fusion. The paper organizes the fusion forms of displacement and stress, displacement and hydrological and meteorological monitoring information, as well as other different types. This paper elaborates on the current research status of slope displacement prediction and landslide risk assessment based on multi-source information fusion. The accuracy of disaster analysis in current research on mine slope landslide disasters heavily depends on the quality of monitoring data and insufficient utilization of knowledge of rock mechanics mechanisms. Based on the above problems, the development trends of research on landslide disasters in mine slopes are pointed out. The multi-source data collection and access standards are unified. The method for analyzing landslide disasters in mine slopes is developed by integrating monitoring data with rock mechanics mechanisms. The spatiotemporal association mining algorithm for multi-source information from the "sky-air-ground" is optimized. The construction of a mine slope landslide disaster warning platform based on multi-source information fusion is strengthened.

Automatic sealing system for goaf along gob-side entry retaining

NIE Baisheng, XIA Xiaofeng, ZHOU Haowen, QIN Feng

2024, 50(6): 16-22. doi: 10.13272/j.issn.1671-251x.2024040042

<Abstract>(315) <HTML> (39) <PDF>(152)

Abstract:
The existing sealing methods for goaf along gob-side entry retaining mainly focus on building sealing walls and sealing wall cracks. The construction period is long and repeated, which consumes a lot of labor costs, has a low degree of automation, and is prone to secondary damage. In order to solve the above problems, an automatic sealing system for goaf along gob-side entry retaining has been designed. The system uses flexible sealing airbags as carriers, placing uninflated airbags between the sealing wall and individual hydraulic pillars in the goaf. The system inflates the airbags to make them in contact with the roof and floor of the goaf and the outer side of the sealing wall in the goaf. The intelligent perception of mine pressure causes deformation of the surrounding rock of the roadway, and the shape of the airbag changes flexibly at any time. That is, when the internal pressure of the airbag rises and exceeds the rated pressure of the safety relief valve, it automatically releases the airbag gas to reduce the volume. The airbag re tightly adheres to the roof and floor surrounding rock. It achieves the effect of continuous sealing of the goaf and suppresses the leakage of dangerous gases in the goaf. The on-site test results show that the safety relief valve opens normally when the internal pressure of the flexible sealing airbag reaches about 4 kPa and stops venting when it reaches about 2.7 kPa. Flexible sealing equipment can sense changes in pressure and shrink its volume to adapt to the shape of surrounding rock, allowing for long-term and sustained sealing of goaf. After the installation of flexible sealing equipment, it has a high degree of adhesion with the sealing wall in the goaf. The volume fraction of gas in front of the sealing wall is reduced by 0.13%, effectively suppressing gas overflow.

Adaptive impact resistance support method for advanced hydraulic supports in mines based on digital twins and deep reinforcement learning

ZHANG Fan, SHAO Guangyao, LI Yuhan, LI Yuxue

2024, 50(6): 23-29, 45. doi: 10.13272/j.issn.1671-251x.2023090004

<Abstract>(476) <HTML> (36) <PDF>(67)

Abstract:
Due to the disturbance of geological disasters such as deep mining and rock burst, there are problems such as poor self perception capability, weak intelligent anti impact adaptive capability, and lack of decision-making and control capability in the advanced support system of the mine. In order to solve the above problems, a adaptive impact resistance support method for advanced hydraulic supports in mines based on digital twins and deep reinforcement learning is proposed. By sensing the roadway environment and advanced hydraulic support status through multiple sensors, a digital twin model of a physical entity is created in a virtual world. The physical model accurately displays the structural features and details of the advanced hydraulic support. The control model realizes adaptive control of the advanced hydraulic support. The mechanism model realizes logical description and mechanism explanation of the adaptive support of the advanced hydraulic support. The data model stores the physical operation data and twin data of the advanced hydraulic support. The simulation model completes the simulation of the advanced hydraulic support column to achieve virtual real interaction between the advanced hydraulic support and the digital twin model. According to the adaptive impact resistance decision-making algorithm based on deep Q-network (DQN) for advanced hydraulic support, intelligent decision-making is made for roadway impact resistance support in the simulation environment. Based on the decision results, control instructions are issued to physical entities and digital twin models to achieve intelligent control of advanced hydraulic support. The experimental results show that the displacement and pressure changes of the column are consistent, indicating that the simulation model design of the advanced hydraulic support column is reasonable, thereby verifying the accuracy of the digital twin model. The adaptive impact resistance decision-making algorithm for advanced hydraulic supports in mines based on DQN can adjust the PID parameters of the hydraulic support controller, adaptively regulate the column pressure, improve the safety level of roadways, and achieve adaptive impact resistance support for advanced hydraulic supports.

