Mining three-dimensional laser ultrasonic roadway deformation detection device and detection method

Abstract

The invention relates to a mining three-dimensional laser ultrasonic roadway deformation detection device and method. The device is composed of a TOF laser distance measuring sensor, an ultrasonic sensor, a data processing module and the like. The TOF laser ranging sensor is responsible for high-precision ranging, and data of the TOF laser ranging sensor is subjected to Kalman filtering to suppress noise; the ultrasonic sensor is responsible for large-range monitoring. The data processing module distinguishes event types by comparing time sequence logics (which is triggered firstly, time difference and duration) of signal changes of the two sensors: (1) a passing moving object is represented as' ultrasonic first change, laser later change and short time difference ', and a system neglects laser data in the time period; (2) the real deformation of the wall body is shown as'simultaneous change of the wall body and the wall body 'or'continuous change is detected by ultrasonic detection', the system immediately enters an'emergency mode ', and the updating frequency of a Kalman filtering module is greatly improved, so that a stable high-precision deformation value is output within seconds, and sound-light alarm and remote data reporting are synchronously triggered.

Description

Technical Field

[0001] This invention relates to the field of mine safety detection technology, specifically to a three-dimensional laser ultrasonic tunnel deformation detection device and method for mines. Background Technology

[0002] During long-term mining operations, underground mine roadways are susceptible to deformation, spalling, and even collapse due to geological pressure and groundwater erosion, posing a serious threat to the safety of personnel and equipment. Therefore, real-time and accurate detection of roadway deformation is crucial.

[0003] Currently, tunnel deformation monitoring mainly employs technologies such as laser ranging, total stations, and fiber optic sensing. Among these, laser ranging sensors (such as TOF type) are widely used due to their high accuracy. However, laser sensors typically have a small field of view and are susceptible to interference from temporary obstacles in the detection path (such as passing mine cars or personnel), leading to distorted detection data and even false alarms. Ultrasonic sensors, on the other hand, have advantages such as a large field of view, fast response speed, and low cost, and are often used for object presence detection. However, their ranging accuracy is relatively low, making them difficult to use alone for millimeter-level deformation detection.

[0004] Existing technologies typically use a single sensor or multiple sensors in parallel, failing to effectively resolve the contradiction between "high-precision static deformation measurement" and "dynamic moving object recognition." When an object passes by, the system may misjudge it as tunnel deformation, leading to frequent false alarms; while when actual deformation occurs, traditional fixed-frequency filtering algorithms respond slowly and cannot provide timely data support for emergency decision-making. Summary of the Invention

[0005] To address the aforementioned technical problems, this invention provides a three-dimensional laser ultrasonic tunnel deformation detection device and method for mining applications, the technical solution of which is as follows: A three-dimensional laser ultrasonic tunnel deformation detection device for mining includes: a TOF laser ranging sensor, an ultrasonic sensor, a data processing module, a power supply module, a sensor data collection and monitoring station, and a surface server. The power module is electrically connected to the TOF laser rangefinder, ultrasonic sensor, and data processing module respectively, and is used to provide working power. The TOF laser rangefinder is used to detect static distance data on the tunnel wall, and its output is connected to the Kalman filter module in the data processing module. The ultrasonic sensor has a larger field of view than the TOF laser rangefinder, and is used to detect changes in objects in the tunnel. Its output is directly connected to the data processing module. The data processing module is preset with a detection time difference threshold and an ultrasonic deformation detection time threshold, which are used to simultaneously receive and analyze the raw distance data from the TOF laser rangefinder and the detection data from the ultrasonic sensor; the data processing module determines the type of the object to be detected based on whether the data from the two sensors changes, the temporal relationship of the changes, and the duration of the changes detected by the ultrasonic sensor.

[0006] Furthermore, the determination is divided into two cases: ① If the interference is determined to be from a moving object, the TOF laser ranging data corresponding to that time period will be filtered out and not input into the Kalman filter module; ② If the wall deformation is determined, the Kalman filter module is controlled to increase the filtering frequency, perform accelerated filtering on the raw data of the TOF laser rangefinder, and trigger an alarm.

