Roadway deformation monitoring device based on millimeter wave radar

Abstract

The invention discloses a roadway deformation monitoring device based on a millimeter wave radar. The roadway deformation monitoring device comprises a main shell, the millimeter wave radar, a laser sensor, a PCBA (Printed Circuit Board Assembly) and a bubble level, and an integrated multi-modal data fusion analysis module, a dynamic measurement range intelligent adjustment module, a real-time deformation prediction and control module and a self-calibration and fault diagnosis module. The multi-modal fusion system fuses radar and laser data through a Kalman filtering algorithm, and full-scale high-precision monitoring is realized by using median filtering, Gaussian fitting preprocessing and dynamic weight distribution; the dynamic range measurement module intelligently switches measurement and frequency sweeping ranges based on a decision tree algorithm according to mining progress, geological conditions and distance fluctuation coefficients; the real-time prediction module adopts an ARIMA model to predict the deformation in advance, and automatically adjusts the scanning angle and the acquisition frequency; the self-calibration module uses a high-precision reflecting plate to correct errors everyday; the system supports 5G communication and remote control, has the advantages of high precision, self-adaption and high reliability, and is suitable for long-term deformation monitoring of mine roadways.

Description

Technical Field

[0001] The present invention relates to the technical field of roadway safety monitoring, and particularly to a roadway deformation monitoring device based on millimeter-wave radar. Background Art

[0002] In underground projects such as mine exploitation and tunnel construction, the deformation of roadway surrounding rock is a key hidden danger threatening safe production. Traditional monitoring means such as mechanical measuring rods and total stations have problems such as low coverage rate of single-point measurement, limited frequency of manual inspections, and insufficient real-time performance, and are difficult to meet the dynamic deformation monitoring requirements under complex geological conditions.

[0003] At present, most roadway deformation monitoring devices are based on radar or optical sensors for monitoring, but single sensors have significant limitations. Although millimeter-wave radar has the advantages of all-weather and strong penetrability, the measurement accuracy at close range is greatly affected by the multipath effect. The laser rangefinder has high accuracy at close range, but is significantly interfered by dust and water vapor at long range, and the scanning angle is fixed, unable to dynamically focus on the deformation area.

[0004] In addition, existing devices generally lack an adaptive measurement range adjustment mechanism. When the mining progress changes or the geological conditions are complex, monitoring blind spots or data redundancy are likely to occur. At the same time, the sensor drift problem during long-term operation lacks automatic calibration and fault diagnosis capabilities, resulting in a decline in monitoring accuracy over time and a risk of delayed safety warning.

[0005] Therefore, how to provide a roadway deformation monitoring device based on millimeter-wave radar is an urgent problem to be solved by those skilled in the art. Summary of the Invention

[0006] An object of the present invention is to provide a roadway deformation monitoring device based on millimeter-wave radar. The present invention realizes high-precision, adaptive, and highly reliable roadway deformation monitoring and early warning through multi-modal data fusion, intelligent adjustment of dynamic measurement range, real-time deformation prediction, and self-calibration technology.

[0007] A roadway deformation monitoring device based on millimeter-wave radar according to an embodiment of the present invention includes a main housing, a rear cover, a millimeter-wave radar, a laser, a PCBA, and a bubble level. The millimeter-wave radar and the laser are arranged in the main housing and face the monitoring area. The PCBA is integrated with a data processing and control module. A gland for the penetration and sealing of external cables is provided on the side of the main housing. The bubble level is embedded in the top surface of the main housing. The rear cover is detachably installed on the rear side of the main housing. The device further includes a multi-modal data fusion analysis system, a dynamic measurement range intelligent adjustment module, a real-time feedback and prediction control module, and a self-calibration and fault diagnosis module integrated on the PCBA; The multi-modal data fusion analysis system performs fusion processing on the distance data collected by the millimeter-wave radar and the distance data collected by the laser through the processor built in the PCBA, and generates fused distance data by combining the measurement advantages of both through the Kalman filtering algorithm. The fusion process satisfies the state equation as and the observation equation , where is the state vector at time k, including distance and speed, is the observation vector, including the millimeter-wave radar and laser measurement values, , is the state transition matrix, and are the process and observation noises; The dynamic measurement range intelligent adjustment module uses the processor to calculate the distance fluctuation coefficient according to the initial measurement data, mining progress data, and geological condition data through the decision tree algorithm: where, is the distance measured at the nth time, is the average value of the previous 3 measurements, and the measurement range of the millimeter-wave radar can be dynamically adjusted according to ; The real-time feedback and predictive control module constructs a deformation prediction model based on the ARIMA model, predicts the deformation trend in real time, and adjusts the measurement strategy. The expression of the prediction model is: where, is the predicted deformation at time t, and are the autoregressive and moving average coefficients respectively, is the residual at time t, is the autoregressive order, which determines the number of time steps of past data considered in the autoregressive model, represents a data sequence value related to deformation at a certain time, is the moving average order, which determines the number of time steps of past residuals considered in the moving average model; The self-calibration and fault diagnosis module regularly calls the built-in 2-meter standard distance reflector to emit signals through the processor, and calibrates the measurement error based on the correction formula: Calibrate the measurement error; where, is the radar measurement error rate, is the laser measurement error rate, and monitors equipment failures through the working parameters of the sensors.

[0008] Furthermore, the multi-modal data fusion analysis system includes a preprocessing unit, a weight allocation unit, and a Kalman filtering unit. The preprocessing unit can use median filtering to remove external pulse noise, To measure the standard deviation of the data, Gaussian fitting is performed on the laser data based on the least squares method. Determine the center of the light spot. , where is the light spot intensity, is the coordinate of the center of the light spot, is the standard deviation of the light spot. The fitting error is verified by the root mean square error RMSE and satisfies RMSE ≤ 0.5 mm. The weight allocation unit dynamically allocates weights according to the measured distance. When the distance ≤ 5 m, the weight ratio of the laser data is ≥ 70%. When the distance > 5 m, the weight ratio of the millimeter wave radar data is ≥ 60%. The Kalman filter unit iteratively calculates through the state equation and the observation equation and outputs the fused distance measurement value.