Spatiotemporal multi-step prediction of hydraulic support pressure based on LSTM-Informer model

YU Qiongfang, YANG Pengfei, TANG Gaofeng

2024, 50(6): 30-35. doi: 10.13272/j.issn.1671-251x.2023120009

<Abstract>(266) <HTML> (55) <PDF>(44)

Abstract:
Currently, most multi-step hydraulic support pressure predictions are cumulative predictions of single step hydraulic support pressure. The more times a single step accumulates, the greater the cumulative error, which affects the prediction precision. In order to solve the above problems, a spatiotemporal multi-step prediction method of hydraulic support pressure based on long short term memory (LSTM)-Informer model is proposed. After using Kalman filtering to eliminate vibration noise in hydraulic support pressure data, two spatiotemporal datasets (Dataset 1 and Dataset 2) are established by selecting 5 adjacent hydraulic support pressure data at the end and middle of the working face. The spatiotemporal data is standardized and preprocessed. The method inputs spatiotemporal data into the LSTM model to extract spatiotemporal features, and inputs the extracted spatiotemporal features into the encoder of the Informer model. After position encoding, the method outputs multi head probability sparse self attention to focus on the changing features of the pressure sequence. After maximum pooling and one-dimensional convolution, the method eliminates the redundant combination of output feature map. By utilizing multi head probability sparse self attention to further focus on pressure sequence features, the decoder of the Informer model is changed to a fully connected layer to obtain the prediction results of hydraulic support pressure. The experimental results show that compared with prediction methods based on gated recurrent unit (GRU), LSTM, and Informer models, prediction methods based on LSTM-Informer model has the smallest root mean square error (RMSE) and mean absolute error (MAE) in predicting hydraulic support pressure at 6, 12, and 24 step sizes. The RMSE of the 6-step hydraulic support pressure predicted based on dataset 1 decreases by 41.63%, 49.74%, and 11.85%, and the MAE decreases by 41.75%, 50.00%, and 12.00%, respectively. The RMSE of the 6-step hydraulic support pressure predicted based on dataset 2 decreases by 48.15%, 59.86%, and 19.88%, and MAE decreases by 49.87%, 54.90%, and 13.16%, respectively.

Application research of all fiber optic microseismic monitoring technology in monitoring water inrush from floor

HUANG Gang, HAN Yunchun, YU Guofeng, LUO Yong, REN Bo, YE Zan, WANG Lichao, ZHAO Jing, XU Yifan

2024, 50(6): 36-45. doi: 10.13272/j.issn.1671-251x.2024030037

<Abstract>(240) <HTML> (32) <PDF>(26)

Abstract:
Currently, most fiber optic microseismic monitoring systems in China are based on optical grating sensing technology. However, fiber optic grating wavelength demodulation limits the detection frequency and sensitivity of the system, and there are few successful cases of long-term, continuous and uninterrupted microseismic monitoring. In order to solve the above problems, a new type of all fiber microseismic monitoring system is proposed. Taking the monitoring of water inrush from the floor during the mining process of Pan'er Coal Mine 11023 working face as the engineering background, a comparison is made between the all fiber optic microseismic monitoring system and the ESG microseismic monitoring system. It is found that the all fiber optic microseismic monitoring system has the following advantages. The recorded waveform spectrum features are clearer, showing a high signal-to-noise ratio advantage. The monitoring range for disturbance depth is larger, and the remote monitoring effect is better. The distribution of seismic source positioning results is more reasonable and more in line with the actual mining situation of the working face. During the monitoring of the entire mining cycle of the working face, the relationship between the floor failure and microseismic activity in the fault abnormal area of the 11023 working face is analyzed. Near the fault and coal seam thinning abnormal area, the number and intensity of microseismic events increase. During the initial mining period of the working face, stress is concentrated and released. Due to the influence of mining, the floor is severely damaged. Relatively high energy events are mainly distributed in the floor of the fault anomaly area, with a depth of about 27 meters of damage to the floor. Micro seismic events do not form a line or accumulate in a plane below 60 meters of the 3 coal seam floor. It indicates that cracks have not expanded and no water conducting channels have been formed. The working face can be safely mined.