[0007] Furthermore, the power module is a 127V mining explosion-proof power supply, which can convert non-intrinsically safe power into intrinsically safe power that complies with underground mine safety regulations.

[0008] Furthermore, the field of view of the ultrasonic sensor is not less than 60°, and the field of view of the TOF laser rangefinder is not greater than 30°.

[0009] Furthermore, the data processing module integrates a storage module for storing raw detection data, filtered deformation values, and alarm event information.

[0010] Furthermore, the normal operating filtering frequency of the Kalman filter module is 10Hz-50Hz, and when wall deformation is detected, the filtering frequency is increased to 100Hz-200Hz.

[0011] Furthermore, it also includes a cloud service platform that aggregates sensor data at the monitoring station, connects to downhole sensors and data processing modules via a 485 communication bus, and communicates with the surface server via an Ethernet port.

[0012] Furthermore, the detection method using a three-dimensional laser ultrasonic tunnel deformation detection device for mining includes the following steps: S1: Device initialization, the TOF laser rangefinder and ultrasonic sensor are activated and begin to continuously collect tunnel environment data; S2: The data processing module synchronously receives the raw distance data from the TOF laser rangefinder and the detection data from the ultrasonic sensor; S3: The data processing module performs joint analysis on the data and executes object type determination logic. S3-1: If the ultrasonic sensor detects an object first, and then the data from the TOF laser rangefinder changes, and the time difference between the two is greater than the preset detection time difference threshold, then it is determined that a moving object has passed by, and data filtering is performed. S3-2: If the TOF laser rangefinder and the ultrasonic sensor detect a change in distance at the same time, or if the duration of the change detected by the ultrasonic sensor exceeds the preset ultrasonic deformation detection time threshold, then it is determined to be wall deformation. S4: Perform the corresponding operation based on the result of S3: S4-1: If the object is moving, discard the TOF laser ranging data that is interfered with during this period and retain the valid background data; S4-2: If it is wall deformation, the accelerated filtering mode is activated to run the Kalman filter algorithm at a higher frequency, quickly process the TOF laser raw data to obtain accurate deformation value, and trigger an alarm and upload the deformation data and alarm information.

[0013] Furthermore, the detection time difference threshold ranges from 0.5 seconds to 2 seconds, and the ultrasonic deformation detection time threshold ranges from 3 seconds to 10 seconds.

[0014] Furthermore, it also includes a data backtracking step: the well server or cloud platform stores all historical detection data, deformation records and alarm information, supporting users to query, filter and analyze data by time range and detection location.

[0015] The beneficial effects of this invention are as follows: This invention provides a three-dimensional laser ultrasonic tunnel deformation detection device and method for mining. The device consists of a TOF laser ranging sensor, an ultrasonic sensor, and a data processing module. The TOF laser ranging sensor is responsible for high-precision ranging, and its data is filtered by Kalman filtering to suppress noise; the ultrasonic sensor is responsible for large-area monitoring. The data processing module distinguishes event types by comparing the temporal logic of the changes in the two sensor signals (which one triggers first, the time difference, and the duration): ① When a moving object passes through, the system ignores the laser data during that time period, as the ultrasonic signal changes first, followed by the laser signal, and the time difference is short. ② When the actual deformation of the wall is shown as "both changing simultaneously" or "the ultrasonic sensor detects a continuous change," the system immediately enters "emergency mode," significantly increasing the update frequency of the Kalman filter module, thereby outputting a stable, high-precision deformation value within seconds and simultaneously triggering audible and visual alarms and remote data reporting. This invention can intelligently distinguish between actual tunnel deformation and interference from moving objects, achieving a unified approach to high-precision deformation detection and rapid emergency response, significantly improving the safety detection level of underground mining operations. Attached Figure Description

