Mining data monitoring optimization method, device, equipment and medium based on 5G communication
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- CHINALCO ZHONGZHOU MINING CO LTD
- Filing Date
- 2026-03-13
- Publication Date
- 2026-06-16
Smart Images
Figure CN122221148A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of mining operation monitoring, and in particular to a mining data monitoring optimization method, apparatus, equipment, and medium based on 5G communication. Background Technology
[0002] In the field of 5G communication and mining operation monitoring, existing technologies have gradually adopted wireless sensor networks to collect and transmit underground environmental data. By deploying multiple fixed dust and vibration sensors and transmitting data back to the ground control center via 5G networks, remote monitoring of underground environmental parameters has been achieved. This technology utilizes the low latency of 5G to meet real-time requirements to a certain extent, while improving the reliability of monitoring through multi-sensor data fusion. Existing research has also explored the application of millimeter-wave radar in dusty environments, such as using traditional phased array radar for non-contact dust concentration detection, providing new technical means for underground environmental monitoring.
[0003] However, existing technologies still face key bottlenecks in achieving real-time monitoring of data throughout the entire process. Current systems lack sufficient dynamic adaptability to dusty environments, especially in mining areas where dust concentrations fluctuate drastically. The measurement accuracy of traditional fixed sensors is easily affected by interference, and they cannot acquire key parameters such as dust particle size distribution in real time, thus limiting the accuracy of environmental risk assessment. Furthermore, existing 5G communication systems lack dynamic optimization mechanisms for channel coding strategies in high-dust underground environments, making it difficult to balance data transmission reliability and spectral efficiency. Summary of the Invention
[0004] In view of this, the purpose of this invention is to provide a mining data monitoring and optimization method, device, equipment, and medium based on 5G communication. This method can construct a closed-loop optimization system from environmental sensing to data transmission, enabling the device to accurately acquire and stably transmit dust status and equipment operation data throughout the entire mining process. The specific solution is as follows: In a first aspect, this application discloses a mining data monitoring optimization method based on 5G communication, applied to a data monitoring device, comprising: Collect dust concentration data of each section of the mine roadway, determine the dust concentration vector based on the dust concentration data, and obtain the dust particle size distribution matrix of each section of the mine roadway based on the preset radar equipment, and output the raw point cloud data based on the dust particle size distribution matrix. The number of information bits of the polar code and the number of iterations of the linear block code are determined based on the dust concentration vector, and the original point cloud data is processed by a preset clustering algorithm to obtain dust particle size characteristics and spatial distribution heat map. Polarization configuration instructions are generated based on the dust particle size characteristics and the spatial distribution heat map. Enhanced point cloud data is output according to the polarization configuration instructions. The enhanced point cloud data, the dust concentration vector, and the mining equipment vibration data are fused through preset edge computing nodes to obtain a joint state matrix. A target 5G channel is constructed based on the number of information bits and the number of iterations of the linear block code. The joint state matrix is then uploaded to the target cloud for state analysis via the target 5G channel. Finally, the scanning frequency of the data monitoring device is optimized through feedback control measures and the spatial distribution heat map.
[0005] Optionally, the step of collecting dust concentration data from each zone of the mine roadway to determine a dust concentration vector based on the dust concentration data includes: The original light intensity signal is output to each section of the mine roadway by a preset laser device, and the scattered laser signal scattered by dust is received, so as to output a current signal group based on the scattered laser signal. The current signal group is amplified and filtered to obtain the current signal to be converted, and the current signal to be converted is converted into a standard voltage signal, and then the standard voltage signal is converted into the corresponding target digital signal. The target digital signal is processed based on Mie scattering to obtain dust concentration data for each section of the mine roadway. The dust concentration data is then denoised and vectorized to obtain the corresponding dust concentration vector.
[0006] Optionally, the step of acquiring the dust particle size distribution matrix of each zone of the mine roadway based on a preset radar device, and outputting raw point cloud data based on the dust particle size distribution matrix, includes: A radio frequency signal is output through a preset signal generator, and the radio frequency signal is converted into a corresponding electromagnetic wave. Then, the echo signal of the electromagnetic wave scattered by the dust is captured. The echo signal is quadrature demodulated to obtain the corresponding baseband signal, and the baseband signal is subjected to fast Fourier transform to generate the corresponding target spectrum. Based on the target spectrum, a dust particle size distribution matrix for each zone of the mine roadway is generated, and the original point cloud data is output based on the dust particle size distribution matrix.
[0007] Optionally, the step of determining the number of information bits of the polar code and the number of iterations of the linear block code based on the dust concentration vector, and processing the original point cloud data through a preset clustering algorithm to obtain dust particle size characteristics and a spatial distribution heatmap, includes: Based on the dust concentration vector, the current channel quality index is output, and the bit information corresponding to the channel quality index is queried in the preset mapping table to obtain the number of initial information bits of the polar code. The initial number of information bits is corrected based on the historical transmission bit error rate to obtain the target number of information bits, and the number of iterations is calculated based on the target number of information bits and the real-time signal-to-noise ratio to obtain the number of iterations for the linear block code. Outlier removal and downsampling are performed on the original point cloud data to obtain preprocessed point cloud data. The preprocessed point cloud data is then clustered using a preset clustering algorithm to output dust particle cluster labels for several dust particle clusters corresponding to the preprocessed point cloud data. The equivalent volume diameter of each particle cluster in the plurality of dust particle clusters is determined, and the equivalent volume diameter of each particle cluster is used as the dust particle size distribution array. The dust particle size distribution array is then divided into grids according to the tunnel space to obtain the corresponding three-dimensional grid data. A dust concentration gradient map is generated based on the dust particle size distribution array and the three-dimensional grid data, and a corresponding spatial distribution heat map is generated based on the dust concentration gradient map.
[0008] Optionally, the step of generating polarization configuration instructions based on the dust particle size characteristics and the spatial distribution heatmap, and outputting enhanced point cloud data according to the polarization configuration instructions, includes: The basic polarization angle is calculated based on the dust particle size characteristics and the spatial distribution heat map, and the basic polarization angle is corrected according to the concentration gradient in the spatial distribution heat map to obtain the optimized polarization angle. The optimized polarization angle is converted into a corresponding phase control codeword, and the radiation beam is reconstructed according to the phase control codeword to obtain the reconstructed radiation beam signal. Environmental scanning is performed based on the reconstructed radiation beam signal, and enhanced echo signals are received. Then, three-dimensional imaging processing is performed based on the enhanced echo signals to obtain enhanced point cloud data.
[0009] Optionally, the step of fusing the enhanced point cloud data, the dust concentration vector, and the mining equipment vibration data through preset edge computing nodes to obtain a joint state matrix includes: Vibration data of mining equipment is collected in real time, and a raw data stream is generated based on the enhanced point cloud data, the dust concentration vector and the vibration data of mining equipment. The raw data stream is then timestamped to obtain the corresponding synchronized data group. The spatial coordinates corresponding to the synchronous data set are transformed to the roadway reference coordinate system to obtain a standardized data set, and the dust concentration gradient and equipment vibration spectrum characteristics are extracted from the standardized data set. Based on the dust concentration gradient and the vibration spectrum characteristics of the equipment, a corresponding joint feature vector is generated, and the joint feature vector is spatiotemporally calibrated to obtain a synchronized data set. The feature weights of the synchronous data set are adjusted based on the adjusted dust concentration gradient in the synchronous data set to obtain a weighted feature matrix; Spatial features are extracted from the weighted feature matrix to obtain spatiotemporal feature blocks, and corresponding state codes are generated based on the spatiotemporal feature blocks. Then, the state codes are mapped to the corresponding joint state matrix.
[0010] Optionally, the step of uploading the joint state matrix to the target cloud via the target 5G channel for state analysis, and then optimizing the scanning frequency of the data monitoring device through feedback control measures and the spatial distribution heat map, includes: The joint state matrix is uploaded to the target cloud via the target 5G channel, so that the target cloud can generate a corresponding risk feature vector based on the feature parameters decoded from the joint state matrix, and input the risk feature vector into a preset model to output a corresponding risk score; based on the risk score and historical maintenance records, a corresponding risk level is generated, and control measures corresponding to the risk level are matched from a preset database, and then the control measures are fed back to the data monitoring device. The received control measures are executed, and the corresponding instruction response delay and execution success rate are determined according to the corresponding execution status. Based on the spatial distribution heat map, dust accumulation areas are determined, and time slot allocation weights are determined by the command response delay and the accumulation areas. The optimal scanning frequency for the radar device is determined based on the time slot allocation weight, the execution success rate, and the aggregation area, and the data acquisition frequency of the radar device is optimized based on the optimal scanning frequency.
[0011] Secondly, this application discloses a mining data monitoring and optimization device based on 5G communication, applied to a data monitoring device, comprising: The point cloud data determination module is used to collect dust concentration data of each section of the mine roadway, determine the dust concentration vector based on the dust concentration data, and obtain the dust particle size distribution matrix of each section of the mine roadway based on the preset radar equipment, and output the original point cloud data based on the dust particle size distribution matrix. The point cloud data processing module is used to determine the number of information bits of the polar code and the number of iterations of the linear block code based on the dust concentration vector, and to process the original point cloud data through a preset clustering algorithm to obtain dust particle size characteristics and spatial distribution heat map. The data fusion module is used to generate polarization configuration instructions based on the dust particle size characteristics and the spatial distribution heat map, output enhanced point cloud data according to the polarization configuration instructions, and fuse the enhanced point cloud data, the dust concentration vector and the mining equipment vibration data through a preset edge computing node to obtain a joint state matrix. The data monitoring and optimization module is used to construct a target 5G channel based on the number of information bits and the number of linear block code iterations, and upload the joint state matrix to the target cloud for state analysis through the target 5G channel. Then, the scanning frequency of the data monitoring device is optimized through feedback control measures and the spatial distribution heat map.
[0012] Thirdly, this application discloses an electronic device, including: Memory, used to store computer programs; A processor is used to execute the computer program to implement the aforementioned 5G communication-based mining data monitoring optimization method.