Features and application of seismic-while-excavating signals during TBM excavation in coal mine rock roadways

DANG Baoquan, GUO Liquan, ZHANG Yanxi, REN Yongle, LI Shenglin

2024, 50(6): 46-53, 60. doi: 10.13272/j.issn.1671-251x.2024010094

<Abstract>(259) <HTML> (42) <PDF>(20)

Abstract:
The advanced seismic-while-excavating detection technology can achieve parallel exploration and excavation, providing the possibility of real-time and accurate geological support in the scenario of rapid and intelligent excavation of roadways. The signals generated by the excavation seismic source are complex, variable frequency, and continuous. The recognition of signal features directly affects the accuracy of data processing and imaging. However, currently, the recognition of seismic-while-excavating signal features for rock tunnel boring machine (TBM) is still unclear, and there is currently no targeted research on signal processing and imaging. In order to solve the above problems, taking the TBM advanced seismic-while-excavating detection test of the gas control roadway in Xieqiao Coal Mine as an example, the time domain, frequency domain, and frequency domain features of the cutterhead pilot signal and the rock wall received signal are analyzed. The proportion of different amplitude energy components in the rock roadway TBM seismic-while-excavating signal show a pyramid shape. But the distribution is random and the degree of asymmetry is high. The energy of the mechanical operation signal is relatively high, and the strength of the cutterhead pilot signal is about 200 times that of the signal received by the rock wall. The frequency domain frequency conversion features are obvious. The basic frequency of the mechanical operation signal is relatively low, and the frequency components of the cutterhead pilot signal are mainly concentrated in the range of 10-80 Hz and 150-200 Hz, with a main frequency of 36.99 Hz. The frequency components of the rock wall received signal are mainly concentrated in the range of 50-200 Hz, with a main frequency of 137.97 Hz. The frequency domain energy distribution of the cutterhead pilot signal is more regular than that of the rock wall received signal, and the phenomenon of multiple source excitation is obvious. The difference features between energy clusters indicate the randomness of amplitude energy and duration during multiple source excitations. The data processing and imaging experiments of TBM seismic-while-excavating signals in rock roadways are carried out using the pulse algorithm and diffraction stacking migration imaging method. The results show the following points. ① The pulse equivalent single shot record has strong consistency with the advanced detection single shot record obtained from conventional seismic-while-excavating sources, with clear and continuous in-phase axes, which can meet the needs of on-site detection analysis.② The advanced prediction results of the rock mass situation within the detection range are consistent with the actual exposure, indicating that TBM advanced seismic-while-excavating detection in rock roadways can provide effective geological support.

A method for constructing a knowledge graph of coal mine roof disaster prevention and control

LUO Xiangyu, DU Hao, HUA Ying, XIE Panshi, LYU Wenyu

2024, 50(6): 54-60. doi: 10.13272/j.issn.1671-251x.2023120032

<Abstract>(485) <HTML> (37) <PDF>(48)

Abstract:
At present, the decision-making of coal mine roof disaster prevention and control measures and the analysis of accident causes mainly rely on manual experience, and the level of intelligence is relatively low. The knowledge graph of roof disaster prevention and control can integrate knowledge and experience of roof disaster prevention and control, assist in analyzing the causes of roof disaster accidents and making decisions on roof disaster prevention and control measures. A method for constructing a knowledge graph of coal mine roof disaster prevention and control has been proposed. The ontology method is used to complete the knowledge modeling of coal mine roof disaster prevention and control. The concepts in the field of roof disaster prevention and control are divided into mine geology, mining technology, prevention and control measures, and accident characterization. The relationships between concepts are defined as usage, triggering, susceptibility, control, prevention, and applicability. The knowledge modeling lays the foundation for the knowledge extraction of coal mine roof disaster prevention and control (entity extraction and relationship extraction). Based on the characteristics of entity overlapping between a large number of nested entities and relationships in the field of coal mine roof disaster prevention and control, a span based entity extraction method and a dependency syntax tree guided entity representation based relationship extraction method are determined. The method constructs a corpus in the field of roof disaster prevention and control, and uses the Neo4j graph database to store data, providing data source support for the application of knowledge graph of roof disaster prevention and control. The partial construction results of the knowledge graph of coal mine roof disaster prevention and control are displayed. It indicates that this knowledge graph can assist in the analysis of roof disaster accident causes and decision-making of prevention and control measures, thereby improving the intelligence level of roof management. It is pointed out that based on this knowledge graph, combined with natural language processing and knowledge reasoning technologies, knowledge Q&A on roof management can be achieved.