[0016] Figure 1 This is a schematic diagram of the overall system architecture and a description of the connection types of the present invention; Figure 2 This is a schematic diagram of the power module architecture of the present invention; Figure 3 This is a schematic diagram of the sensor and data acquisition module architecture of the present invention; Figure 4 This is a schematic diagram of the communication network architecture of the present invention; Figure 5 This is a flowchart illustrating the data processing logic of the present invention. Detailed Implementation

[0017] This invention relates to a three-dimensional laser ultrasonic tunnel deformation detection device and method for mining applications, such as... Figure 1 As shown, the detection device of the present invention is set up in units of a "detection node". Each node is installed in the roof or wall of the mine roadway that needs to be detected. Several "detection nodes" along the line are connected to each other through an RS485 bus and connected to the surface server.

[0018] I. A "detection node" contains the following hardware units: Sensor probe assembly, data processing and communication unit, intrinsically safe power supply module, and sensor data collection and monitoring station.

[0019] 1. Sensor Probe Assembly: This assembly comprises a TOF laser rangefinder (e.g., VL53L1CXV0FY / 1, 940nm wavelength, measurement range 0-4m, accuracy ±20mm) and an ultrasonic sensor (e.g., HC-SR04, center frequency 40kHz, detection angle ≥60°), both encapsulated in an explosion-proof, dustproof, and waterproof protective housing. The TOF laser rangefinder is used to detect static deformation data in the tunnel. Its output is connected to a Kalman filter module to filter out environmental noise interference and improve static detection accuracy. Its optical axis is essentially perpendicular to the tunnel wall, aiming at a fixed monitoring point, emitting laser light at a fixed frequency (e.g., 50Hz) and receiving the reflected signal to measure the precise distance from the point to the wall. The ultrasonic sensor has a larger field of view than the TOF laser rangefinder (ultrasonic sensor field of view ≥ 60°, TOF laser rangefinder field of view ≤ 30°), and is used to quickly detect changes in objects in the tunnel. Its output is directly connected to the data processing module, which can transmit detection data in real time. It is installed at a large tilt angle (such as a certain pitch angle and yaw angle) so that its fan-shaped detection area can cover a fan-shaped tunnel space in front of and to the side of the laser measuring point (for example, covering an area 4 meters in front of the laser point and 0.9 meters to the left and right), and is used to sense the entry and exit of objects in this space.

[0020] 2. Intrinsically safe power supply module: such as Figure 2As shown, a 127V mine explosion-proof power supply is used to convert non-safe power into intrinsically safe DC power that meets the standards for underground operation, so as to power the sensor probe group (TOF laser rangefinder sensor, ultrasonic sensor) and data processing unit (STM32 sensor main controller, 485 communication module) to ensure stable operation of the device.

[0021] 3. Data Processing and Communication Unit: This unit is installed in an intrinsically safe junction box near the sensor probe assembly. It contains a data processing module, a power management module, a storage module (e.g., W25Q128), and communication interfaces (RS485, Ethernet). The data processing module (e.g., STM32F103C8T6) is the core control unit, running the core decision-making algorithm and Kalman filter program (e.g., ...). Figure 3 (As shown). Preset detection time difference thresholds (which can be set according to the typical passage speed of vehicles or personnel in the tunnel, for example, 0.5s-2s) and ultrasonic deformation detection time thresholds (used to determine whether it is instantaneous interference or continuous deformation, usually set to 3s-10s; exceeding this time is considered as the long-term presence of an object or wall deformation). These are used to receive raw data from the TOF laser rangefinder and detection data from the ultrasonic sensor. By analyzing the change characteristics and time difference of the detection data from the two sensors, the type of detected object is determined—whether it is a moving object or wall deformation.

[0022] like Figure 5 As shown, the data processing module continuously performs the following judgments: Check for significant changes in the ultrasonic signal (such as sudden changes in echo intensity or jumps in distance measurement). If there are no changes, the system is in "quiet monitoring" mode, and the TOF laser data sensor performs Kalman filtering at a normal frequency (e.g., 20Hz), continuously outputting smooth distance values ​​for trend observation. If the ultrasonic signal detects a change, start timing and closely monitor the TOF laser data.