[0009] Furthermore, the adjustment method of the dynamic measurement range intelligent adjustment module includes the following steps: S1. Obtain the mining progress data and geological condition data through the external RS485 interface. The mining progress data includes the excavation depth and the support type. The geological condition data includes the rock compressive strength and the joint density; S2. If the distance fluctuation coefficient of three consecutive measurements ≤ 5% and the excavation speed ≤ 0.5 m / h, set the measurement range of the millimeter wave radar to 0 - 10 m; S3. If the fluctuation coefficient > 5% or the excavation speed > 0.5 m / h, automatically expand the measurement range to 0 - 30 m; where the parameter thresholds are stored in the EEPROM of the PCBA.

[0010] Furthermore, the real-time feedback and prediction control module includes a prediction model construction unit, a real-time warning unit, and a measurement strategy adjustment unit. The prediction model construction unit trains an ARIMA model using historical fused distance data, optimizes the order through the BIC criterion, and establishes the functional relationship between the deformation amount and time. The real-time warning unit drives an external audible and visual alarm when the predicted deformation amount of the stoping roadway in the next 2 hours ≥ 50 mm or the deformation amount of the transportation roadway ≥ 80 mm. The measurement strategy adjustment unit controls the sensor scanning angle to deflect ±15° to focus on the deformation area and increases the acquisition frequency from once every 60 seconds to once every 10 seconds when the monitored deformation rate ≥ 10 mm / h.

[0011] Furthermore, the self-calibration and fault diagnosis module includes a self-calibration unit and a fault diagnosis unit. The self-calibration unit triggers the calibration program at 0:00 every day, controls the millimeter wave radar and the laser to emit signals to the 2 m standard reflector built into the main housing, and the reflectivity of the reflector ≥ 95%, the reference distance is 2 m ± 0.1 mm, and the measurement error is calculated according to the reflection time. and , and correct the sensor parameters. The fault diagnosis unit can monitor the transmission power of the millimeter-wave radar and the spot offset of the laser in real time. The normal range of the transmission power of the millimeter-wave radar is 20 - 25 dBm, and the normal range of the spot offset is ±3°. If the transmission power < 18 dBm or the offset > 5°, it is determined that the sensor is faulty and a fault code is generated and stored in the FLASH.

[0012] Furthermore, the PCBA integrates a 5G communication module. The 5G communication module can upload the fused distance data, prediction results, and fault codes to the cloud server via the MQTT protocol, and can receive remote control instructions, supporting remote adjustment of the measurement range, acquisition frequency, and warning threshold parameters.

[0013] Furthermore, the preprocessing unit uses median filtering with a 5×5 window for the millimeter-wave radar data, and uses the least squares method for the Gaussian fitting of the laser data. The fitting error is verified by the root mean square error, satisfying RMSE ≤ 0.5 mm.

[0014] Furthermore, the input parameters of the decision tree algorithm also include roadway support materials and monitoring periods. The output is the sweep frequency range of the millimeter-wave radar. The sweep frequency range of the millimeter-wave radar is adjustable from 24 GHz to 77 GHz. The roadway support materials include bolts and cables. The monitoring periods include the tunneling period and the stable period. Among them, the sweep frequency automatically expands to 60 - 77 GHz during the tunneling period and shrinks to 24 - 30 GHz during the stable period.

[0015] Furthermore, in the real-time warning unit, the roadway types of the extraction roadway and the transportation roadway are read through the RFID tags externally connected to the PCBA or input by keys. The tags store the roadway type codes, and the system automatically matches the warning thresholds according to the codes.

[0016] Furthermore, the standard reflector built into the main housing is a high-precision diffuse reflector, installed at the exact center of the inner wall of the main housing, with a surface roughness Ra ≤ 0.2 μm, ensuring the stable reflection of the calibration signal and the consistency of the measurement reference.

[0017] The beneficial effects of the present invention are as follows: 1. In the present invention, a multi-modal data fusion scheme of millimeter-wave radar and laser sensor is adopted. By dynamically combining the advantages of both through the Kalman filtering algorithm, and designing a preprocessing unit and a dynamic weight distribution mechanism, it solves the problems that the accuracy of traditional single laser sensors drops sharply due to dust interference at long distances and the accuracy of single radar sensors is insufficient at short distances. The full-range accuracy can be effectively improved after fusion, meeting the millimeter-level deformation monitoring requirements of mine roadways. And by removing radar pulse noise through median filtering and correcting laser spot offset through Gaussian fitting, the data efficiency can be greatly improved in the underground environment with high dust concentration, significantly reducing the impact of invalid data on the monitoring results.

[0018] 2. The present invention is based on a decision tree algorithm, integrates mining progress, geological conditions and real-time measurement data, and dynamically adjusts the measurement range and scanning range of the millimeter-wave radar, effectively avoiding the defects of fixed-range monitoring of traditional devices. No redundant monitoring is required during the stable period, and the range is automatically expanded during the excavation period to capture key deformation areas, taking into account both monitoring accuracy and equipment energy consumption. The scanning strategy is automatically switched according to the support type and monitoring period, which improves the adaptability to complex geological conditions such as joint development zones and soft rock sections, and realizes "on-demand monitoring and precise measurement".

[0019] 3. The present invention constructs a deformation prediction model based on the ARIMA model, optimizes the model order through the BIC criterion, predicts the deformation in the next 2 hours in real time, and adjusts the measurement strategy in a linked manner. Compared with the traditional device that can only display data in real time, the present invention can identify the deformation trend in advance, and the warning time is advanced by 2 hours, which buys precious time for the tunnel support and reinforcement, and effectively reduces the risk of uncontrolled deformation. When the deformation rate is ≥10mm / h, the sensor scanning angle is automatically deflected and the acquisition frequency is increased, so that the data density of the deformation area is increased from 1 group per minute to 6 groups per minute, and its ability to capture subtle deformation is improved, thereby effectively identifying local sudden deformation.