Foreign object detection of coal mine conveyor belt based on improved YOLOv8

HONG Yan, WANG Lei, SU Jingming, WANG Hantao, LI Mushi

2024, 50(6): 61-69. doi: 10.13272/j.issn.1671-251x.2024050006

<Abstract>(289) <HTML> (53) <PDF>(52)

Abstract:
The existing deep learning based foreign object detection models for conveyor belts are relatively large and difficult to deploy on edge devices. There are errors and omissions in detecting foreign objects of different sizes and small objects. In order to solve the above problems, a foreign object detection method for coal mine conveyor belts based on improved YOLOv8 is proposed. The depthwise separable convolution, squeeze-and-excitation (SE) networks are used to reconstruct the Bottleneck of the C2f module in the YOLOv8 backbone network as a DSBlock, which improves the detection performance while keeping the model lightweight. To enhance the capability to obtain information from objects of different sizes, an efficient channel attention (ECA) mechanism is introduced. The input layer of ECA is subjected to adaptive average pooling and adaptive maximum pooling operations to obtain a cross channel interactive MECA module, which enhances the global visual information of the module and further improves the precision of foreign object recognition. The method modifies the 3 detection heads of YOLOv8 to 4 lightweight small object detection heads to enhance sensitivity to small objects and effectively reduce the missed and false detection rates of small object foreign objects. The experimental results show that the improved YOLOv8 achieves a precision of 91.69%, mAP@50 reached 92.27%, an increase of 3.09% and 4.07% respectively compared to YOLOv8. The detection speed of improved YOLOv8 reaches 73.92 frames/s, which can fully meet the demand for real-time detection of foreign objects on conveyor belts in coal mines. The improved YOLOv8 outperforms mainstream object detection algorithms such as SSD, Faster-RCNN, YOLOv5, and YOLOv7-tiny in terms of precision, mAP@50, number of parameters, weight size, and number of floating point operations.

Research on multivariate abnormal image detection in coal mine transportation system

LYU Donghan, HU Eryi, HUANG Yipo, LI Wenzhang

2024, 50(6): 70-78. doi: 10.13272/j.issn.1671-251x.2024050001

<Abstract>(80) <HTML> (24) <PDF>(11)

Abstract:
There are various types and scenarios of abnormal risks in coal mine transportation systems. The occurrence of abnormal accidents at coal mine sites is accidental, and the number of abnormal samples obtained is much smaller than that of normal samples, resulting in an imbalance of positive and negative samples. In order to solve the above problems, a multivariate abnormal image detection method for coal mine transportation systems based on hypersphere reconstructed data description (HRDD) is proposed. On the basis of full convolutional data description (FCDD), an image reconstruction auxiliary task is introduced. The mean square error loss function is selected as the objective function of the image reconstruction auxiliary task. Abnormal image detection and positioning are quantified as an inequality constrained optimization problem. The seamless fusion technology is used to fuse auxiliary datasets and abnormal samples into normal samples, in order to reduce the difference between abnormal fusion samples and normal samples, expand the total number of abnormal samples, and balance the proportion of abnormal and normal samples. Through multiple sets of noise simulation experiments and on-site experiments, it has been proven that adding Gaussian noise to the resistance zone with a certain probability for enhanced training can improve the noise resistance efficiency, generalization capability, detection accuracy, and other aspects of the HRDD model. The results of the ablation experiment show that the auxiliary dataset effectively improves the problem of sample imbalance, with an accuracy increase of 36.5%. The introduction of image reconstruction auxiliary tasks can ensure that deep features can be accurately mapped to abnormal positions, resulting in an IoU improvement of 33.4%. There is a strong coupling effect between the auxiliary dataset and the image reconstruction auxiliary task. The combination of the two can further stimulate the performance potential of the HRDD algorithm. The addition of seamless fusion samples and Gaussian noise enhancement has to some extent improved the generalization capability of the HRDD model. The comparative experimental results show that the accuracy and IoU of the HRDD algorithm are better than other mainstream algorithms. Compared with the FCDD algorithm, the accuracy and IoU of the HRDD algorithm have increased by 4.6% and 7.0% respectively, making it more suitable for coal mine sites.