[0013] Fourthly, this application discloses a computer-readable storage medium for storing a computer program, wherein the computer program, when executed by a processor, implements the aforementioned mining data monitoring optimization method based on 5G communication.
[0014] In this application, dust concentration data of each zone in the mine roadway can be collected to determine a dust concentration vector based on the dust concentration data. A dust particle size distribution matrix of each zone in the mine roadway can be obtained based on a preset radar device, and raw point cloud data can be output based on the dust particle size distribution matrix. The number of information bits of the polarization code and the number of iterations of the linear block code are determined based on the dust concentration vector. The raw point cloud data is processed using a preset clustering algorithm to obtain dust particle size characteristics and a spatial distribution heatmap. A polarization configuration instruction is generated based on the dust particle size characteristics and the spatial distribution heatmap. Enhanced point cloud data is output based on the polarization configuration instruction. The enhanced point cloud data, the dust concentration vector, and the vibration data of the mining equipment are fused using a preset edge computing node to obtain a joint state matrix. A target 5G channel is constructed based on the number of information bits and the number of iterations of the linear block code. The joint state matrix is uploaded to a target cloud for state analysis via the target 5G channel. Then, the scanning frequency of the data monitoring device is optimized using feedback control measures and the spatial distribution heatmap.
[0015] Therefore, the method of this application can determine the dust concentration vector and original point cloud data based on dust data from the mining area; determine the number of information bits of the polar code and the number of iterations of the linear block code based on the dust concentration vector, and construct the target 5G channel based on these two parameters; obtain the dust particle size characteristics and spatial distribution heat map from the original point cloud data, and determine the enhanced point cloud data based on these two parameters; fuse the enhanced point cloud data, dust concentration vector, and mining equipment vibration data through preset edge computing nodes, and upload the fused joint state matrix to the cloud for analysis through the target 5G channel; and optimize the current scanning frequency through feedback control measures and spatial distribution heat map. In this way, a closed-loop optimization system from environmental perception to data transmission can be constructed, enabling the device to accurately acquire and stably transmit dust status and equipment operation data throughout the entire mining process. Attached Figure Description
[0016] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.
[0017] Figure 1 This application discloses a flowchart of a mining data monitoring optimization method based on 5G communication. Figure 2 This is a schematic diagram of the processing flow of a mining data monitoring and optimization method based on 5G communication disclosed in this application; Figure 3 This is a flowchart of the first specific mining data monitoring and optimization method based on 5G communication disclosed in this application; Figure 4 This is a flowchart of the second specific mining data monitoring optimization method based on 5G communication disclosed in this application; Figure 5 This is a flowchart of the third specific mining data monitoring optimization method based on 5G communication disclosed in this application; Figure 6 This is a flowchart of the fourth specific mining data monitoring optimization method based on 5G communication disclosed in this application; Figure 7 This is a schematic diagram of a mining data monitoring and optimization device based on 5G communication disclosed in this application; Figure 8 This is a structural diagram of an electronic device disclosed in this application. Detailed Implementation
[0018] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.
[0019] Currently, key bottlenecks remain in achieving real-time monitoring of the entire process. Existing systems lack dynamic adaptability to dusty environments, especially in mining areas where dust concentration fluctuates drastically. The measurement accuracy of traditional fixed sensors is easily affected by interference, and they cannot acquire key parameters such as dust particle size distribution in real time, thus limiting the accuracy of environmental risk assessment. Furthermore, existing 5G communication systems lack dynamic optimization mechanisms for channel coding strategies in high-dust underground environments, making it difficult to balance data transmission reliability and spectral efficiency.
[0020] To overcome the aforementioned technical problems, this application discloses a mining data monitoring and optimization method, device, equipment, and medium based on 5G communication. This method can construct a closed-loop optimization system from environmental perception to data transmission, enabling the device to accurately acquire and stably transmit dust status and equipment operation data throughout the entire mining process.
[0021] See Figure 1 As shown, this embodiment of the invention discloses a mining data monitoring optimization method based on 5G communication, applied to a data monitoring device, including: Step S11: Collect dust concentration data of each section of the mine roadway, determine the dust concentration vector based on the dust concentration data, and obtain the dust particle size distribution matrix of each section of the mine roadway based on the preset radar equipment, and output the original point cloud data based on the dust particle size distribution matrix.
[0022] In this embodiment, as Figure 2As shown, the process begins with collecting dust concentration data via the data acquisition module of the data monitoring device. Then, the dust concentration data is processed to obtain the corresponding vector. Specifically, a pre-set laser device outputs raw light intensity signals to each section of the mine roadway and receives scattered laser signals from the dust, outputting a current signal group based on these scattered laser signals. Specifically, the laser emitting unit emits a 650nm laser beam to each section of the roadway, outputting the raw light intensity signal. The photoelectric sensor array receives the scattered laser signals from the dust and outputs a current signal group. Next, the current signal group needs to be amplified and filtered to obtain the current signal to be converted. This current signal is then converted into a standard voltage signal, and finally, the standard voltage signal is converted into the corresponding target digital signal. Specifically, the signal conditioning circuit amplifies and filters the current signal group, outputting a standardized voltage signal. The AD (Analog-to-Digital) conversion module converts the standardized voltage signal into a digital signal. Finally, the target digital signal is processed based on Mie scattering to obtain dust concentration data for each zone of the mine roadway. The dust concentration data is then denoised and vectorized to obtain the corresponding dust concentration vector. Specifically, a microprocessor uses Mie scattering theory to obtain the concentration value of the digital signal, outputting the dust concentration data for each zone. Then, a Kalman filter is used to denoise the dust concentration data for each zone, outputting a standardized dust concentration vector.
[0023] Furthermore, such as Figure 2As shown, the dust particle size distribution matrix needs to be obtained by scanning with appropriate radar equipment, and then the raw point cloud data is output. Specifically, firstly, a radio frequency (RF) signal needs to be output through a preset signal generator and converted into a corresponding electromagnetic wave. Then, the echo signal of the electromagnetic wave scattered by the dust is captured. That is, a 60GHz frequency-modulated continuous wave can be emitted by a millimeter-wave signal generator to output an RF signal. Then, a metasurface antenna array converts the RF signal into a polarization-tunable electromagnetic beam, and a dust reflection signal receiver captures the echo signal scattered by the dust. Further, the echo signal needs to be orthogonally demodulated to obtain the corresponding baseband signal, and a fast Fourier transform is performed on the baseband signal to generate the corresponding target spectrum. Then, the dust particle size distribution matrix of each section of the mine roadway is generated based on the target spectrum, and the raw point cloud data is output based on the dust particle size distribution matrix. That is, the wave signal is orthogonally demodulated by the IQ (In-phase / Quadrature) demodulation module to output the baseband signal, and the FFT (Fast Fourier Transform) processor performs fast Fourier transform on the baseband signal to generate the range-Doppler spectrum (target spectrum). The point cloud generation unit inverts the three-dimensional spatial coordinates based on the range-Doppler spectrum and outputs the original point cloud data.
[0024] Step S12: Determine the number of information bits of the polar code and the number of iterations of the linear block code based on the dust concentration vector, and process the original point cloud data through a preset clustering algorithm to obtain the dust particle size characteristics and spatial distribution heat map.
[0025] In this embodiment, as Figure 2 As shown, the dust concentration vector needs to be dynamically calculated by the adaptive coding control module in the data monitoring device. Then, the original point cloud data is processed by a clustering algorithm to output dust particle size characteristics and a spatial distribution heatmap. Specifically, firstly, the current channel quality index needs to be output based on the dust concentration vector, and the corresponding bit information is looked up in a preset mapping table to obtain the initial information bit count for the polar code. That is, the channel state interface receives the standardized dust concentration vector, outputs the current channel quality index, and the Polar code configuration unit looks up the preset mapping table based on the current channel quality index to output the initial information bit count. Then, the initial information bit count needs to be corrected based on the historical transmission error rate to obtain the target information bit count, and the number of iterations is calculated based on the target information bit count and the real-time signal-to-noise ratio to obtain the linear block code iteration count. That is, the dynamic adjustment algorithm, combined with the historical transmission error rate, corrects the initial information bit count and outputs the target information bit count. The LDPC (Low Density Parity Check Code) control engine calculates the number of iterations based on the target information bit count and the real-time signal-to-noise ratio, and outputs the LDPC iteration count.
[0026] Furthermore, outlier removal and downsampling are required on the original point cloud data to obtain preprocessed point cloud data. A preset clustering algorithm is then used to cluster the preprocessed point cloud data, outputting dust particle cluster labels for several dust particle clusters corresponding to the preprocessed point cloud data. Then, the equivalent volume diameter of each dust particle cluster is determined, and this equivalent volume diameter is used as the dust particle size distribution array. In other words, the data preprocessing module removes outliers and downsamples the original point cloud data, outputting cleaned preprocessed point cloud data. The DBSCAN (Density-Based Spatial Clustering of Applications with Noise) clustering engine clusters the cleaned point cloud data based on Euclidean distance, outputting dust particle cluster labels. The feature extraction unit obtains the equivalent volume diameter of each cluster and outputs a dust particle size distribution array. Finally, the dust particle size distribution array is divided into grids according to the tunnel space to obtain corresponding three-dimensional grid data. A dust concentration gradient map is then generated based on the dust particle size distribution array and the three-dimensional grid data, and a corresponding spatial distribution heat map is generated based on the dust concentration gradient map. In other words, the spatial gridding processor divides the dust particle size distribution array into grids according to the tunnel space, outputs three-dimensional grid data, and the heat map generator combines the dust particle size distribution array and the three-dimensional grid data to generate a dust concentration gradient map and output a spatial distribution heat map.
[0027] Step S13: Generate polarization configuration instructions based on the dust particle size characteristics and the spatial distribution heat map, output enhanced point cloud data according to the polarization configuration instructions, and fuse the enhanced point cloud data, the dust concentration vector and the mining equipment vibration data through preset edge computing nodes to obtain a joint state matrix.