[0020] 4. The present invention uses a built-in 2-meter high-precision diffuse reflection plate to automatically trigger the calibration program at 0:00 every day. The radar and laser error rates are integrated through the correction formula, the radar transmission power and laser spot offset are monitored in real time, and fault codes can be generated and stored and uploaded. Through daily self-calibration, the error rate of long-term measurement of millimeter-wave radar and laser sensors can be effectively reduced. There is no need for manual regular calibration, which reduces maintenance costs. It is suitable for long-term unattended mine monitoring scenarios, and the fault diagnosis response time is less than 30 seconds. It can accurately identify radar power anomalies and automatically switch to redundant monitoring mode. The fault missed rate is greatly reduced, and the system reliability is improved.

[0021] 5. The present invention integrates a 5G communication module, uses RFID tags or buttons to input the tunnel type, automatically matches the warning threshold, supports remote adjustment of measurement parameters, reads RFID tags or inputs the tunnel type by buttons, and greatly shortens the on-site configuration time to avoid manual parameter setting errors. The cloud-based remote control function supports real-time response to complex working conditions, greatly improving equipment deployment efficiency. A three-dimensional monitoring network is built through the 5G network to achieve data synchronization and fusion analysis of multiple devices, which has a larger coverage than single-point monitoring and provides full-dimensional data support for the overall stability assessment of the tunnel. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings: Figure 1Schematic diagram of the overall structure of a roadway deformation monitoring device based on millimeter-wave radar proposed by the present invention; Figure 2 Flowchart of the multi-modal data fusion analysis system of a roadway deformation monitoring device based on millimeter-wave radar proposed by the present invention; Figure 3 Intelligent adjustment logic block diagram of the dynamic measurement range of a roadway deformation monitoring device based on millimeter-wave radar proposed by the present invention; Figure 4 Flowchart of the real-time feedback and prediction control module of a roadway deformation monitoring device based on millimeter-wave radar proposed by the present invention; Figure 5 Hardware connection diagram of the self-calibration and fault diagnosis module of a roadway deformation monitoring device based on millimeter-wave radar proposed by the present invention.

[0023] In the figure: 1, main housing; 2, rear cover; 3, millimeter-wave radar; 4, laser sensor; 5, gland; 6, bubble level; 7, PCBA. Detailed implementation mode

[0024] In the description of the present invention, the orientation or positional relationship indicated by terms such as "upper", "lower", "inner", "outer", etc. is usually based on the orientation when the device is normally installed and used, and is only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, so it should not be construed as a limitation of the present invention. Unless otherwise clearly specified and defined, terms such as "set", "installed", "connected", etc. should be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be directly connected, or indirectly connected through an intermediate medium, and it can be the communication inside two elements. For those of ordinary skill in the art, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0025] In this application, unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the art. In case of inconsistency, it shall be based on the meaning described in this specification or the meaning obtained according to the content recorded in this specification. The terms used herein are only for the purpose of describing the embodiments of this application and are not intended to limit this application.

[0026] The following will describe in detail the embodiments of the present invention with reference to the accompanying drawings. Taking a specific mine roadway monitoring scenario as an example, the working process of the roadway deformation monitoring device based on the millimeter-wave radar 3 will be described.

[0027] Please refer to Figures 1 - 5, A roadway deformation monitoring device based on a millimeter-wave radar 3 includes a main housing 1, a rear cover 2, a millimeter-wave radar 3, a laser sensor 4, a PCBA 7, and a bubble level 6. The main housing 1 and the rear cover 2 are made of high-strength and corrosion-resistant engineering plastics or metal materials, such as aluminum alloy. This material not only ensures the structural strength of the device to adapt to the complex environment underground in mines but also has good corrosion resistance to prevent equipment damage caused by factors such as humidity and corrosive gases underground. The main housing 1 and the rear cover 2 are connected by screws or buckles, facilitating the installation and maintenance of internal components.

[0028] The millimeter-wave radar 3 and the laser sensor 4 are installed inside the main housing 1, and the detection directions of both are towards the roadway area to be monitored, so as to obtain the distance data of the roadway in real time.

[0029] The millimeter-wave radar 3 is selected with a model having high resolution and high-precision measurement capabilities, and can accurately measure the distance information of the target object. The laser sensor 4 is a high-precision laser ranging sensor. It uses the linear propagation characteristic of laser light and calculates the distance by measuring the time difference between laser emission and reception, with high precision and stability.

[0030] The PCBA 7 integrates a data processing and control module, which is responsible for receiving and processing the data from the millimeter-wave radar 3 and the laser sensor 4, and controlling the operation of each functional module. On the PCBA 7, components such as a microprocessor, a storage chip, and a communication module are included. The microprocessor needs to have high-speed data processing capabilities and rich interface resources to meet the real-time processing requirements of the data collected by the millimeter-wave radar 3 and the laser sensor 4. The storage chip is used to store the collected data, processing algorithms, and system configuration parameters, etc., such as a large-capacity Flash memory. The communication module integrates 5G communication functions and selects a 5G communication chip that meets the standards to ensure that the data can be stably and quickly transmitted to the cloud server.

[0031] A gland 5 is provided on the side of the main housing 1 for the penetration and sealing of external cables. The gland is made of stainless steel and has an IP68 protection level. There is a rubber sealing ring inside, which can effectively prevent dust and water vapor underground in mines from entering the inside of the housing. After the cable penetrates through the gland, it is fixed by a compression nut to ensure a firm connection and reliable sealing, meeting the requirements of explosion-proof and moisture-proof for underground equipment in the coal mine safety regulations.

[0032] The bubble level 6 uses a high-precision circular or tubular bubble level 6. By observing the position of the bubble in the level, the operator can easily adjust the device to the horizontal state, ensuring that the measurement directions of the millimeter-wave radar 3 and the laser sensor 4 are accurate and error-free, and avoiding measurement errors caused by the inclination of the device.