A coal mine underground image enhancement method based on multi-scale gradient domain guided image filtering

MU Qi, GE Xiangfu, WANG Xinyue, LI Lei, LI Zhanli

2024, 50(6): 79-88, 111. doi: 10.13272/j.issn.1671-251x.2023080126

<Abstract>(112) <HTML> (29) <PDF>(16)

Abstract:
There are serious issues with uneven lighting and noise interference in coal mine underground images. The existing Retinex based methods are directly applied to enhance coal mine underground images, which are prone to problems such as halo artifacts, blurred edges, over enhancement, and noise amplification. In order to solve the above problems, a coal mine underground image enhancement method based on multi-scale gradient domain guided image filtering is proposed. Firstly, the multi-scale idea is introduced into gradient domain guided image filtering to achieve accurate estimation of non-uniform lighting, effectively solving the problems of halo artifacts and edge blurring in enhanced images. Secondly, the Retinex model is used to separate the lighting component and reflection component. For the lighting component, the lighting information is corrected pixel by pixel through an adaptive gamma correction function, which enhances the dark areas of the image while suppressing the over enhancement of the bright areas. The image contrast is adjusted using a contrast limited adaptive histogram equalization method. For the reflection component, gradient domain guided image filtering is combined with multi-scale detail enhancement to accurately remove noise and improve texture details, avoiding the problem of noise amplification during image enhancement. Finally, the processed lighting and reflection components are fused, and the image gain coefficient is calculated. The linear color restoration method is used to enhance the original RGB image pixel by pixel, improving the processing efficiency of the method. The experimental results show that, from a subjective and objective perspective, compared with existing methods, the images processed by the proposed method have achieved better enhancement effects in color preservation, contrast, noise suppression, detail preservation, and other aspects, while also having higher processing efficiency.

A defogging algorithm for coal mine underground images based on dark channel guided filtering and lighting correction

GAI Yonggang

2024, 50(6): 89-95. doi: 10.13272/j.issn.1671-251x.2024030048

<Abstract>(86) <HTML> (31) <PDF>(10)

Abstract:
The existing methods for defogging coal mine images fail to perform lighting correction while extracting deep level feature information, resulting in the loss of detail information or image darkening in the processed images. A defogging algorithm for coal mine underground images based on dark channel guided filtering and lighting correction is proposed. Firstly, the original underground images are subjected to bilateral filtering, lighting estimation, and dark primary color processing using the image differentiation module (IDM) to obtain lighting maps, dark primary color maps, and lighting reflection maps. Secondly, the method preprocesses the dark primary color map and uses it as a weight guidance parameter to guide the filtering of the lighting reflection map, in order to restore the image's detailed feature information. Finally, the lighting map is used as a weight parameter to perform lighting correction and feature extraction on the image. The color distortion problem is solved through multiple lighting corrections, while increasing the network depth to remove degradation in dark areas, achieving reconstruction of image details and obtaining clear images. The subjective evaluation results indicate that the coal mine underground image defogging algorithm based on dark channel guided filtering and lighting correction retains more structural textures and background details while removing fog. It makes the entire image closer to the corresponding clear image. The objective evaluation results show that compared with the suboptimal algorithm PMS-Net, the information entropy on the training and testing sets is increased by 0.32 and 0.11, the standard deviation is increased by 3.58 and 1.89, and the average gradient is increased by 0.008 and 0.004, respectively. This indicates that the proposed algorithm can effectively reduce the fog in coal mine underground images. The results of ablation experiments show that the proposed algorithm has higher information entropy, standard deviation, and average gradient on the test dataset than other network models. It indicates that the defogging effect is the best and it can effectively preserve image details and edge information.

A fusion positioning method for underground personnel based on UWB and PDR

JIA Yutao, LI Guanhua, PAN Hongguang, CHEN Haijian, WEI Xuqiang, BAI Junming

2024, 50(6): 96-102, 135. doi: 10.13272/j.issn.1671-251x.2024010071

<Abstract>(131) <HTML> (43) <PDF>(17)