[0023] For moving objects: After the ultrasonic wave changes, the TOF laser rangefinder data also shows a brief "spik" change within a short time (e.g., 0.8 seconds), after which both return to their original state. This is consistent with the characteristics of a mine car or personnel passing by quickly. The data processing module filters this detection data, does not trigger the filtering function, and does not input it into the Kalman filter to avoid interfering with the filtering results. It also controls the storage module to output valid detection data.

[0024] Wall Deformation: After the ultrasonic wave detects a change, the TOF laser data also begins to drift slowly but continuously. Alternatively, if the object detected by the ultrasonic wave (actually a locally protruding rock mass) persists for more than a set time (e.g., 5 seconds), the system determines it as real deformation. Execute Immediately: a. Trigger local and remote alarms.

[0025] b. Switch the update frequency of the Kalman filter from 20Hz to 150Hz.

[0026] c. Store the raw laser and ultrasonic data.

[0027] d. The high-speed filter converges within seconds and outputs accurate deformation data (e.g., "the east side wall has shifted inward by 15 mm").

[0028] The data processing module controls the storage module to output raw data, while simultaneously accelerating the filtering frequency and outputting alarm information to the host computer. After filtering is completed, the roadway deformation value is output and transmitted to the host computer to achieve remote early warning. The data processing module will aggregate the generated "effective detection data" or "deformation event comprehensive data package" (including deformation value, raw data, and alarm information) through the sensor data collection and monitoring station and push it to the 485 communication module in real time. The host computer can remotely export the data through the cloud service platform to facilitate subsequent data backtracking and analysis.

[0029] Kalman filtering is optimized to account for the ±2cm random error (modeled as Gaussian white noise) of the TOF laser rangefinder. Since the monitored object (tunnel wall) can be considered quasi-static over a short timescale, i.e. ,in For process noise, based on the state update equation ; The specific meanings of the expressions in the above equations are as follows: ① As a system default, This is the current state. This represents the system state at the previous moment. This is process noise, which in a static system is the measured value being the state at the previous moment plus the system error; ② For the state update equation, Let k be the posterior state estimate at time k. For the prior state estimate at time k, For Kalman gain, The measured value at time k is the current filter value, which is the filter value of the previous time plus the measured value minus the filter value of the previous time multiplied by the Kalman gain.

[0030] The Kalman filter function can be derived (as shown below). Rapid estimation is achieved by adjusting the weights of the process noise covariance Q and the measurement noise covariance R online, and by increasing the call frequency in emergency mode. Here, x represents the state estimate, P represents the estimated covariance, Q represents the process noise covariance, and R represents the measurement noise covariance.

[0031] ; The default filtering frequency of the Kalman filter module is 10Hz-50Hz, and when wall deformation is detected, the filtering frequency is accelerated to 100Hz-200Hz.

[0032] 4. Sensor Data Collection and Monitoring Station: Downhole, a detection node is typically set up every 50 meters. Centered on this monitoring station, multiple monitoring nodes along the line are connected radially within a 1000-meter radius and integrated into an RS485 bus interface. For example... Figure 4 As shown, the monitoring station has a built-in protocol converter that packages the 485 bus data / alarm information into TCP / IP protocol and connects it to the surface server via a network (such as an underground Ethernet ring network or fiber optic network).

[0033] 5. It also includes an in-ground server: The in-ground server receives data from each monitoring station (each detection node), runs the database and web services. The in-ground server also has a monitoring platform, providing a human-machine interface with B / S or C / S architecture, realizing functions such as map display, real-time data curves, alarm pop-ups, and report generation.

[0034] The mine explosion-proof power supply converts the 127V non-safe power supply to an intrinsically safe power supply. The intrinsically safe circuit is integrated into the sensor data processing motherboard and connected to it. The data processing motherboard is connected to the I2C interface of the laser sensor via the I2C communication protocol to read the data set by the laser sensor. It is connected to the ultrasonic sensor via GPIO to transmit and receive ultrasonic waves and calculate the time difference. The data processing motherboard integrates flash memory to store parameter data, raw data and post-processed data. It also sends the data that needs to be sent to the monitoring station via RS-485. Finally, the monitoring station receives the data and uploads it to the surface server via the network port.