[0033] In this embodiment, first, a suitable installation location is selected. Generally, a location on the roadway wall with a wide field of view and capable of clearly observing the monitoring area is chosen. The main housing 1 is firmly installed on the roadway wall using expansion bolts or other fixing devices. During the installation process, by observing the bubble level 6, the angle of the main housing 1 is adjusted so that the bubble is located at the center of the level, ensuring that the device is in a horizontal state. After the installation is completed, the data lines and power lines between the millimeter-wave radar 3, the laser sensor 4, and the PCBA 7 are connected to ensure a secure connection and avoid abnormal data transmission or equipment power-off caused by loosening. Finally, the rear cover 2 is covered to further protect the internal components.

[0034] Please refer to Figure 2 , the multi-modal data fusion analysis system includes a preprocessing unit, a weight assignment unit, and a Kalman filtering unit.

[0035] Specifically, the preprocessing unit can perform preliminary processing on the data collected by the millimeter-wave radar 3 and the laser sensor 4. For the data of the millimeter-wave radar 3, a median filtering algorithm with a 5×5 window is used to remove external pulse noise. In the actual mine environment, there are various electromagnetic interference sources, such as underground electrical equipment, communication equipment, etc. These interferences may cause abnormal pulse noise in the data collected by the millimeter-wave radar 3. The principle of the median filtering algorithm is to arrange the data within a 5×5 window in ascending order and take the median value as the data output at the center position of the window.

[0036] For example, assume the data within the window is: [12, 5, 20, 3, 25, 18, 30, 8, 40, 15, 50, 35, 60, 22, 45, 7, 32, 55, 47, 52, 10, 27, 37, 42, 62]; Sort these data in ascending order to get: [3, 5, 7, 8, 10, 12, 15, 18, 20, 22, 25, 27, 30, 32, 35, 37, 40, 42, 45, 47, 50, 52, 55, 60, 62]; There are 25 data in the 5×5 window, so the median value is 30, and the value 30 can be output at the center position of this window, thus achieving the effect of removing abnormal values such as pulse noise.

[0037] In this way, pulse noise can be effectively removed, improving the stability and reliability of the data.

[0038] For the data of the laser sensor 4, the least squares method is used for Gaussian fitting to determine the center of the light spot. Since the laser is affected by factors such as dust and fog in the roadway during propagation, the shape of the light spot is distorted, affecting the accuracy of distance measurement. The principle of the least squares method is to find a set of parameters that minimize the sum of the squares of the errors between the measured data and the theoretical model.

[0039] In this embodiment, it is assumed that the theoretical model of the laser light spot is a Gaussian function , where is the light spot intensity, is the light spot center coordinate, is the standard deviation of the light spot. By performing least squares fitting on multiple light spot data points collected by the laser sensor 4, continuously adjust , , and values, so that the sum of the squares of the errors between the fitting curve and the actual data points is minimized. After a large number of experimental verifications, the root mean square error RMSE of this fitting method satisfies RMSE ≤ 0.5mm, which can accurately determine the position of the light spot center and provide an accurate data basis for subsequent distance calculations.

[0040] The weight allocation unit dynamically allocates weights to the data of the millimeter-wave radar 3 and the laser sensor 4 according to the measured distance. When the measured distance ≤ 5 meters, the measurement accuracy of the laser sensor 4 is higher, and at this time, the weight ratio of the laser data is ≥ 70%. This is because in the short-distance range, the laser sensor 4 is relatively less affected by environmental factors and can provide more accurate distance measurement values.

[0041] For example, in the case of a distance of 3 meters, after multiple experimental verifications, the measurement error of the laser sensor 4 is within ±1mm, while the measurement error of the millimeter-wave radar 3 is about ±3mm. Therefore, when fusing the data, giving higher weight to the laser data can improve the accuracy of the fusion result.

[0042] When the distance > 5 meters, the millimeter-wave radar 3 has a certain advantage in long-distance measurement, and its data weight ratio is ≥ 60%. As the measured distance increases, the signal intensity of the laser sensor 4 will gradually weaken, and it will be more affected by environmental factors. The millimeter-wave radar 3 can maintain good measurement performance at a relatively long distance by emitting millimeter-wave signals.

[0043] For example, at a distance of 10 meters, the measurement error of the millimeter-wave radar 3 is about ±5mm, while the measurement error of the laser sensor 4 may increase to more than ±10mm. At this time, increasing the weight of the data of the millimeter-wave radar 3 can better reflect the actual distance situation.

[0044] The Kalman filter unit is based on the state equation for and the observation equation , where is the state vector at time k, including distance and speed, and are the process noise and the observation noise respectively. The process noise mainly comes from the uncertainties within the system, such as the impact of electronic noise of equipment, mechanical vibration and other factors on the measurement results; the observation noise is mainly caused by the measurement errors of the millimeter-wave radar 3 and the lidar 4 itself; is the observation vector, covering the measurement values of the millimeter-wave radar 3 and the lidar 4; is the state transition matrix, which is used to describe the relationship between the state and time. When obtaining the measurement values, it is necessary to synchronize the data of the millimeter-wave radar 3 and the lidar 4 to ensure data fusion at the same moment.

[0045] In actual calculations, it is necessary to estimate and process these two kinds of noises. Usually, the statistical characteristics of the noises, such as the mean and variance, can be obtained through the statistical analysis of a large amount of historical data, and then corresponding processing can be carried out in the Kalman filter algorithm to improve the accuracy of the fusion result.

[0046] In practical applications, the distance information can be directly obtained from the measurement data of the millimeter-wave radar 3 and the lidar 4, while the speed information can be obtained by performing differential calculations on the distance data at consecutive moments. For example, the distance at time is the distance at time is , and the sampling time interval is , then the speed

[0047] In this embodiment, it is assumed that the state transition matrix , where is the sampling time interval. This matrix indicates that within a sampling period, the change in distance is related to the speed, while the speed remains unchanged.