Abstract:
Most existing fusion positioning methods for ultra-wideband (UWB) and pedestrian dead reckoning (PDR) ignore the correction of positioning errors in non line of sight (NLOS) environments. The methods use simple threshold division as the basis for NLOS environment judgment, which is largely related to the positioning scene and site size. In order to solve the above problems, a fusion positioning method for underground personnel considering NLOS environment is proposed. Firstly, UWB technology is used to calculate the position of underground personnel. After obtaining the preliminary position of personnel through the trilateral positioning algorithm, the least squares method is used to optimize the position. Polynomial fitting is used to achieve the fitting between the actual value and the measured value between the base station and the tag in the NLOS environment, reducing the ranging error in the NLOS environment and improving the positioning precision. Secondly, the PDR algorithm is used for gait recognition and analysis. The PDR algorithm uses gait data collected by inertial navigation sensors to update the target position through gait recognition, step size estimation, and direction estimation. Thirdly, the convolutional neural network (CNN) - long short term memory (LSTM) network is used to analyze the features of channel impulse response (CIR) and achieve line of sight (LOS)/NLOS recognition. It solves the problem of scene limitations in NLOS environment judgment. Finally, the fusion coefficient is determined based on the LOS/NLOS recognition results to achieve the fusion of UWB and PDR positioning results. The experimental results show that after polynomial fitting, the average ranging error of UWB is reduced by 0.59 m. The average accuracy of LOS/NLOS recognition is 95.3%, and the recall rate and F₁ score are both above 90%, verifying that CNN-LSTM has good recognition performance. The average error of the fusion positioning method is 0.31 m, which is 1.57 m lower than UWB and 1.41 m lower than PDR.

A method for estimating the step size of underground personnel based on generative adversarial networks

WANG Taiji

2024, 50(6): 103-111. doi: 10.13272/j.issn.1671-251x.2024020039

<Abstract>(105) <HTML> (26) <PDF>(10)

Abstract:
In response to the problems of cumulative errors in step size estimation and the large sample size required by traditional deep learning methods in the pedestrian dead reckoning (PDR) based underground personnel positioning system in coal mines, a step size estimation method for underground personnel based on generative adversarial network (GAN) is proposed. The GAN model mainly includes two parts: generative model and discriminative model, both of which are implemented using deep neural networks (DNNs). The generative model aims to generate continuous result distributions (i.e. labels) based on input data. Its output layer uses a linear activation function to preserve the linear features of the network, allowing the model to predict the step size of any personnel during walking. The discriminant model aims to distinguish whether the input data and labels are real labels or labels generated by the generator. Its output layer uses a Sigmoid activation function to achieve binary classification of results. After determining the generative model and discriminant model, the GAN model combines two models for training. By constructing and optimizing the dynamic competition between the generator and discriminator, the generator can learn to generate more realistic and indistinguishable data samples in continuous iterations. The experimental results show that under the same training and testing sets, the average error of the GAN model is 0.14 m, and the standard deviation and root mean square error are both smaller than those of the DNNs model, with the minimum values being 0.74 m. The outdoor test results show that the GAN based underground personnel step estimation method has a minimum error of 3.21% and a maximum error of 4.79% in uphill and downhill scenarios. Compared to uphill and downhill scenarios, the error in playground scenarios is smaller, with a maximum error of 1.91%.

Path planning of coal mine foot robot by integrating improved A^* algorithm and dynamic window approach

WANG Limin, SUN Ruifeng, ZHAI Guodong, ZHANG Jiawei, XU Hong, ZHAO Jie, HUA Yihang

2024, 50(6): 112-119. doi: 10.13272/j.issn.1671-251x.2024020042

<Abstract>(136) <HTML> (27) <PDF>(23)

Abstract:
In order to improve the operational efficiency, search precision, and obstacle avoidance flexibility of the path planning algorithm for coal mine foot robot, a path planning method for coal mine foot robots is proposed, which integrates the improved A^* algorithm and the dynamic window approach (DWA). Firstly, the A^* algorithm is improved by reducing the length of the planned path through a redundant node removal strategy. The method improves the neighborhood search method and cost function to increase the speed of path planning, and uses segmented second-order Bessel curves for path smoothing. The path nodes planned by the improved A^* algorithm are sequentially used as local target points for local path planning DWA for algorithmic fusion. The method filters neighboring obstacle nodes to shorten the path length again, and improves obstacle avoidance performance by adjusting the weight ratio in the DWA cost function. In response to the problem of robots falling into a "feigned death" state when encountering unavoidable obstacles, the method starts from the current initial point, the fusion algorithm is called up again. The global path planning is carried out again, and the new nodes obtained replace the original local target points, and the subsequent work is carried out according to the new route. The simulation results show that, while ensuring the safety and stability of robot walking, the improved A^* algorithm reduces the calculation time by 65%, the path length by 24.1%, and the number of path nodes by 27.65% compared to the traditional A^* algorithm, resulting in a smoother path. The fusion algorithm further enhances the global path planning capability, enabling it to bypass newly added dynamic and static obstacles in multi obstacle environments. When the robot encounters an L-shaped obstacle and enters a "feigned death" state, it reconducts global path planning at the "feigned death" position, updates its walking path, and successfully reaches the final target point. The experimental results of path planning for JetHexa hexapod robot based on fusion algorithm have verified the effectiveness and superiority of the fusion algorithm.