[0028] In this embodiment, as Figure 2As shown, the radar control module needs to generate polarization configuration commands based on dust particle size characteristics and spatial distribution heatmaps to output enhanced point cloud data. Specifically, the basic polarization angle needs to be calculated based on dust particle size characteristics and spatial distribution heatmaps, and then corrected according to the concentration gradient in the spatial distribution heatmaps to obtain an optimized polarization angle. The optimized polarization angle is then converted into corresponding phase control codewords, and the radiation beam is reconstructed based on the phase control codewords to obtain the reconstructed radiation beam signal. That is, the polarization control solver receives dust particle size characteristics and spatial distribution heatmaps, calculates the basic polarization angle, corrects the basic polarization angle according to the concentration gradient in the spatial distribution heatmaps, outputs the optimized polarization angle, the FPGA (Field Programmable Gate Array) configurator converts the optimized polarization angle into phase control codewords for metasurface units, the antenna drive circuit loads the phase control codewords to adjust the resonant characteristics of each metasurface unit, and outputs the reconstructed radiation beam. Ultimately, an environmental scan is performed based on the reconstructed radiation beam signal to receive the enhanced echo signal. Then, three-dimensional imaging processing is performed on the enhanced echo signal to obtain enhanced point cloud data. In other words, the radar echo processor uses the reconstructed radiation beam to scan the environment and receive the enhanced echo signal, while the point cloud reconstruction unit performs three-dimensional imaging processing on the enhanced echo signal and outputs enhanced point cloud data.
[0029] In this embodiment, as Figure 2As shown, the enhanced point cloud data, dust concentration vector, and mining equipment vibration data need to be fused using a fusion module to obtain a joint state matrix. Specifically, firstly, mining equipment vibration data needs to be acquired in real time, and a raw data stream needs to be generated based on the enhanced point cloud data, dust concentration vector, and mining equipment vibration data. The raw data stream is then timestamped to obtain a corresponding synchronized data set. That is, mining equipment vibration data is acquired in real time using an accelerometer at a sampling frequency of 2kHz. The receiving interface acquires enhanced point cloud data, dust concentration vector, and mining equipment vibration data, outputs a raw data stream, and performs timestamping processing on the raw data stream to output a synchronized data set. Further, the spatial coordinates corresponding to the synchronized data set need to be transformed to the roadway reference coordinate system to obtain a standardized data set. The dust concentration gradient and equipment vibration spectrum features are then extracted from the standardized data set. That is, the spatial coordinates in the synchronized data set are transformed to the roadway reference coordinate system, standardized data is output, and the feature extraction engine extracts the dust concentration gradient and equipment vibration spectrum features from the standardized data, outputting an environment-equipment feature vector. Then, based on the dust concentration gradient and equipment vibration spectrum characteristics, a joint feature vector is generated, and spatiotemporal calibration is performed on the joint feature vector to obtain a synchronized data set. Based on the adjusted dust concentration gradient in the synchronized data set, the feature weights are adjusted to obtain a weighted feature matrix. That is, the data alignment module receives the environment-equipment feature vector, performs spatiotemporal calibration, and outputs the synchronized data set; the feature weighting unit dynamically adjusts the weights of each feature according to the dust concentration gradient and outputs the weighted feature matrix. Finally, spatial features are extracted from the weighted feature matrix to obtain spatiotemporal feature blocks, and corresponding state codes are generated based on these blocks. The state codes are then mapped to the corresponding joint state matrix. In other words, a 3D convolutional layer extracts spatial features from the weighted feature matrix, outputting spatiotemporal feature blocks; an LSTM time series analyzer processes the time series dependencies in the spatiotemporal feature blocks, outputting state codes; and a fully connected classifier maps the state codes to the environment-equipment joint state matrix.
[0030] Step S14: Construct a target 5G channel based on the number of information bits and the number of iterations of the linear block code, and upload the joint state matrix to the target cloud for state analysis through the target 5G channel. Then, optimize the scanning frequency of the data monitoring device through feedback control measures and the spatial distribution heat map.
[0031] In this embodiment, as Figure 2As shown, control measures need to be output through the environmental analysis module, and the time slot allocation weights and radar scanning frequency need to be adjusted through the network optimization module. Specifically, a target 5G channel needs to be constructed based on the number of information bits and the number of linear block code iterations. The joint state matrix is then uploaded to the target cloud for state analysis through the target 5G channel. The target cloud generates a corresponding risk feature vector based on the feature parameters decoded from the joint state matrix, and inputs the risk feature vector into a preset model to output a corresponding risk score. Based on the risk score and historical maintenance records, a corresponding risk level is generated, and control measures corresponding to the risk level are matched from a preset database. The control measures are then fed back to the data monitoring device. Specifically, the 5G URLLC (Ultra-reliable & Low-latency Communication) encapsulator hierarchically packages the environment-equipment joint state matrix, outputs encrypted data packets, and uploads them to the cloud through the 5G channel. The deep inference engine inputs the risk feature vector into a pre-trained 3D CNN-BiLSTM hybrid model, outputs the original risk score, and corrects it based on the equipment's historical maintenance records to output a calibrated risk level. Based on the pre-set rule base of the calibrated risk level, output control measures, perform safety and conflict detection on the control measures, and output the risk level and control measures.
[0032] Furthermore, the received control measures need to be executed, and the corresponding command response delay and execution success rate need to be determined based on the execution status. Dust accumulation areas are identified based on the spatial distribution heatmap, and time slot allocation weights are determined using the command response delay and accumulation areas. Based on the time slot allocation weights, execution success rate, and accumulation areas, the optimized scanning frequency for the radar equipment is determined, and the data acquisition frequency of the radar equipment is optimized based on the optimized scanning frequency. Specifically, the execution status parsing module receives the feedback execution status (Feedback_Status), parses out the command response delay and execution success rate, and obtains the dust concentration gradient and accumulation areas from the spatial distribution heatmap. The time slot weight calculation unit dynamically adjusts the proportion of emergency time slots based on the command response delay and accumulation areas, outputs the time slot allocation weights, and, combined with the execution success rate and accumulation areas, obtains the radar scanning frequency. Then, the data acquisition frequency of the radar equipment is optimized based on the optimized scanning frequency.
[0033] In this embodiment, dust concentration vector and raw point cloud data can be determined based on dust data from the mining area; the number of information bits of the polar code and the number of iterations of the linear block code can be determined based on the dust concentration vector, so as to construct the target 5G channel based on these two parameters; dust particle size characteristics and spatial distribution heat map can be obtained through the raw point cloud data, and enhanced point cloud data can be determined based on these two parameters; the enhanced point cloud data, dust concentration vector and mining equipment vibration data can be fused through preset edge computing nodes, and the fused joint state matrix can be uploaded to the cloud for analysis through the target 5G channel, and the current scanning frequency can be optimized through feedback control measures and spatial distribution heat map. In this way, through the polarization angle control equation and metasurface antenna reconstruction steps in the radar control module, dynamic optimization of the electromagnetic wave polarization direction in a dusty environment is achieved. This enables the radar beam to adapt to changes in dust particle size distribution and spatial concentration gradient, significantly improving the point cloud data quality and 3D imaging accuracy of dust monitoring. Furthermore, through the dynamic calculation of Polar code information bits and LDPC iterations in the adaptive coding control module, real-time matching and optimization of 5G communication parameters with dust concentration characteristics are achieved, ensuring reliable transmission of monitoring data in high-dust environments. These two core steps work synergistically to construct a closed-loop optimization system from environmental perception to data transmission, enabling the device to accurately acquire and stably propagate dust status and equipment operation data throughout the entire mining process.
[0034] As can be seen from the foregoing embodiments, the method of this application can determine the dust concentration vector based on the collected dust concentration data of the mining area, and output the original point cloud data based on the collected dust particle size distribution matrix. Therefore, this embodiment provides a detailed description of how to determine the dust concentration vector and the original point cloud data based on the collected data. See also... Figure 3 As shown in the figure, this invention discloses a mining data monitoring and optimization method based on 5G communication, including: Step S21: Output the original light intensity signal to each section of the mine roadway through the preset laser equipment, and receive the scattered laser signal scattered by dust, so as to output a current signal group based on the scattered laser signal.
[0035] In this embodiment, a preset laser device is first used to output raw light intensity signals to each section of the mine tunnel and receive scattered laser signals from dust particles. Based on these scattered laser signals, a current signal set is output. Specifically, the laser emitting unit directionally emits a 650 nm laser beam to each section of the tunnel. The laser beam scatters with dust particles in the tunnel air, forming a scattered light signal. The photoelectric sensor array consists of multiple spatially distributed silicon photodiodes. Each photodiode receives the scattered light signal at its corresponding spatial angle and converts it into an analog current signal. The current signals output by multiple photodiodes together form a current signal set. The amplitude of the current signal set is positively correlated with the dust concentration at the corresponding spatial location. This current signal set serves as the raw electrical signal for dust concentration detection and is output to subsequent processing stages. The photoelectric sensor array is arranged with equal angular intervals, for example, one photodiode every 15 degrees, ensuring omnidirectional coverage of the tunnel space. The laser emitting unit operates using pulse modulation, for example, square wave modulation at a frequency of 1 kHz to reduce ambient light interference. The current signal set output by the photoelectric sensor array maintains signal integrity after passing through an impedance matching circuit.
[0036] Step S22: Amplify and filter the current signal group to obtain the current signal to be converted, convert the current signal to be converted into a standard voltage signal, and then convert the standard voltage signal into the corresponding target digital signal.