[0048] Through continuous iteration, the Kalman filter unit outputs the fused high-precision distance measurement value. In each iteration process, first, according to the state vector at the previous moment and the state transition matrix F, the state vector at the current moment is predicted, that is . Then, according to the observation vector and the predicted state vector , the Kalman gain is calculated, and the calculation formula is , where is the covariance matrix of the predicted state vector, is the observation matrix, is the covariance matrix of the observation noise. Finally, according to the Kalman gain the predicted state vector is corrected to obtain the optimal estimated state vector at the current moment, that is . Through such an iterative process, the fusion result is continuously optimized to make it closer to the true distance value.

[0049] The dynamic measurement range intelligent adjustment module obtains the mining progress data and geological condition data through an external RS485 interface. The RS485 interface is a commonly used industrial communication interface, which has the advantages of long transmission distance and strong anti-interference ability, and is very suitable for use in the underground mine environment. The mining progress data includes key information such as the excavation depth and the support type. The excavation depth can be obtained through a depth sensor installed on the excavation equipment, and the support type can be obtained through manual input or from the production management system of the mine. The geological condition data covers the rock compressive strength and the joint density, etc. The rock compressive strength can be obtained through on-site sampling and compressive tests in the laboratory; the joint density can be obtained through observations and statistics by geological exploration personnel at the roadway site. These data are transmitted to the data processing and control module of PCBA7 through the RS485 interface according to a specific communication protocol.

[0050] The adjustment strategy can first calculate the distance fluctuation coefficient according to the decision tree algorithm, where is the distance measured at the nth time, is the average value of the previous 3 measurements, and the measurement range of the millimeter-wave radar 3 can be dynamically adjusted according to . The decision tree algorithm is a classification and prediction algorithm based on a tree structure. In this embodiment, it takes the initial measurement data, mining progress data, geological condition data, roadway support materials, and monitoring time period, etc. as input parameters, and outputs the sweep range of the millimeter-wave radar 3.

[0051] If the distance fluctuation coefficient of three consecutive measurements ≤ 5% and the excavation speed ≤ 0.5 m / h, it indicates that the roadway deformation is relatively stable. At this time, the measurement range of the millimeter-wave radar 3 is set to 0 - 10 meters. This is because in the case of stable roadway deformation, a smaller measurement range is sufficient to meet the monitoring requirements, and at the same time, it can reduce the power consumption and data processing volume of the equipment. For example, in a certain stable roadway section, after continuous measurements, the distance fluctuation coefficient is about 3% and the excavation speed is 0.3 m / h. At this time, setting the measurement range of the millimeter-wave radar 3 to 0 - 10 meters can not only ensure the effective monitoring of the roadway deformation but also reduce unnecessary resource consumption.

[0052] Conversely, if the fluctuation coefficient > 5% or the excavation speed > 0.5 m / h, it indicates that the roadway deformation may be relatively large or change relatively fast, then the measurement range is automatically extended to 0 - 30 meters. In this case, the larger measurement range can more comprehensively monitor the deformation of the roadway and timely detect potential safety hazards. For example, during the roadway tunneling process, the excavation speed is relatively fast, reaching 0.8 m / h, and at the same time, the distance fluctuation coefficient is also relatively large, exceeding 5%. At this time, extending the measurement range to 0 - 30 meters can monitor the deformation of the roadway at a farther distance in real time and provide guarantee for construction safety.

[0053] The relevant parameter thresholds are stored in the EEPROM of PCBA7. The EEPROM has the characteristic that data is not lost when the power is off, which is convenient for retaining the previously set parameter thresholds in cases such as device restart or battery replacement. The staff can modify and adjust the parameter thresholds in the EEPROM according to the actual situation through the debugging interface or remote control instructions on PCBA7.

[0054] The input parameters of the decision tree algorithm also include roadway support materials such as bolts and cable bolts, and monitoring time periods during the tunneling period and the stable period. During the tunneling period, since the roadway is continuously being excavated, the surrounding rock stress changes greatly and the deformation risk is relatively high. Therefore, the sweep frequency range of the millimeter-wave radar 3 is automatically extended to 60 - 77 GHz. The higher sweep frequency range can provide higher resolution and more accurate measurement results to timely detect the minor deformation of the roadway during the tunneling process. While during the stable period, the roadway deformation is relatively small. In order to reduce the device power consumption and data processing volume, the sweep frequency range is shrunk to 24 - 30 GHz.

[0055] The real-time feedback and predictive control module includes a prediction model construction unit, a real-time warning unit, and a measurement strategy adjustment unit.

[0056] Among them, the prediction model construction unit: trains the ARIMA model using historical fusion distance data. The ARIMA model is a commonly used time series prediction model, which can effectively capture the trend, seasonality, and randomness characteristics in time series data.

[0057] In this embodiment, first, collect historical fusion distance data over a period of time, such as the fusion distance data collected at regular time intervals, such as every 1 hour, in the past month. Then, preprocess these data, including removing outliers, stationary processing, etc. The stationary processing can be achieved through differential operations, that is, performing first-order or multi-order differences on the time series data to make it meet the stationarity conditions.

[0058] Optimize the model order through BIC. The BIC criterion is a commonly used criterion in model selection. It takes into account both the goodness of fit of the model and the complexity of the model. When training an ARIMA model, try different autoregressive orders p, differencing orders d, and moving average orders q, and calculate the BIC value for each model. Select the model with the smallest BIC value as the optimal model, which can avoid overfitting of the model while ensuring the fitting effect of the model. For example, after multiple experiments, it is found that when the BIC value of the model is the smallest. At this time, the established ARIMA(2,1,1) model can better fit the historical fusion distance data and accurately predict the future deformation trend.