A method for completing coal wall point cloud in fully mechanized working face based on residual optimization

WANG Weibing, HOU Xueqian, ZHAO Shuanfeng, HE Haitao, XING Zhizhong, LU Zhengxiong

2024, 50(6): 120-128. doi: 10.13272/j.issn.1671-251x.2024020014

<Abstract>(82) <HTML> (34) <PDF>(11)

Abstract:
The digital 3D reconstruction process of coal mine fully mechanized working face roadways requires complete and dense coal wall point cloud data. Due to factors such as occlusion and limited viewing angle, the collected coal wall point cloud data of the fully mechanized working face is often incomplete and sparse, which affects downstream tasks and requires coal wall point cloud repair and completion. At present, there is a lack of datasets and network models for underground point cloud completion tasks. Existing models used for coal wall point cloud completion suffer from uneven distribution of point cloud density and loss of point cloud feature information. In order to solve the above problems, a coal wall point cloud completion network model based on residual optimization is designed. Supervised learning is used to learn point cloud feature information, and the complete point cloud is output by minimizing density sampling and iteratively optimizing the residual network. The method collects real coal wall point cloud data of fully mechanized working face underground, preprocesses and screens available data. The method creates a coal wall point cloud missing dataset by simulating random cavities. The missing dataset is used to train the residual optimization-based coal wall point cloud complementary network model. The experimental results show that compared with the classic FoldingNet, TopNet, AtlasNet, PCN, and 3D-Capsule point cloud completion network models, the residual optimization-based coal wall point cloud completion network model achieves the optimal level of chamfer distance, ground shift distance, and F₁ score for the constructed missing and sparse coal wall point clouds, with the best overall completion effect. It is able to achieve effective completion for the actual missing coal wall point clouds.

Design of height measurement sensor for hydraulic support in fully mechanized working face

GAO Siwei, GU Minyong, LI Dianpeng

2024, 50(6): 129-135. doi: 10.13272/j.issn.1671-251x.2024010089

<Abstract>(104) <HTML> (29) <PDF>(17)

Abstract:
A hydraulic support height measueement sensor based on Pascal's law is designed to address the difficulties in measuring the height, reliability, and stability of hydraulic supports in fully mechanized working faces. The height measurement sensor is equipped with a slender liquid tube, which is filled with methyl silicone oil. Pressure sensors are installed at both ends of the height measuremenr sensor. By measuring the pressure at both ends of the sealed liquid tube, the height difference between the two ends of the height measuremenr sensor is obtained. Methyl silicone oil has obvious thermal expansion and contraction characteristics when the ambient temperature changes, which can cause a sharp change in pressure in a closed space and affect measurement precision. A compensation method is proposed. The method stores a portion of methyl silicone oil in a corrugated tube, and uses the elasticity of the corrugated tube itself to compensate for the volume change caused by thermal expansion and contraction of methyl silicone oil. The method calibrates the density change caused by the volume change of methyl silicone oil through algorithms in the software to ensure the measurement precision of the height measurement sensor. The test results show that the height measuremenr sensor can operate in an environment of 0-40 ℃, with a measurement error of less than 4 cm. The on-site application results of the hydraulic support in the fully mechanized working face of thin and medium thick coal seams show that compared with manual measurement results, the error of the height measurement sensor is within 5 cm, indicating high reliability of the sensor.

Precise and fast digital twin mapping method for hydraulic support attitude

LIU Meng, FU Xiang, JIANG Yulong, LIU Bin, YANG Yuqi, QIN Yifan, SUN Yan

2024, 50(6): 136-141, 158. doi: 10.13272/j.issn.1671-251x.18180

<Abstract>(126) <HTML> (34) <PDF>(18)