[0037] In this embodiment, the current signal group needs to be amplified by a programmable gain amplifier. The amplification factor is dynamically adjusted according to the dust concentration range, for example, using a gain of 100 times in the low concentration region and a gain of 10 times in the high concentration region. The amplified signal is then filtered by a bandpass filter to remove high-frequency noise and low-frequency drift. The passband range of the bandpass filter is set to 100Hz to 5kHz to match the laser modulation frequency. The filtered signal enters a precision rectifier circuit to be converted into a DC voltage signal, and then passes through a voltage follower to output a standardized voltage signal. The amplitude range of the standardized voltage signal is limited to between 0 and 5V to adapt to A The input requirements for the AD conversion module are as follows: the AD conversion module uses a 16-bit resolution analog-to-digital converter to convert the standardized voltage signal into a digital signal, and the sampling rate is set to 10kSPS to ensure signal integrity; the microprocessor receives the digital signal output by the AD conversion module, and calculates the dust concentration value based on the light intensity-concentration conversion algorithm established by Mie scattering theory. The algorithm takes into account the scattering characteristics of 650nm wavelength laser on typical mineral dust particles; the microprocessor performs spatial mapping of the concentration value of each photodiode channel, and outputs the dust concentration data of each zone. The zone division corresponds to the scanning area of the laser emitting unit.
[0038] Step S23: Process the target digital signal based on Mie scattering to obtain dust concentration data for each section of the mine roadway, and perform noise reduction and vectorization processing on the dust concentration data to obtain the corresponding dust concentration vector.
[0039] In this embodiment, a time-series-based state-space model is established based on the dust concentration data of each zone output by the microprocessor received by the Kalman filter. The state variables include the current dust concentration value and the rate of change of concentration. The Kalman filter performs iterative calculations through two stages: prediction and update. In the prediction stage, the dust concentration at the current moment is estimated based on the state at the previous moment. In the update stage, the state estimate is corrected by comparing the actual measured value with the predicted value. The Kalman gain matrix is dynamically adjusted based on the measurement noise covariance and the process noise covariance. The measurement noise covariance is determined by the calibration data of the photoelectric sensor array, and the process noise covariance reflects the natural fluctuation characteristics of the dust concentration. After multiple iterations, the Kalman filter outputs smoothed dust concentration estimates for each zone, eliminating random measurement errors and instantaneous interference. Finally, the processed dust concentration estimates for each zone are arranged in spatial order to form a standardized dust concentration vector. The vector dimension is consistent with the number of lane zones, and each element represents the normalized dust concentration value of the corresponding zone, with a normalization range of 0 to 1.
[0040] Step S24: Output a radio frequency signal through a preset signal generator, convert the radio frequency signal into a corresponding electromagnetic wave, and then capture the echo signal of the electromagnetic wave scattered by the dust.
[0041] In this embodiment, a frequency-modulated continuous wave (FM) signal with a center frequency of 60 GHz is generated by a millimeter-wave signal generator, and the modulation bandwidth is set to 2 GHz to achieve high distance resolution. The radio frequency (RF) signal is transmitted to a metasurface antenna array via a low-loss coaxial cable. The metasurface antenna array consists of M×N programmable metasurface units. Each metasurface unit performs wavefront shaping on the incident RF signal by loading different phase offsets to form an electromagnetic beam with a specific polarization direction. The polarization direction is dynamically adjusted according to a preset polarization angle control equation. The electromagnetic beam propagates in the tunnel space and is scattered by dust particles, with some energy being reflected to form an echo signal. The dust reflection signal receiver adopts the same metasurface antenna array structure as the transmitter. By switching the operating mode, it switches from the transmitting state to the receiving state. The low-noise amplifier in the receiver pre-amplifies the weak echo signal, and the amplified signal is down-converted to an intermediate frequency (IF) by a mixer. The IF signal is then output to the subsequent processing stage after being bandpass filtered. The passband range of the bandpass filter matches the modulation bandwidth of the FM continuous wave, for example, it is set to 1 MHz to 100 MHz to retain effective Doppler information.
[0042] Step S25: Perform quadrature demodulation on the echo signal to obtain the corresponding baseband signal, and perform fast Fourier transform on the baseband signal to generate the corresponding target spectrum.
[0043] In this embodiment, the intermediate frequency (IF) signal output from the dust reflection signal receiver is received through the IQ demodulation module. The IF signal is decomposed into in-phase and quadrature components by a quadrature mixer, which uses a local oscillator synchronized with the transmitted signal as a reference signal. The in-phase and quadrature components are filtered by a low-pass filter to remove high-frequency components and then combined into a complex baseband signal. The cutoff frequency of the low-pass filter is set to the maximum Doppler shift of the frequency-modulated continuous wave, for example, 500kHz, to cover the range of dust particle movement speeds in the tunnel. The FFT processor performs a two-dimensional fast Fourier transform on the baseband signal. The first dimension of the FFT processes the distance information, and the number of transform points is consistent with the number of sweep points of the frequency-modulated continuous wave. The second dimension of the FFT processes the Doppler information, and the number of transform points is consistent with the number of accumulated pulses. The transform result generates a distance-Doppler spectrum, which is also the target spectrum.
[0044] Step S26: Generate the dust particle size distribution matrix of each zone of the mine roadway based on the target spectrum, and output the original point cloud data based on the dust particle size distribution matrix.
[0045] In this embodiment, the distance-Doppler spectrum, i.e., the peak points in the target spectrum, needs to be analyzed by the point cloud generation unit. The three-dimensional coordinates of each scattering point in the tunnel coordinate system are calculated using the triangulation principle. The beam pointing angle information of the metasurface antenna array is considered during the coordinate transformation process. The final output raw point cloud data contains the three-dimensional coordinates and reflection intensity information of each detection point. The data format is an N×4 matrix, where N is the number of detection points, and the four columns correspond to the X, Y, and Z coordinates and the reflection intensity value, respectively.
[0046] In this embodiment, Mie scattering processing can effectively detect aerosols, smoke, and dust in the atmosphere, thereby improving the accuracy of the final determined dust concentration vector. Furthermore, the output point cloud data can provide a high-precision three-dimensional environment of the mine tunnel.
[0047] As can be seen from the foregoing embodiments, the method of this application can determine the number of information bits of the polar code and the number of iterations of the linear block code based on the dust concentration vector, and process the original point cloud data through a preset clustering algorithm to obtain dust particle size characteristics and a spatial distribution heatmap. Therefore, this embodiment provides a detailed description of how to determine the number of information bits of the polar code and the number of iterations of the linear block code, as well as how to determine the dust particle size characteristics and the spatial distribution heatmap. See [link to documentation]. Figure 4 As shown in the figure, this invention discloses a mining data monitoring and optimization method based on 5G communication, including: Step S31: Output the current channel quality index based on the dust concentration vector, and query the bit information corresponding to the channel quality index in the preset mapping table to obtain the number of initial information bits corresponding to the polar code.
[0048] In this embodiment, a standardized dust concentration vector output by a Kalman filter is received through a channel state interface. The dust concentration value is mapped to a channel quality index through a linear transformation, and the transformation coefficients are determined based on measured data from 5G channels in the mine. The channel quality index ranges from 0 to 15, corresponding to the CQI (Channel Quality Indicator) level defined by the 3GPP (3rd Generation Partnership Project) standard. The Polar code configuration unit stores a preset mapping table of channel quality indices and coding parameters. The mapping table records the recommended number of Polar code information bits under different channel quality indices. The query process uses a binary search algorithm to quickly locate the coding parameter range corresponding to the current channel quality index, and the initial number of information bits is the median of the range. The initial number of information bits ensures that the coding efficiency is within the range of 30% to 70%. For example, when the channel quality index is 8, the initial number of information bits corresponds to 50% of the total length of the coding block. The mapping table is constructed based on a large amount of measured data from mine channels, and the optimal coding parameters are determined through curve fitting.
[0049] Step S32: Correct the initial number of information bits according to the historical transmission error rate to obtain the target number of information bits, and calculate the number of iterations based on the target number of information bits and the real-time signal-to-noise ratio to obtain the number of iterations for the linear block code.
[0050] In this embodiment, the initial number of information bits needs to be corrected by dynamically adjusting the algorithm in conjunction with the historical transmission bit error rate, and the target number of information bits is output. The LDPC control engine calculates the number of iterations based on the number of information bits and the real-time signal-to-noise ratio, and outputs the number of LDPC iterations. The formula is expressed as follows: ; in, This represents the number of LDPC iterations. Based on the number of iterations, This is the signal-to-noise ratio adjustment factor. For real-time signal-to-noise ratio, This is the total length of the coded block. for, For bitrate compensation factors, The target signal-to-noise ratio.
[0051] Step S33: Remove outliers and downsample the original point cloud data to obtain preprocessed point cloud data. Then, cluster the preprocessed point cloud data using a preset clustering algorithm to output dust particle cluster labels for several dust particle clusters corresponding to the preprocessed point cloud data.
[0052] In this embodiment, a further data preprocessing module receives the raw point cloud data output by the point cloud generation unit. First, a statistical outlier removal algorithm is used to remove noise points. The algorithm determines the outlier threshold based on the average distance between each point and its k nearest neighbors. For example, k is set to 20 and the distance threshold is set to 1.5 times the average distance. Downsampling is performed using a voxel grid filtering method. The voxel size is set to 0.1m × 0.1m × 0.1m according to the spatial resolution requirements of the alleyway. The point closest to the centroid is retained within each voxel. The cleaned point cloud data is input into the DBSCAN clustering engine. The engine parameters include the neighborhood radius eps (Epsilon) and the minimum number of points minPts. The eps value is related to the average spacing of dust particles, for example, it is set to 0.15m. minPts is set to 5 to ensure effective clustering. The clustering process calculates the Euclidean distance between each point and other points in the neighborhood, and points with density connections are grouped into the same cluster. Point cloud data with cluster labels is output.
[0053] Step S34: Determine the equivalent volume diameter of each particle cluster in the plurality of dust particle clusters, so as to use the equivalent volume diameter of each particle cluster as a dust particle size distribution array, and divide the dust particle size distribution array into grids according to the tunnel space to obtain the corresponding three-dimensional grid data.