[0059] Predict the future deformation trend by establishing a functional relationship between the deformation amount and time. Assume that the expression of the established ARIMA model is , where is the predicted deformation amount at time t, and are the autoregressive and moving average coefficients respectively, and is the residual at time t. In practical applications, substitute the fusion distance data before the current time into the model to calculate the model parameters and , and then predict the deformation amounts at different future times according to these parameters. Real-time warning unit: When it is predicted that the deformation amount of the mining roadway within the next 2 hours ≥ 50 mm or the deformation amount of the transportation roadway ≥ 80 mm, immediately drive the external sound and light alarm to sound an alarm. Different warning thresholds are set for different types of roadways because there are differences in functions, structures, and usage frequencies between the mining roadway and the transportation roadway. The mining roadway is directly connected to the coal mining face during the coal mining process, and its deformation is directly affected by coal mining activities. The deformation speed may be relatively fast, and the safety requirements are higher, so a lower warning threshold is set. The transportation roadway is mainly used for the transportation of coal and equipment, and its deformation is relatively slow. However, once a large deformation occurs, it will also affect the transportation safety, so a relatively higher warning threshold is set.

[0060] Select a sound and light alarm with high brightness and high volume to ensure that in the noisy underground mine environment, the staff can hear the alarm sound and see the warning light in time. When the warning condition is met, PCBA7 sends a trigger signal to the sound and light alarm through the control interface to make it sound an alarm, reminding the staff to take corresponding measures in time, such as strengthening support, stopping work, etc.

[0061] When the measurement strategy adjustment unit detects that the deformation rate ≥ 10 mm / h, in order to more accurately monitor the deformation area, it controls the scanning angle of the sensor to deflect ±15° to focus on this area. In this embodiment, the scanning angles of the millimeter-wave radar 3 and the laser sensor 4 are controlled by the motor drive device. The motor drive device receives the control signal sent by the PCBA 7, drives the motor to rotate according to the instruction of the signal, and thus drives the scanning mechanism of the sensor to rotate, realizing the adjustment of the scanning angle. For example, when it is detected that the deformation rate of a certain area reaches 12 mm / h, the PCBA 7 sends an instruction to the motor drive device to deflect the scanning angle of the sensor by 15° towards the deformation area, so as to obtain the distance data of this area more concentratedly.

[0062] At the same time, the acquisition frequency is increased from once every 60 seconds to once every 10 seconds. This is achieved by adjusting the sampling time interval of the data acquisition module in the PCBA 7. In the case of a relatively fast deformation rate, increasing the acquisition frequency can obtain denser measurement data and more timely grasp the changes in the deformation area. For example, within the time period of a relatively high deformation rate, data is acquired once every 10 seconds, which can more accurately capture the details of the roadway deformation and provide more sufficient data support for subsequent analysis and decision-making.

[0063] The roadway type is read through the RFID tag externally connected to the PCBA 7 or input by keys. The RFID tag stores the roadway type code. When the tag approaches the RFID reading module on the PCBA 7, the module transmits a radio frequency signal through the antenna, activates the tag and reads the stored roadway type code, such as an 8-bit binary code. "00000001" represents the mining roadway, "00000010" represents the transportation roadway, "00000011" represents the return airway, etc. The RFID reading module uses the 13.56 MHz high-frequency band, and the reading distance ≤ 5 cm to ensure that only the specified tag can be recognized when it approaches, avoiding misreading.

[0064] If the key input method is selected, there are 3 function keys "Type +", "Type -", and "Confirm" set on the PCBA 7 panel, and the roadway type code is input through the combination of long pressing and short pressing. For example, long press the "Type +" key for 3 seconds to enter the setting mode, and short press to switch the code. At this time, the screen synchronously displays the icon of the currently selected roadway type. For example, the "coal mining face" icon is displayed for the mining roadway, and the "track" icon is displayed for the transportation roadway. After pressing the "Confirm" key, the code is stored in the EEPROM of the PCBA 7.

[0065] The system preferentially recognizes RFID tag input. When a valid tag is detected, it automatically overwrites the currently stored roadway type, regardless of whether it was previously input via buttons or set by default. If no tag is detected, the code input via buttons is used as the roadway type. Both input methods support dynamic modification during the operation of the device. For example, when the roadway transitions from the tunneling period to the stable period, by replacing the RFID tag or resetting via buttons, the system will automatically match the corresponding warning threshold and measurement strategy according to the new roadway type.

[0066] To prevent misoperation, the button input requires continuous input of the correct verification code. If the input is not completed within 30 seconds in the setting mode, it will automatically exit to prevent parameter confusion caused by accidental touch by personnel. The RFID tag uses a unique coding binding mechanism. Each roadway type corresponds to a unique tag ID. The mapping relationship between the tag and the roadway type can be configured on the cloud server through the management software. For example, through the background management system, the tag ID "ABC123" can be bound to "Transportation Roadway - East Section", and the geological parameters and historical deformation data of this area can be associated to achieve intelligent configuration and management of the monitoring system.

[0067] Through the above two input methods, the system can quickly and accurately identify the roadway type, avoid manual configuration errors, ensure the accurate matching of the warning threshold and measurement strategy with the actual roadway attributes, and improve the adaptability and reliability of the monitoring system. For example, when the device is migrated from the mining roadway to the transportation roadway, simply bring it close to the corresponding RFID tag or input the new type via buttons, without the need to re - debug the sensor parameters, greatly reducing the on - site operation difficulty and maintenance cost.

[0068] The self - calibration and fault diagnosis module includes a self - calibration unit and a fault diagnosis unit.

[0069] Among them, the self - calibration unit realizes regular calibration of the millimeter - wave radar 3 and the laser sensor 4 through the built - in high - precision diffuse reflection plate to ensure long - term measurement accuracy. The calibration program is automatically triggered by the real - time clock module of the PCBA7 at 0:00 every day. The specific process is as follows: Hardware design of the reflection plate: The standard reflection plate is of the diffuse reflection type, made of polytetrafluoroethylene material or a high - reflectivity coating material, with a reflectivity ≥ 95%, the reference distance is strictly calibrated to 2 meters ± 0.1 mm, and the surface roughness Ra ≤ 0.2 μm to ensure uniform and stable reflected signals. The reflection plate is fixedly installed at the exact center of the inner wall of the main housing 1, perpendicular to the emission directions of the millimeter - wave radar 3 and the laser sensor 4. The installation deviation is ensured to be ≤ 0.05° through a precision mechanical structure to avoid calibration reference errors caused by installation tilt.