Abstract:
A precise and fast digital twin mapping method for hydraulic support attitude in fully mechanized working face is proposed to address the problems of low precision, large time delay, and difficulty in balancing precision and time delay in implementing attitude mapping using digital twins. The study designs a digital twin system architecture for hydraulic support attitude, which includes physical perception layer, data layer, business logic layer, and presentation layer. At the business logic layer, by retaining the external shape of the high-precision model of the hydraulic support and merging key parts into lightweight structural components, a digital twin model of the hydraulic support attitude is established. It reduces rendering delay. By establishing a conversion relationship between the angle measured by the inclination sensor and the rotation angle of the digital twin model, consistent of virtual-real mapping of the digital twin model of the hydraulic support attitude and real entities of the hydraulic support is achieved, ensuring the precision of the hydraulic support attitude mapping. At the data layer, transmission intervals are divided from each data transmission link and parameter constraints are applied. Parameter programming is applied to optimize the data update interval of adjustable links, reducing resource waste and latency in the data transmission process of digital twin systems. A precise and fast digital twin mapping platform for hydraulic support attitude is built, and the time delay and precision of virtual real mapping for hydraulic support attitude are tested. The results show that this method has a lower time delay while ensuring the precision of hydraulic support attitude mapping.

Diagnosis method for planetary gear faults in shearer under strong noise interference

LI Yong, ZHANG Qizhi, ZHUANG Deyu, QIU Jinbo, CHENG Gang

2024, 50(6): 142-149. doi: 10.13272/j.issn.1671-251x.18177

<Abstract>(107) <HTML> (33) <PDF>(23)

Abstract:
The health status of the planetary gears in the cutting section of shearer's rocker arm directly affects the cutting efficiency. The strong noise interference caused by multiple impacts during the cutting of coal and rock by the shearer, the complex gear structure, and the variable transmission path make it difficult to extract fault features. In order to solve the above problems, a fault diagnosis method for planetary gears in shearer based on spectral average denoising and correlation spectrum is proposed. Based on the distribution features of signal spectrum and the random features of noise, the spectrum average denoising method is adopted to suppress the interference of noise on the signal spectrum and obtain the signal denoising spectrum. The method constructs relevant spectra to establish the intrinsic relationship between few sample denoising spectra and multi sample denoising spectra, and reduce the demand for sample size for average spectrum denoising. The method uses a one-dimensional convolutional neural network (1D CNN) to establish an accurate mapping relationship between correlation spectra and fault categories, with correlation spectra as input and fault categories as output, to achieve planetary gear fault classification and recognition. The experimental verification of the fault diagnosis method for planetary gears in shearer based on spectral average denoising and correlation spectrum is carried out on the drivetrain diagnostics simulator transmission system fault diagnosis experimental platform. The results show that the method can enhance the key frequency that characterizes the fault features. The overall recognition rate for five types of health status signals of planetary gears, including normal, broken teeth, wear, missing teeth, and cracks, reaches 96%. Gear fault diagnosis can be effectively and accurately achieved when the signal-to-noise ratio is not less than 15 dB.

Analysis of dynamic and static features of intrinsically safe electromagnetic valves and optimization of influencing parameters

YAO Zhuo, WANG Wei, WEI Wenshu, LU Delai, LI Xiangbo

2024, 50(6): 150-158. doi: 10.13272/j.issn.1671-251x.2024020048

<Abstract>(97) <HTML> (27) <PDF>(8)

Abstract:
In response to the problems of insufficient electromagnetic driving force and slow response speed of electromagnetic valves under the constraints of driving power and electromagnetic valve volume, the dynamic and static features of electromagnetic valves are analyzed. Through simulation analysis and prototype experiments, it is verified that improving the intrinsically safe electromagnetic force can significantly improve the response features of electromagnetic valves. A plan to improve the response features of electromagnetic valves by optimizing the electromagnetic force is determined. The evaluation indicators for the features of intrinsically safe electromagnets, including effective stroke index, average electromagnetic force index, and static comprehensive performance index, have been proposed to solve the problem of difficult performance evaluation of electromagnets caused by different stroke. The Maxwell electromagnetic simulation software is used to analyze the effects of changes in guide tube depth, armature radius, non working air gap, non working air gap, and pot mouth height on the static features of the electromagnet. The sensitivity of different parameters to the static features is obtained, providing a reference for selecting the size control range in parameter optimization. A second-order response surface model is constructed based on the results of orthogonal experiments to evaluate the comprehensive features of electromagnets using iron core structural parameters. Genetic algorithm is used to optimize the iron core parameters. The prototype test results show that the optimized electromagnetic force in the horizontal section of the electromagnet has increased by 56%, the effective stroke has increased by 26%, and the opening response time of the intrinsically safe electromagnetic valve has been shortened by 52.5%.

2024 Vol. 50, No. 6

Submission System

Latest AcceptedMore>