[0054] In this embodiment, the feature extraction unit performs principal component analysis on the point set of each cluster, calculates the distribution characteristics of the point set in the three principal directions, and obtains the equivalent volume diameter by equating the volume of the point set within the cluster to the diameter of a sphere. The dust particle size distribution array is divided into 0.01mm intervals to statistically analyze the frequency of occurrence of the equivalent diameter of each cluster, and the array length covers the typical range of mineral dust particle sizes, such as 0.1μm to 100μm. Further, the spatial grid processor receives the dust particle size distribution data output by the feature extraction unit and establishes a three-dimensional rectangular coordinate system according to the actual dimensions of the roadway. The coordinate system uses the mining face as the reference plane, with the X-axis along the roadway direction, the Y-axis perpendicular to the roadway direction, and the Z-axis along the height direction. The grid is divided using an equal-interval segmentation method, for example, one grid every 1 meter in the X direction, one grid every 0.5 meters in the Y direction, and one grid every 0.5 meters in the Z direction. Each grid cell records the number and particle size distribution data of dust particles falling within the spatial range, and the output format is three-dimensional grid data in a three-dimensional array structure.
[0055] Step S35: Generate a dust concentration gradient map based on the dust particle size distribution array and the three-dimensional grid data, and generate a corresponding spatial distribution heat map based on the dust concentration gradient map.
[0056] In this embodiment, a heatmap generator is used to read the dust particle density information from the three-dimensional mesh data, and a bilinear interpolation method is used to calculate the concentration value at the mesh nodes. The concentration gradient map uses color mapping to represent the dust concentration differences in different regions, with colors ranging from blue to red indicating concentrations from low to high. For example, blue represents 0-10. Red represents 90-100 The final output of the spatial distribution heatmap is a two-dimensional matrix with coordinate information. The matrix element values correspond to the normalized concentration values of each grid cell. The normalization range is 0 to 1, where 0 represents no dust and 1 represents the highest concentration.
[0057] In this embodiment, by dynamically calculating the number of Polar code information bits and the number of LDPC iterations in the adaptive coding control module, real-time matching and optimization of 5G communication parameters and dust concentration characteristics are achieved, ensuring reliable transmission of monitoring data in high dust environments.
[0058] As can be seen from the foregoing embodiments, the method of this application can generate polarization configuration instructions based on the dust particle size characteristics and the spatial distribution heatmap, output enhanced point cloud data according to the polarization configuration instructions, and perform feature fusion on various data including the enhanced point cloud data. Therefore, this embodiment provides a detailed description of how to output enhanced point cloud data and how to perform feature fusion. See [link to documentation]. Figure 5 As shown in the figure, this invention discloses a mining data monitoring and optimization method based on 5G communication, including: Step S41: Calculate the basic polarization angle based on the dust particle size characteristics and the spatial distribution heat map, and correct the basic polarization angle according to the concentration gradient in the spatial distribution heat map to obtain the optimized polarization angle.
[0059] In this embodiment, the polarization control solver needs to receive the dust particle size characteristics and spatial distribution heatmap to calculate the fundamental polarization angle. Specifically, the formula is as follows: ; in, Based on the polarization angle, The wavelength of electromagnetic waves, The median particle size of the dust. The wind speed in the alleyway.
[0060] Furthermore, it is necessary to receive the concentration gradient values of each grid cell in the spatial distribution heatmap through a polarization control solver, and correct the basic polarization angle using a linear weighting method. The weighting coefficients are proportional to the concentration gradient; for example, for every 10° increase in the concentration gradient... The polarization angle was increased by 0.5 degrees to obtain an optimized polarization angle.
[0061] Step S42: Convert the optimized polarization angle into the corresponding phase control codeword, and reconstruct the radiation beam according to the phase control codeword to obtain the reconstructed radiation beam signal.
[0062] In this embodiment, the optimized polarization angle is rounded to the nearest integer and then output to the FPGA configurator. The FPGA configurator stores a phase control code table for the metasurface units, which records the correspondence between the polarization angle and the phase offset of each unit. The lookup process uses a lookup table method to convert the optimized polarization angle into an 8-bit binary phase control codeword. The codeword is transmitted to the antenna drive circuit through a serial peripheral interface. The antenna drive circuit includes multiple digital potentiometers. Each potentiometer adjusts the bias voltage of the corresponding metasurface unit according to the phase control codeword. The voltage adjustment range is, for example, 0-5V, corresponding to a phase change of 0-360 degrees. After the bias voltage is applied, each metasurface unit changes its resonant characteristics, and the radiation characteristics of the entire array are adjusted accordingly, forming an electromagnetic beam with a new polarization direction. The polarization direction of the reconstructed radiation beam achieves optimal matching with the dust concentration gradient distribution, the main lobe width of the beam is kept less than 10 degrees, and the sidelobe level is lower than -20dB.
[0063] Step S43: Perform an environmental scan based on the reconstructed radiation beam signal and receive the enhanced echo signal. Then, perform three-dimensional imaging processing based on the enhanced echo signal to obtain enhanced point cloud data.
[0064] In this embodiment, the radar echo processor controls the metasurface antenna array to transmit reconstructed radiation beams to scan the tunnel environment. Beam scanning employs a combination of mechanical rotation and electronic scanning. The mechanical rotation ranges from 0 to 180 degrees horizontally, while electronic scanning achieves beam deflection from -30 to +30 degrees vertically. The receiving channel collects echo signals scattered by dust particles. After processing by a low-noise amplifier and a bandpass filter, the signal is input to a quadrature demodulator. The demodulator reference signal maintains coherence with the transmitted signal. The demodulated baseband signal is digitized by an analog-to-digital converter with a sampling rate of... The system is set to 100 MSPS to ensure signal fidelity. The point cloud reconstruction unit performs pulse compression processing on the digitized echo signal, and the compressed signal is selected by a range gate to extract the effective scattering point information. The three-dimensional imaging algorithm is based on multi-view radar image fusion technology, which combines beam pointing angle information and distance measurement values to calculate the spatial coordinates of each scattering point. The enhanced point cloud data contains the three-dimensional coordinates, reflection intensity, and signal-to-noise ratio information of each detection point. The data format is an N×5 matrix, where N is the number of effective scattering points, and the 5 columns correspond to the X, Y, and Z coordinates, reflection intensity value, and signal-to-noise ratio value, respectively.
[0065] Step S44: Collect vibration data of mining equipment in real time, generate a raw data stream based on the enhanced point cloud data, dust concentration vector and vibration data of mining equipment, and timestamp the raw data stream to obtain the corresponding synchronized data group.
[0066] In this embodiment, accelerometers installed at key parts of the mining equipment continuously collect triaxial vibration signals at a sampling frequency of 2kHz. The signals are transmitted to the data receiving interface via an anti-interference shielded cable. The data receiving interface simultaneously receives enhanced point cloud data output from the point cloud reconstruction unit, standardized dust concentration vector output from the Kalman filter, and equipment vibration data output from the accelerometers. These three types of data are encapsulated into a raw data stream using a unified data frame format. The timestamp synchronization of the raw data stream uses the PTP (Precision Time Protocol), with the main clock source being the GPS (Global Positioning System) signal from the 5G base station. The system (Global Positioning System) synchronizes the clock, with the slave clock being the local clock of each data acquisition terminal. The synchronization process calculates the clock offset of each data source and aligns the time axes of data with different sampling rates using a linear interpolation method. For example, it matches 2kHz vibration data with 10Hz dust concentration data on the timestamp. The synchronized data is grouped by time windows, each window being 100ms long and containing complete data samples of various types. The output format is a time-aligned synchronized data group. The enhanced point cloud data in the synchronized data group retains the original three-dimensional coordinate information, the dust concentration vector maintains normalization characteristics, and the equipment vibration data includes three-axis acceleration waveforms.
[0067] Step S45: Transform the spatial coordinates corresponding to the synchronous data set to the roadway reference coordinate system to obtain a standardized data set, and extract the dust concentration gradient and equipment vibration spectrum characteristics from the standardized data set.
[0068] In this embodiment, the enhanced point cloud data in the synchronous data group needs to be transformed to the roadway reference coordinate system through a coordinate transformation matrix. The transformation matrix is determined by the installation position and attitude angle of the metasurface radar. The transformed spatial coordinates are completely aligned with the roadway design drawings, and the output format is standardized data in a unified coordinate system. The feature extraction engine performs spatial difference calculation on the dust concentration values in the standardized data to obtain the concentration gradient components in the X, Y, and Z directions. The gradient calculation adopts the central difference method, and the step size is set to 1 times the grid spacing. The equipment vibration data is transformed by 512-point FFT to extract the spectral features of the 0-1kHz frequency band. The energy ratio of the spectrum is calculated in 20 frequency bands.
[0069] Step S46: Generate a corresponding joint feature vector based on the dust concentration gradient and the vibration spectrum characteristics of the equipment, and perform spatiotemporal calibration on the joint feature vector to obtain a synchronized data set.
[0070] In this embodiment, a joint feature vector needs to be generated based on the dust concentration gradient and the vibration spectrum characteristics of the equipment. The dust concentration gradient component and the vibration spectrum characteristics are merged into an environment-equipment feature vector, which is also a joint feature vector. The vector dimension is fixed at 23 dimensions, with the first 3 dimensions being the concentration gradient and the last 20 dimensions being the vibration spectrum energy distribution. The feature vector is subjected to Min-Max normalization to unify the numerical range of each dimension to between 0 and 1. The output format is a time series feature matrix.
[0071] Furthermore, based on the environment-device feature vectors output by the feature extraction engine received by the data alignment module, the data streams with different sampling rates are aligned using a time series interpolation method. The interpolation algorithm uses cubic spline curve fitting. The spatiotemporal calibration process uses the GPS clock of the 5G base station as the reference time source, with a calibration accuracy of ±1ms, and outputs a synchronized data set with strictly synchronized time.
[0072] Step S47: Adjust the feature weights of the synchronous data group based on the adjusted dust concentration gradient in the synchronous data group to obtain a weighted feature matrix.