[0070] Emission of calibration signal and error calculation: After triggering the calibration, the millimeter - wave radar 3 emits a continuous - wave signal of a specific frequency, and the laser sensor 4 emits a pulsed laser signal. Both synchronously emit signals towards the reflection plate. Calculate the measured distance based on the round - trip time of the signal: Measured distance of millimeter-wave radar 3 , where is the speed of light, is the signal round-trip time; Measured distance of laser sensor 4 , where is the flight time of the laser signal.

[0071] Calculate the radar measurement error rate , and the laser measurement error rate .

[0072] Sensor parameter correction: According to the correction formula , the distance measurement models of millimeter-wave radar 3 and laser sensor 4 are corrected for parameters. For example, if the measured radar error rate is 2% and the measured laser error rate is 1%, then in subsequent measurements, the radar data needs to be multiplied by 0.98 and the laser data by 0.99, and then fusion calculation is performed. The corrected parameters are stored in the EEPROM of PCBA7 and are not lost when power is off.

[0073] The fault diagnosis unit realizes the equipment status evaluation by real-time monitoring of key sensor parameters. The specific mechanism is as follows: Monitoring of millimeter-wave radar 3: Real-time read the value of the transmit power register of the radar chip, and the normal range is 20 - 25 dBm. When the transmit power < 18 dBm, it may be due to the failure of the power amplifier module or the loose antenna interface, resulting in signal attenuation. At this time, the radar is determined to be faulty, and a fault code "0x01" is generated and stored in the FLASH. The staff can read the code through remote instructions or the local debugging interface to check the hardware connection or replace the radar module.

[0074] Monitoring of laser sensor 4: Monitor the offset of the laser spot through the built-in angle sensor. The reference direction of the laser emission module is aligned with the axis of the main housing 1. When the spot offset > 5°, it indicates that the laser emission component may have mechanical displacement, and a fault code "0x02" is generated. The system simultaneously triggers the "equipment failure" prompt of the audible and visual alarm to avoid the invalidation of measurement data due to the deviation of the sensor pointing.

[0075] Fault code management: The FLASH stores the last 50 fault records, including fault codes, occurrence times, sensor status parameters, etc., and supports remote export through the 5G communication module, which is convenient for preventive maintenance.

[0076] The 5G communication module integrated in PCBA7 is used to realize data transmission and remote control. The 5G communication module supports NSA / SA dual mode, and the working frequency band covers mainstream frequency bands such as n1 / n3 / n5 / n8 / n41 / n78 / n79, ensuring the network connection stability in the complex environment of the mine shaft. The data transmission and control functions are realized as follows: Upward data transmission: Integrate distance data, the results of the ARIMA model predicting the deformation amount in the next 2 hours, fault codes, etc., and encapsulate them through the MQTT protocol. The topic format is " / mine / device / {device ID} / data", where the device ID is the 12-bit hexadecimal encoded unique identifier built into PCBA7.

[0077] Data encryption: Use the TLS / DTLS protocol to encrypt the transmitted data to prevent data eavesdropping in the mine network and ensure the security of the monitoring data.

[0078] Downward control instructions: Remote parameter adjustment: Support the cloud server to send instructions through MQTT to modify the measurement range from 0 - 10 meters to 0 - 30 meters, the acquisition frequency from 60 seconds per time to 10 seconds per time, and the roadway deformation warning threshold from 50 mm to 40 mm. The instruction parsing module is located in the control software of PCBA7 and takes effect immediately after receiving the instruction, and returns parameter confirmation information.

[0079] Remote calibration and fault reset: Support manually triggering the calibration program, remotely sending the "CALIBRATE" instruction, or resetting the fault status to clear the fault code in the FLASH for easy remote maintenance.

[0080] It should be noted that the bubble level 6 is embedded in the top surface of the main housing 1. A high-precision tubular level is used, and the graduation value is 0.5 mm / m, that is, when it tilts 0.5 mm per meter, the bubble shifts one grid. During installation, adjust the bracket screws to center the bubble to ensure that the detection direction of the sensor is perpendicular to the roadway wall. After actual measurement, when the tilt angle of the device < 0.1°, the distance measurement error < 0.3 mm, meeting the millimeter-level monitoring accuracy requirements.

[0081] The above is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present invention, according to the technical solution and inventive concept of the present invention, making equivalent replacements or changes, should be covered by the protection scope of the present invention.

Claims

1. A roadway deformation monitoring device based on a millimeter-wave radar, characterized in that, It includes a main housing, a rear cover, a millimeter-wave radar, a laser, a PCBA, and a bubble level. The millimeter-wave radar and the laser are arranged inside the main housing and face the monitoring area. The PCBA is integrated with a data processing and control module. A gland for the penetration and sealing of external cables is provided on the side of the main housing. The bubble level is embedded in the top surface of the main housing. The rear cover is detachably installed on the rear side of the main housing. The device further includes a multi-modal data fusion analysis system, a dynamic measurement range intelligent adjustment module, a real-time feedback and prediction control module, and a self-calibration and fault diagnosis module integrated on the PCBA; The multi-modal data fusion analysis system processes the distance data collected by the millimeter-wave radar and the distance data collected by the lidar through the processor built in the PCBA, and generates fused distance data by combining the measurement advantages of both through the Kalman filtering algorithm. The fusion process satisfies the state equation as and the observation equation , where is the state vector at time k, including distance and velocity, is the observation vector, including the millimeter-wave radar and lidar measurements, , is the state transition matrix, and are the process and observation noises; The dynamic measurement range intelligent adjustment module uses the processor to calculate the distance fluctuation coefficient through the decision tree algorithm according to the initial measurement data, mining progress data, and geological condition data: where is the measured distance at the nth measurement, is the average value of the previous 3 measurements, and the measurement range of the millimeter-wave radar can be dynamically adjusted according to ; The real-time feedback and predictive control module constructs a deformation prediction model based on the ARIMA model, predicts the deformation trend in real time and adjusts the measurement strategy. The expression of the prediction model is as follows: Wherein, is the predicted deformation at time t, and are the autoregressive and moving average coefficients respectively, is the residual at time t, is the autoregressive order, which determines the number of time steps of past data considered in the autoregressive model, represents the value of a data sequence related to deformation at time is the moving average order, which determines the number of time steps of past residuals considered in the moving average model; The self-calibration and fault diagnosis module periodically transmits signals by calling a built-in 2-meter standard distance reflector through the processor, based on the correction formula: Calibrate the measurement error; Among them, is the radar measurement error rate, is the laser measurement error rate, and the working parameters of the sensor are used to monitor equipment failures.