[0035] II. The detection method for each "detection node" includes the following steps: (1) After the device is powered on, each module completes self-test, the TOF laser ranging sensor and ultrasonic sensor start up and begin to continuously detect the roadway environment data, and the Kalman filter module is in standby mode at the default frequency (e.g., 20Hz). (2) The TOF laser ranging sensor collects static distance data of the tunnel in real time and transmits it to the data processing module and the Kalman filter module; the ultrasonic sensor collects data on changes in objects in the tunnel in real time and transmits it to the data processing module. (3) The data processing module receives the detection data from the two sensors and analyzes whether the data changes and the order of the changes: ① If the ultrasonic sensor detects the object first, and then the original data of the TOF laser rangefinder changes, and the time difference between the two detections is greater than the preset detection time difference threshold, it is determined that a moving object has passed by. The detection data is filtered and not input into the Kalman filter module. The storage module outputs the valid detection data. ② If the TOF laser rangefinder and the ultrasonic sensor detect deformation at the same time, or the duration of the deformation detected by the ultrasonic sensor exceeds the preset ultrasonic deformation detection time threshold, it is determined that the wall is deformed. The storage module outputs the original data, and the data processing module triggers the Kalman filter module to accelerate the filtering frequency. (4) The Kalman filter module filters the raw data of the TOF laser rangefinder at an accelerated frequency and outputs accurate tunnel deformation values ​​after processing. (5) When the data processing module determines that the wall is deformed, it synchronously triggers the alarm module to output an alarm signal and stores the original data, alarm information and filtered deformation value.

[0036] The core of this invention lies in sensor fusion and intelligent decision-making, achieving a unified approach to dynamic and static separation, intelligent judgment, and rapid emergency response. This provides a reliable and efficient solution for mine roadway stability detection, ensuring safety in underground mine operations. Its advantages are as follows: ① Balancing detection accuracy and field of view coverage: The TOF laser rangefinder ensures high accuracy in static deformation detection, while the ultrasonic sensor quickly captures changes in objects within the tunnel with its wide field of view. The two work together to meet the needs of accurate detection while expanding the detection range and avoiding blind spots. ② Accurately distinguish between moving objects and roadway deformation: By analyzing the detection time difference between the two sensors and the duration of ultrasonic deformation detection, interference from moving objects such as mine cars and personnel can be effectively filtered out, avoiding false alarms and improving the reliability of detection. ③ Dynamically adjust the filter frequency: During normal testing, the filter is performed at the default frequency to ensure data stability; when wall deformation is detected, the Kalman filter frequency is accelerated to quickly output accurate deformation values ​​and provide timely support for emergency response. ④ Alarm and Data Storage: It has a remote upper computer alarm function to ensure timely transmission of alarm information; the storage module completely records detection data, alarm information and deformation results, supports data backtracking and trend analysis, and provides a scientific basis for tunnel maintenance; ⑤ Adaptable to underground mining environment: The power supply module adopts a 127V mine explosion-proof power supply, which meets intrinsic safety requirements. The network transmission module is adapted to the complex network environment in underground mines, ensuring stable operation of the device.

[0037] The above description is merely a preferred embodiment of the present invention and does not limit the scope of the patent. Any equivalent structural transformations made using the contents of the specification and drawings of the present invention under the inventive concept of the present invention, or direct / indirect applications in other related technical fields, should be included within the scope of protection of this patent.