[0073] In this embodiment, feature weights need to be adjusted based on the adjusted dust concentration gradient in the synchronous data set to obtain a weighted feature matrix. Specifically, the feature weighting unit parses the dust concentration gradient components in the synchronous data set and dynamically calculates the weight coefficients of each feature according to the gradient magnitude. The weight calculation uses a Sigmoid function mapping; for example, for every 10 increments in the concentration gradient... The corresponding feature weights are increased by 0.1. The weighting process multiplies each dimension of the environment-device feature vector by the corresponding weight coefficient, and the output format is a weighted feature matrix that retains the original dimensions. The row vectors of the weighted feature matrix represent time series samples, the column vectors represent feature dimensions, and the matrix element values are the product of the original feature values and the weight coefficients.
[0074] Step S48: Extract spatial features from the weighted feature matrix to obtain spatiotemporal feature blocks, generate corresponding state codes based on the spatiotemporal feature blocks, and then map the state codes to the corresponding joint state matrix.
[0075] In this embodiment, a weighted feature matrix output by the feature weighting unit is received through a three-dimensional convolutional layer. A sliding window calculation is performed in the spatial dimension using 5×5×3 convolutional kernels, with each kernel corresponding to a spatial feature pattern. The convolution operation employs the ReLU (Linear Rectification Function) activation function and SAME padding, with a stride of 2×2×1. The output is a spatiotemporal feature block containing local spatial features. The number of channels in the spatiotemporal feature block is consistent with the number of convolutional kernels; for example, using 32 convolutional kernels outputs a 32-channel feature map. The LSTM (Long Short-Term Memory) temporal analyzer is configured with 128 hidden units. The input gate, forget gate, and output gate use the Sigmoid activation function. Cell state updates use Tanh (Hyperbolic Tangent Activation)... Function (hyperbolic tangent activation function); The LSTM temporal analyzer processes spatiotemporal feature blocks step by step. Each time step takes a 32-dimensional feature vector as input and outputs the hidden state and cell state. The final state encoding is obtained by projecting the hidden state of the last time step through a fully connected layer. The encoding dimension is 64-dimensional and includes a compressed representation of temporal dependencies.
[0076] Furthermore, the 64-dimensional state code received from the LSTM time series analyzer by the fully connected classifier needs to be nonlinearly transformed through a three-layer fully connected network. The first fully connected layer maps the 64-dimensional input to a 128-dimensional feature space, the second fully connected layer compresses it to a 32-dimensional feature space, and the third fully connected layer outputs the environment-equipment joint state matrix. The row vectors of the environment-equipment joint state matrix represent time series samples, and the column vectors contain three types of indicators: dust concentration level, equipment health score, and risk flag.
[0077] In this embodiment, the polarization direction of electromagnetic waves in a dusty environment is dynamically optimized through the polarization angle control equation and the metasurface antenna reconstruction step. This enables the radar beam to adapt to changes in dust particle size distribution and spatial concentration gradient, significantly improving the point cloud data quality and three-dimensional imaging accuracy of dust monitoring.
[0078] As can be seen from the foregoing embodiments, the method of this application allows the joint state matrix to be uploaded to the target cloud for state analysis via the target 5G channel. Then, the scanning frequency of the data monitoring device is optimized using feedback control measures and the spatial distribution heatmap. This embodiment provides a detailed explanation of how to perform this optimization. See [link to documentation]. Figure 6 As shown in the figure, this invention discloses a mining data monitoring and optimization method based on 5G communication, including: Step S51: Upload the joint state matrix to the target cloud via the target 5G channel, so that the target cloud can generate a corresponding risk feature vector based on the feature parameters decoded from the joint state matrix, and input the risk feature vector into a preset model to output a corresponding risk score; generate a corresponding risk level based on the risk score and historical maintenance records, match the control measures corresponding to the risk level from the preset database, and then feed the control measures back to the data monitoring device.
[0079] In this embodiment, the environment-device joint state matrix needs to be divided into three priority data streams—urgent, important, and normal—based on risk level using a 5G URLLC encapsulator. The urgent data stream is encrypted using AES-256, the important data stream is encrypted using AES-128, and the normal data stream is not encrypted. During the encapsulation process, a timestamp, sequence number, and CRC (Cyclic Redundancy Check) check code are added to each data packet. The urgent data packet is assigned the highest QoS level with a latency budget of less than 10ms, and the important data packet has a latency budget of less than 50ms. The encrypted data packets are transmitted through the URLLC channel of 5G NR (5G New Radio). The physical layer uses Polar code encoding, and the modulation method is adaptively selected according to the channel quality—either QPSK (Quadrature Phase Shift Keying) or 16QAM (16-Quadrature Amplitude Modulation). The final output encrypted data packet conforms to the 3GPP Release 16 URLLC standard format and is relayed to the cloud data center through the 5G base station.
[0080] Furthermore, the cloud-stored environment-equipment joint state matrix is transmitted to the local processing unit via a 5G URLLC channel. During reception, CRC checks are first performed, and data packets failing the checks are discarded. Data packets that pass the checks are decrypted using either AES-256 or AES-128 algorithms based on their encryption level to restore the original data. The decryption keys are stored and managed via a secure chip. The decoded state data retains the original timestamp and sequence number information, and its format is completely consistent with the environment-equipment joint state matrix sent from the sending end. The data processing unit parses the column vector structure of the environment-equipment joint state matrix, extracting the first column (dust concentration level data), the second column (equipment stress index), and the third column (vibration spectrum characteristics). The dust concentration level data is converted to absolute concentration values, with the conversion coefficient dynamically adjusted according to the spatial location of the tunnel. The equipment stress index and vibration spectrum characteristics undergo inverse normalization to restore their original physical dimensions. The final generated risk feature vector contains three key parameters: dust concentration value, equipment stress value, and vibration dominant frequency energy ratio. The vector dimension is fixed at 3 dimensions, corresponding to monitoring indicators of different dimensions.
[0081] Next, the deep inference engine receives the 3D risk feature vector output by the data processing unit, reshapes the vector into a 1×1×3 tensor structure, and uses it as input to the 3D CNN-BiLSTM hybrid model; 3D The CNN part uses two convolutional layers to extract spatial features. The first layer uses four 3×3×3 convolutional kernels, and the second layer uses eight 3×3×3 convolutional kernels. Each convolutional layer is followed by ReLU activation and max pooling. The BiLSTM (Bidirectional Long Short-Term Memory) part is configured with 64 hidden units to process temporal features. The output vectors of the forward and backward LSTMs are fused by concatenation. The fully connected layer maps the 128-dimensional features output by the BiLSTM to a 1-dimensional space, and outputs the original risk score in the range of 0-1 after passing through the Sigmoid activation function. The equipment historical maintenance record database stores the maintenance logs of the past 12 months, which include equipment failure type, repair time and parts replacement information. The correction algorithm calculates the equipment reliability coefficient based on the failure frequency in the maintenance records. The coefficient ranges from 0.8 to 1.2. Multiplying it by the original risk score gives the calibrated risk level. The final output calibrated risk level retains two decimal places. The larger the value, the higher the risk level.
[0082] Finally, a pre-defined rule base is used to determine the mapping relationship between risk levels and control measures. The rule base uses an interval division method; for example, risk levels 0-0.3 correspond to normal monitoring, 0.3-0.7 correspond to reduced speed operation, and 0.7-1.0 correspond to emergency shutdown. The query process uses a binary search algorithm to quickly locate the interval to which the calibrated risk level belongs and outputs the corresponding control measure text description. The safety detection unit verifies the compatibility between the control measures and the current state of the equipment, checking the equipment operating mode, load conditions, and historical operation records. The conflict detection unit compares the execution priorities of multiple parallel control measures, retaining the highest priority measure when there is an instruction conflict. Control measures that pass the test are appended with timestamps and digital signatures and encapsulated together with the calibrated risk levels into JSON format output data. The final output risk level and control measure data fields include the risk level value, control measure code, effective time, and validity period. The data packet size is fixed at 128 bytes. Finally, the Feedback_Status data packet containing the control measures is fed back to the data monitoring device.
[0083] Step S52: Execute the received control measures and determine the corresponding instruction response delay and execution success rate according to the corresponding execution status.
[0084] In this embodiment, the execution status parsing module receives the returned JSON format Feedback_Status data packet, parses the timestamp field in it to calculate the delay time from the issuance of the control command to the response, with the delay time accuracy reaching the millisecond level; the execution success rate statistics module records the execution results of the past 10 control commands and calculates the ratio of the number of successful executions to the total number of executions as the execution success rate.
[0085] Step S53: Determine the dust accumulation area based on the spatial distribution heat map, and determine the time slot allocation weight through the instruction response delay and the accumulation area.
[0086] In this embodiment, the concentration gradient of the heatmap needs to be calculated using the Sobel operator by the spatial distribution heatmap processing unit. The gradient magnitude exceeds the threshold of 50. The regions are marked as clustered areas; the clustered area identification algorithm is based on the connected component analysis method, and isolated areas with an area of less than 1 square meter are filtered out; the parsed command response delay and execution success rate are stored together with the dust concentration gradient and the coordinates of the clustered areas as structured data, and the data format is a record array containing 5 fields; each record corresponds to a state snapshot within a control cycle, and the time resolution is synchronized with the heat map update frequency, for example, a record is generated every 10 seconds.
[0087] Furthermore, the time slot weight calculation unit needs to receive the instruction response delay and cluster area data output by the execution status parsing module, and calculate the proportion of emergency time slots using a linear interpolation method. The interpolation coefficient is inversely proportional to the instruction response delay. For example, for every 10ms increase in delay, the proportion of emergency time slots decreases by 5%. The total area of the cluster area is used as a weight adjustment factor. For every 1 square meter increase in area, the proportion of emergency time slots increases by 2%. The final output time slot allocation weight is limited to between 5% and 25% to ensure basic communication needs.
[0088] Step S54: Determine the optimized scanning frequency for the radar device based on the time slot allocation weight, the execution success rate, and the clustering area, and optimize the data acquisition frequency of the radar device based on the optimized scanning frequency.