2. The roadway deformation monitoring device based on a millimeter-wave radar according to claim 1, characterized in that, The multi-modal data fusion analysis system includes a preprocessing unit, a weight allocation unit, and a Kalman filtering unit. The preprocessing unit can use median filtering to remove external pulse noise, is the standard deviation of the measurement data. Gaussian fitting is performed on the laser data based on the least squares method to determine the center of the light spot , where is the light spot intensity, is the coordinate of the center of the light spot, is the standard deviation of the light spot. The fitting error is verified by the root mean square error RMSE and satisfies RMSE ≤ 0.5 mm. The weight allocation unit dynamically allocates weights according to the measured distance. When the distance ≤ 5 m, the weight ratio of the laser data is ≥ 70%. When the distance > 5 m, the weight ratio of the millimeter-wave radar data is ≥ 60%. The Kalman filtering unit iteratively calculates through the state equation and the observation equation, and outputs the fused distance measurement value.

3. The deformation monitoring device for roadway based on millimeter-wave radar according to claim 1, characterized in that, The adjustment method of the dynamic measurement range intelligent adjustment module includes the following steps: S1. Obtain the mining progress data and geological condition data through an external RS485 interface. The mining progress data includes the excavation depth and the support type, and the geological condition data includes the rock compressive strength and the joint density; S2. If the distance fluctuation coefficient of three consecutive measurements ≤ 5% and the excavation speed ≤ 0.5 m / h, set the measurement range of the millimeter-wave radar to 0 - 10 meters; S3. If the fluctuation coefficient > 5% or the excavation speed > 0.5 m / h, automatically expand the measurement range to 0 - 30 meters; Among them, the parameter thresholds are stored in the EEPROM of the PCBA.

4. The deformation monitoring device for roadway based on millimeter-wave radar according to claim 1, wherein The real-time feedback and prediction control module includes a prediction model construction unit, a real-time warning unit, and a measurement strategy adjustment unit. The prediction model construction unit trains an ARIMA model using historical fusion distance data, optimizes the order through the BIC criterion, and establishes a functional relationship between the deformation amount and time. The real-time warning unit drives an external audible and visual alarm when the predicted deformation amount of the extraction roadway in the next 2 hours ≥ 50 mm or the deformation amount of the transportation roadway ≥ 80 mm. The measurement strategy adjustment unit controls the sensor scanning angle to deflect ±15° to focus on the deformation area and increases the acquisition frequency from once every 60 seconds to once every 10 seconds when the monitored deformation rate ≥ 10 mm / h.

5. The roadway deformation monitoring device based on a millimeter-wave radar according to claim 1, characterized in that, The self-calibration and fault diagnosis module includes a self-calibration unit and a fault diagnosis unit. The self-calibration unit triggers the calibration program at 0:00 every day, controls the millimeter-wave radar and the laser to emit signals to a 2-meter standard reflector built into the main housing, and the reflectivity of the reflector is ≥95%, the reference distance is 2 meters ±0.1 mm, and the measurement error is calculated according to the reflection time. and , and corrects the sensor parameters. The fault diagnosis unit can monitor the transmission power of the millimeter-wave radar and the spot offset of the laser in real time. The normal range of the transmission power of the millimeter-wave radar is 20-25 dBm, and the normal range of the spot offset is ±3°. If the transmission power <18 dBm or the offset >5°, it is determined that the sensor is faulty and a fault code is generated and stored in the FLASH.

6. The roadway deformation monitoring device based on a millimeter-wave radar according to claim 1, characterized in that The PCBA is integrated with a 5G communication module. The 5G communication module can upload the fusion distance data, prediction results, and fault codes to the cloud server in the MQTT protocol, and can receive remote control instructions, supporting remote adjustment of the measurement range, acquisition frequency, and warning threshold parameters.

7. The deformation monitoring device for roadway based on millimeter-wave radar according to claim 2, characterized in that, The preprocessing unit performs median filtering on the millimeter-wave radar data with a 5×5 window, uses the least squares method for Gaussian fitting of the laser data, and verifies the fitting error through the root mean square error, satisfying RMSE ≤ 0.5 mm.

8. The roadway deformation monitoring device based on a millimeter-wave radar according to claim 1, characterized in that, The input parameters of the decision tree algorithm further include the roadway support material and the monitoring period. The output is the sweep frequency range of the millimeter-wave radar. The sweep frequency range of the millimeter-wave radar is adjustable from 24 GHz to 77 GHz. The roadway support materials include bolts and cables. The monitoring periods include the tunneling period and the stable period. Among them, it automatically expands to 60 - 77 GHz during the tunneling period and shrinks to 24 - 30 GHz during the stable period.

9. The roadway deformation monitoring device based on millimeter-wave radar according to claim 4, characterized in that, In the real-time warning unit, the roadway types of the extraction roadway and the transportation roadway are read through RFID tags externally connected to the PCBA or input by keys. The tags store the roadway type codes, and the system automatically matches the warning thresholds according to the codes.

10. The roadway deformation monitoring device based on a millimeter-wave radar according to claim 5, characterized in that, The standard reflector built into the main housing is a high-precision diffuse reflector, which is installed at the exact center of the inner wall of the main housing, with a surface roughness Ra ≤ 0.2 μm, ensuring the stable reflection of the calibration signal and the consistency of the measurement reference.

Images