Claims

1. A three-dimensional laser ultrasonic tunnel deformation detection device for mining, characterized in that, include: Laser rangefinder, ultrasonic sensor, data processing module, power supply module, sensor data collection and monitoring station, and well server; The power module is electrically connected to the laser rangefinder, ultrasonic sensor, and data processing module respectively, and is used to provide working power. A laser rangefinder sensor is used to acquire the initial distance information of the tunnel wall, and its output is connected to a data processing module. An ultrasonic sensor, with a field of view larger than that of a laser rangefinder, is used to acquire secondary distance information within the tunnel space, and its output is directly connected to a data processing module. The data processing module is connected to both the laser rangefinder and the ultrasonic sensor. Based on the first and second distance information, it identifies dynamic interference and static wall deformation within the tunnel and suppresses the data from the laser rangefinder during periods of dynamic interference. The data processing module has preset detection time difference thresholds and ultrasonic deformation detection time thresholds, used to simultaneously receive and analyze the raw distance data from the laser rangefinder and the detection data from the ultrasonic sensor. The data processing module determines the type of object being detected based on whether the data from the two sensors changes, the temporal relationship of the changes, and the duration of the change detected by the ultrasonic sensor.

2. The apparatus as claimed in claim 1, characterized in that, The determination is divided into two cases: ① If the interference is determined to be from a moving object, the laser ranging data corresponding to that time period will be filtered out and not input into the Kalman filter module; ② If the wall deformation is determined, the Kalman filter module is controlled to increase the filtering frequency, accelerate the filtering of the raw data from the laser rangefinder, and trigger an alarm.

3. The apparatus as described in claim 1, characterized in that, The power module is a 127V mining explosion-proof power supply, which can convert non-intrinsically safe power into intrinsically safe power that meets the safety standards for underground mines.

4. The apparatus as claimed in claim 1, characterized in that, The field of view of the ultrasonic sensor is not less than 60°, and the field of view of the laser rangefinder is not greater than 30°.

5. The apparatus as claimed in claim 1, characterized in that, The data processing module integrates a storage module to store raw detection data, filtered deformation values, and alarm information.

6. The apparatus as claimed in claim 1, characterized in that, The normal operating frequency of the Kalman filter module is 10Hz-50Hz. When wall deformation is detected, the filter frequency is increased to 100Hz-200Hz.

7. The apparatus as claimed in claim 1, characterized in that, It also includes a cloud service platform that aggregates sensor data at the monitoring station, connects to downhole sensors and data processing modules via a 485 communication bus, and communicates with the surface server via an Ethernet port.

8. A method for detecting deformation in mine tunnels using a three-dimensional laser ultrasonic scanner based on the device described in any one of claims 1 to 7, characterized in that, Includes the following steps: S1: Device initialization, laser rangefinder and ultrasonic sensor start up, and begin to continuously collect tunnel environment data; S2: The data processing module synchronously receives the raw distance data from the laser rangefinder and the detection data from the ultrasonic sensor; S3: The data processing module performs joint analysis on the data and executes object type determination logic. S3-1: If the ultrasonic sensor detects an object first, and then the data from the laser rangefinder changes, and the time difference between the two is greater than the preset detection time difference threshold, then it is determined that a moving object has passed by, and data filtering is performed. S3-2: If the laser rangefinder and the ultrasonic sensor detect a change in distance at the same time, or if the duration of the change detected by the ultrasonic sensor exceeds the preset ultrasonic deformation detection time threshold, then it is determined to be wall deformation. S4: Perform the corresponding operation based on the result of S3: S4-1: If the object is moving, discard the interfered laser ranging data during that time period and retain the valid background data; S4-2: If it is wall deformation, the accelerated filtering mode is activated to run the Kalman filter algorithm at a higher frequency, quickly process the raw laser data to obtain accurate deformation values, and trigger an alarm and upload the deformation data and alarm information.

9. The method according to claim 8, characterized in that, The detection time difference threshold ranges from 0.5 seconds to 2 seconds, and the ultrasonic deformation detection time threshold ranges from 3 seconds to 10 seconds.

10. The method according to claim 8, characterized in that, It also includes a data backtracking step: the well server or cloud platform stores all historical detection data, deformation records and alarm information, and supports users to query, filter and analyze data by time range and detection location.

Images