[0089] In this embodiment, the radar scanning frequency calculation unit needs to combine the execution success rate and the cluster area data. The basic scanning frequency is set to 1Hz. For every 10% decrease in the execution success rate, the frequency increases by 0.2Hz. The concentration gradient amplitude of the cluster area is used as a frequency correction term. For every 10mg / m³ / m increase in the gradient amplitude, the scanning frequency increases by 0.1Hz. The upper limit of the adjusted radar scanning frequency is 5Hz to avoid hardware overload. The frequency value is rounded to 0.1Hz accuracy for output.
[0090] In this embodiment, performing status analysis through the cloud can effectively reduce local computing pressure and save local computing resources. Furthermore, by optimizing the scanning frequency of the device, the device can achieve accurate acquisition and stable transmission of dust status and equipment operation data throughout the entire mining process.
[0091] See Figure 7 This invention discloses a mining data monitoring and optimization device based on 5G communication, applied to a data monitoring device, comprising: The point cloud data determination module 11 is used to collect dust concentration data of each section of the mine roadway, determine the dust concentration vector based on the dust concentration data, and obtain the dust particle size distribution matrix of each section of the mine roadway based on the preset radar equipment, and output the original point cloud data based on the dust particle size distribution matrix. The point cloud data processing module 12 is used to determine the number of information bits of the polar code and the number of iterations of the linear block code based on the dust concentration vector, and to process the original point cloud data through a preset clustering algorithm to obtain dust particle size characteristics and spatial distribution heat map. Data fusion module 13 is used to generate polarization configuration instructions based on the dust particle size characteristics and the spatial distribution heat map, output enhanced point cloud data according to the polarization configuration instructions, and fuse the enhanced point cloud data, the dust concentration vector and the mining equipment vibration data through preset edge computing nodes to obtain a joint state matrix. The data monitoring and optimization module 14 is used to construct a target 5G channel based on the number of information bits and the number of linear block code iterations, and upload the joint state matrix to the target cloud for state analysis through the target 5G channel. Then, it optimizes the scanning frequency of the data monitoring device through feedback control measures and the spatial distribution heat map.
[0092] In this embodiment, dust concentration vectors and raw point cloud data can be determined based on dust data from the mining area. The number of information bits in the polar code and the number of iterations in the linear block code are determined based on the dust concentration vector, and a target 5G channel is constructed based on these two parameters. Dust particle size characteristics and spatial distribution heatmaps are obtained from the raw point cloud data, and enhanced point cloud data is determined based on these two parameters. The enhanced point cloud data, dust concentration vectors, and mining equipment vibration data are fused using preset edge computing nodes, and the fused joint state matrix is uploaded to the cloud for analysis via the target 5G channel. The current scanning frequency is optimized using feedback control measures and the spatial distribution heatmap. In this way, a closed-loop optimization system from environmental perception to data transmission can be constructed, enabling the device to accurately acquire and stably transmit dust status and equipment operation data throughout the entire mining process.
[0093] In some embodiments, the point cloud data determination module 11 may specifically include: The current signal output unit is used to output the original light intensity signal to each section of the mine roadway through a preset laser device, and to receive the scattered laser signal scattered by dust, so as to output a current signal group based on the scattered laser signal. The signal conversion unit is used to amplify and filter the current signal group to obtain the current signal to be converted, convert the current signal to be converted into a standard voltage signal, and then convert the standard voltage signal into a corresponding target digital signal. The dust data processing unit is used to process the target digital signal based on Mie scattering to obtain dust concentration data of each section of the mine roadway, and to perform noise reduction and vectorization processing on the dust concentration data to obtain the corresponding dust concentration vector.
[0094] In some embodiments, the point cloud data determination module 11 may specifically include: An echo signal capture unit is used to output a radio frequency signal through a preset signal generator, convert the radio frequency signal into a corresponding electromagnetic wave, and then capture the echo signal of the electromagnetic wave scattered by dust. The signal conversion unit is used to perform quadrature demodulation on the echo signal to obtain the corresponding baseband signal, and to perform fast Fourier transform on the baseband signal to generate the corresponding target spectrum. The raw point cloud data output unit is used to generate a dust particle size distribution matrix for each section of the mine roadway based on the target spectrum, and output raw point cloud data based on the dust particle size distribution matrix.
[0095] In some embodiments, the point cloud data processing module 12 may specifically include: The bit information determination unit is used to output the current channel quality index based on the dust concentration vector, and query the bit information corresponding to the channel quality index in a preset mapping table to obtain the number of initial information bits of the polar code. The iteration number determination unit is used to correct the initial number of information bits according to the historical transmission bit error rate to obtain the target number of information bits, and calculate the iteration number based on the target number of information bits and the real-time signal-to-noise ratio to obtain the linear block code iteration number; The data clustering unit is used to remove outliers and downsample the original point cloud data to obtain preprocessed point cloud data, and to cluster the preprocessed point cloud data using a preset clustering algorithm to output dust particle cluster labels for several dust particle clusters corresponding to the preprocessed point cloud data. The data partitioning unit is used to determine the equivalent volume diameter of each particle cluster in the plurality of dust particle clusters, so as to use the equivalent volume diameter of each particle cluster as a dust particle size distribution array, and to divide the dust particle size distribution array into a grid according to the roadway space to obtain the corresponding three-dimensional grid data. The heat map generation unit is used to generate a dust concentration gradient map based on the dust particle size distribution array and the three-dimensional grid data, and to generate a corresponding spatial distribution heat map based on the dust concentration gradient map.
[0096] In some embodiments, the data fusion module 13 may specifically include: The data optimization unit is used to calculate the basic polarization angle based on the dust particle size characteristics and the spatial distribution heat map, and to correct the basic polarization angle according to the concentration gradient in the spatial distribution heat map to obtain the optimized polarization angle. A radiation beam reconstruction unit is used to convert the optimized polarization angle into a corresponding phase control codeword, and to reconstruct the radiation beam according to the phase control codeword to obtain the reconstructed radiation beam signal. An enhanced point cloud data generation unit is used to perform environmental scanning based on the reconstructed radiation beam signal, receive the enhanced echo signal, and then perform three-dimensional imaging processing based on the enhanced echo signal to obtain enhanced point cloud data.
[0097] In some embodiments, the data fusion module 13 may specifically include: The first data synchronization unit is used to collect vibration data of mining equipment in real time, generate a raw data stream based on the enhanced point cloud data, the dust concentration vector and the vibration data of mining equipment, and perform timestamp synchronization on the raw data stream to obtain the corresponding synchronized data group. The feature extraction unit is used to transform the spatial coordinates corresponding to the synchronous data set to the roadway reference coordinate system to obtain a standardized data set, and extract the dust concentration gradient and equipment vibration spectrum features from the standardized data set. The second data synchronization unit is used to generate a corresponding joint feature vector based on the dust concentration gradient and the vibration spectrum characteristics of the equipment, and to perform spatiotemporal calibration on the joint feature vector to obtain a synchronized data set. The weight adjustment unit is used to adjust the feature weights of the synchronous data group based on the adjusted dust concentration gradient in the synchronous data group to obtain a weighted feature matrix. The encoding mapping unit is used to extract spatial features from the weighted feature matrix to obtain spatiotemporal feature blocks, generate corresponding state codes based on the spatiotemporal feature blocks, and then map the state codes into corresponding joint state matrices.
[0098] In some embodiments, the data monitoring and optimization module 14 may specifically include: The data processing unit is used to upload the joint state matrix to the target cloud via the target 5G channel, so that the target cloud can generate a corresponding risk feature vector based on the feature parameters decoded from the joint state matrix, and input the risk feature vector into a preset model to output a corresponding risk score; generate a corresponding risk level based on the risk score and historical maintenance records, match the control measures corresponding to the risk level from a preset database, and then feed the control measures back to the data monitoring device. The parameter determination unit is used to execute the received control measures and determine the corresponding instruction response delay and execution success rate according to the corresponding execution status. The missing weight determination unit is used to determine the dust accumulation area based on the spatial distribution heat map, and to determine the time slot allocation weight through the instruction response delay and the accumulation area; The parameter acquisition optimization unit is used to determine the optimized scanning frequency of the radar device based on the time slot allocation weight, the execution success rate, and the aggregation area, and to optimize the data acquisition frequency of the radar device based on the optimized scanning frequency.
[0099] Furthermore, embodiments of this application also disclose an electronic device, Figure 8 An exemplary embodiment illustrates the structure of an electronic device 20. The content of the figure should not be construed as limiting the scope of this application.
[0100] Figure 8 This application provides a schematic diagram of the structure of an electronic device 20. Specifically, the electronic device 20 may include: at least one processor 21, at least one memory 22, a power supply 23, a communication interface 24, an input / output interface 25, and a communication bus 26. The memory 22 stores a computer program, which is loaded and executed by the processor 21 to implement the relevant steps in the 5G communication-based mining data monitoring and optimization method disclosed in any of the foregoing embodiments. Alternatively, the electronic device 20 in this embodiment may specifically be a computer.
[0101] In this embodiment, the power supply 23 is used to provide operating voltage for each hardware device on the electronic device 20; the communication interface 24 can create a data transmission channel between the electronic device 20 and external devices, and the communication protocol it follows can be any communication protocol applicable to the technical solution of this application, and is not specifically limited here; the input / output interface 25 is used to acquire external input data or output data to the outside world, and its specific interface type can be selected according to specific application needs, and is not specifically limited here.
[0102] In addition, the memory 22, as a carrier for resource storage, can be a read-only memory, random access memory, disk or optical disk, etc. The resources stored thereon can include operating system 221, computer program 222, etc., and the storage method can be temporary storage or permanent storage.
[0103] The operating system 221 is used to manage and control the various hardware devices on the electronic device 20 and the computer program 222, which may be Windows Server, Netware, Unix, Linux, etc. In addition to including a computer program capable of performing the 5G communication-based mining data monitoring and optimization method executed by the electronic device 20 as disclosed in any of the foregoing embodiments, the computer program 222 may further include computer programs capable of performing other specific tasks.
[0104] Furthermore, this application also discloses a computer-readable storage medium for storing a computer program; wherein, when the computer program is executed by a processor, it implements the aforementioned mining data monitoring optimization method based on 5G communication. Specific steps of this method can be found in the corresponding content disclosed in the foregoing embodiments, and will not be repeated here.
[0105] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the apparatus disclosed in the embodiments, since it corresponds to the method disclosed in the embodiments, the description is relatively simple; relevant parts can be referred to in the method section.
[0106] Those skilled in the art will further recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.
[0107] The steps of the methods or algorithms described in conjunction with the embodiments disclosed herein can be implemented directly by hardware, a software module executed by a processor, or a combination of both. The software module can be located in random access memory (RAM), main memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, removable disk, CD-ROM, or any other form of storage medium known in the art.
[0108] Finally, it should be noted that in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.
[0109] The technical solutions provided in this application have been described in detail above. Specific examples have been used to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only for the purpose of helping to understand the methods and core ideas of this application. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of this application. Therefore, the content of this specification should not be construed as a limitation of this application.
Claims
1. A mining data monitoring and optimization method based on 5G communication, characterized in that, Applications in data monitoring devices include: Collect dust concentration data of each section of the mine roadway, determine the dust concentration vector based on the dust concentration data, and obtain the dust particle size distribution matrix of each section of the mine roadway based on the preset radar equipment, and output the raw point cloud data based on the dust particle size distribution matrix. The number of information bits of the polar code and the number of iterations of the linear block code are determined based on the dust concentration vector, and the original point cloud data is processed by a preset clustering algorithm to obtain dust particle size characteristics and spatial distribution heat map. Polarization configuration instructions are generated based on the dust particle size characteristics and the spatial distribution heat map. Enhanced point cloud data is output according to the polarization configuration instructions. The enhanced point cloud data, the dust concentration vector, and the mining equipment vibration data are fused through preset edge computing nodes to obtain a joint state matrix. A target 5G channel is constructed based on the number of information bits and the number of iterations of the linear block code. The joint state matrix is then uploaded to the target cloud for state analysis via the target 5G channel. Finally, the scanning frequency of the data monitoring device is optimized through feedback control measures and the spatial distribution heat map.
2. The mining data monitoring and optimization method based on 5G communication according to claim 1, characterized in that, The process of collecting dust concentration data from various sections of the mine roadways to determine a dust concentration vector based on the dust concentration data includes: The original light intensity signal is output to each section of the mine roadway by a preset laser device, and the scattered laser signal scattered by dust is received, so as to output a current signal group based on the scattered laser signal. The current signal group is amplified and filtered to obtain the current signal to be converted, and the current signal to be converted is converted into a standard voltage signal, and then the standard voltage signal is converted into the corresponding target digital signal. The target digital signal is processed based on Mie scattering to obtain dust concentration data for each section of the mine roadway. The dust concentration data is then denoised and vectorized to obtain the corresponding dust concentration vector.
3. The mining data monitoring and optimization method based on 5G communication according to claim 1, characterized in that, The step of acquiring the dust particle size distribution matrix of each zone of the mine roadway based on a preset radar device, and outputting raw point cloud data based on the dust particle size distribution matrix, includes: A radio frequency signal is output through a preset signal generator, and the radio frequency signal is converted into a corresponding electromagnetic wave. Then, the echo signal of the electromagnetic wave scattered by the dust is captured. The echo signal is quadrature demodulated to obtain the corresponding baseband signal, and the baseband signal is subjected to fast Fourier transform to generate the corresponding target spectrum. Based on the target spectrum, a dust particle size distribution matrix for each zone of the mine roadway is generated, and the original point cloud data is output based on the dust particle size distribution matrix.
4. The mining data monitoring and optimization method based on 5G communication according to claim 1, characterized in that, The process of determining the number of information bits of the polar code and the number of iterations of the linear block code based on the dust concentration vector, and processing the original point cloud data using a preset clustering algorithm to obtain dust particle size characteristics and spatial distribution heatmaps, includes: Based on the dust concentration vector, the current channel quality index is output, and the bit information corresponding to the channel quality index is queried in the preset mapping table to obtain the number of initial information bits of the polar code. The initial number of information bits is corrected based on the historical transmission bit error rate to obtain the target number of information bits, and the number of iterations is calculated based on the target number of information bits and the real-time signal-to-noise ratio to obtain the number of iterations for the linear block code. Outlier removal and downsampling are performed on the original point cloud data to obtain preprocessed point cloud data. The preprocessed point cloud data is then clustered using a preset clustering algorithm to output dust particle cluster labels for several dust particle clusters corresponding to the preprocessed point cloud data. The equivalent volume diameter of each particle cluster in the plurality of dust particle clusters is determined, and the equivalent volume diameter of each particle cluster is used as the dust particle size distribution array. The dust particle size distribution array is then divided into grids according to the tunnel space to obtain the corresponding three-dimensional grid data. A dust concentration gradient map is generated based on the dust particle size distribution array and the three-dimensional grid data, and a corresponding spatial distribution heat map is generated based on the dust concentration gradient map.
5. The mining data monitoring and optimization method based on 5G communication according to claim 1, characterized in that, The step of generating polarization configuration instructions based on the dust particle size characteristics and the spatial distribution heatmap, and outputting enhanced point cloud data according to the polarization configuration instructions, includes: The basic polarization angle is calculated based on the dust particle size characteristics and the spatial distribution heat map, and the basic polarization angle is corrected according to the concentration gradient in the spatial distribution heat map to obtain the optimized polarization angle. The optimized polarization angle is converted into a corresponding phase control codeword, and the radiation beam is reconstructed according to the phase control codeword to obtain the reconstructed radiation beam signal. Environmental scanning is performed based on the reconstructed radiation beam signal, and enhanced echo signals are received. Then, three-dimensional imaging processing is performed based on the enhanced echo signals to obtain enhanced point cloud data.
6. The mining data monitoring and optimization method based on 5G communication according to claim 1, characterized in that, The process of fusing the enhanced point cloud data, the dust concentration vector, and the mining equipment vibration data through preset edge computing nodes to obtain a joint state matrix includes: Vibration data of mining equipment is collected in real time, and a raw data stream is generated based on the enhanced point cloud data, the dust concentration vector and the vibration data of mining equipment. The raw data stream is then timestamped to obtain the corresponding synchronized data group. The spatial coordinates corresponding to the synchronous data set are transformed to the roadway reference coordinate system to obtain a standardized data set, and the dust concentration gradient and equipment vibration spectrum characteristics are extracted from the standardized data set. Based on the dust concentration gradient and the vibration spectrum characteristics of the equipment, a corresponding joint feature vector is generated, and the joint feature vector is spatiotemporally calibrated to obtain a synchronized data set. The feature weights of the synchronous data set are adjusted based on the adjusted dust concentration gradient in the synchronous data set to obtain a weighted feature matrix; Spatial features are extracted from the weighted feature matrix to obtain spatiotemporal feature blocks, and corresponding state codes are generated based on the spatiotemporal feature blocks. Then, the state codes are mapped to the corresponding joint state matrix.
7. The mining data monitoring and optimization method based on 5G communication according to any one of claims 1 to 6, characterized in that, The process of uploading the joint state matrix to the target cloud via the target 5G channel for state analysis, and then optimizing the scanning frequency of the data monitoring device through feedback control measures and the spatial distribution heat map, includes: The joint state matrix is uploaded to the target cloud via the target 5G channel, so that the target cloud can generate a corresponding risk feature vector based on the feature parameters decoded from the joint state matrix, and input the risk feature vector into a preset model to output a corresponding risk score; based on the risk score and historical maintenance records, a corresponding risk level is generated, and control measures corresponding to the risk level are matched from a preset database, and then the control measures are fed back to the data monitoring device. The received control measures are executed, and the corresponding instruction response delay and execution success rate are determined according to the corresponding execution status. Based on the spatial distribution heat map, dust accumulation areas are determined, and time slot allocation weights are determined by the command response delay and the accumulation areas. The optimal scanning frequency for the radar device is determined based on the time slot allocation weight, the execution success rate, and the aggregation area, and the data acquisition frequency of the radar device is optimized based on the optimal scanning frequency.
8. A mining data monitoring and optimization device based on 5G communication, characterized in that, Applications in data monitoring devices include: The point cloud data determination module is used to collect dust concentration data of each section of the mine roadway, determine the dust concentration vector based on the dust concentration data, and obtain the dust particle size distribution matrix of each section of the mine roadway based on the preset radar equipment, and output the original point cloud data based on the dust particle size distribution matrix. The point cloud data processing module is used to determine the number of information bits of the polar code and the number of iterations of the linear block code based on the dust concentration vector, and to process the original point cloud data through a preset clustering algorithm to obtain dust particle size characteristics and spatial distribution heat map. The data fusion module is used to generate polarization configuration instructions based on the dust particle size characteristics and the spatial distribution heat map, output enhanced point cloud data according to the polarization configuration instructions, and fuse the enhanced point cloud data, the dust concentration vector and the mining equipment vibration data through a preset edge computing node to obtain a joint state matrix. The data monitoring and optimization module is used to construct a target 5G channel based on the number of information bits and the number of linear block code iterations, and upload the joint state matrix to the target cloud for state analysis through the target 5G channel. Then, the scanning frequency of the data monitoring device is optimized through feedback control measures and the spatial distribution heat map.
9. An electronic device, characterized in that, include: Memory, used to store computer programs; A processor for executing the computer program to implement the mining data monitoring optimization method based on 5G communication as described in any one of claims 1 to 7.
10. A computer-readable storage medium, characterized in that, Used to store a computer program, wherein the computer program, when executed by a processor, implements the mining data monitoring and optimization method based on 5G communication as described in any one of claims 1 to 7.