Mine blasting safety monitoring and early warning system
The mine blasting safety monitoring system, which integrates multi-source sensors and deep learning models, solves the problems of single monitoring and isolated data analysis in traditional systems, and achieves precise monitoring and efficient emergency management of the blasting process, thereby improving the safety of mine blasting.
Patent Information
- Application Number
- CN202511277514.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-09-09
- Publication Date
- 2025-12-02
AI Technical Summary
Traditional mine blasting safety monitoring systems rely on single vibration monitoring, lack a correlation model between blasting energy and secondary disasters, have statically set warning thresholds, and fail to effectively integrate multimodal data, resulting in poor accuracy and reliability of the monitoring system under complex geological conditions.
It employs a multi-source sensing module, edge computing nodes, dynamic risk assessment module, intelligent early warning decision-making module, and emergency linkage execution module, integrating a vibration sensor array, shock wave overpressure sensor, toxic gas monitoring unit, and high-definition video monitoring unit. Through edge computing and deep learning models, it dynamically adjusts the early warning threshold to achieve multi-source data fusion and accurate early warning.
It enables comprehensive monitoring of the blasting process, accurately predicts the probability and scope of secondary disasters, dynamically adjusts early warning thresholds, improves the accuracy of early warnings and the efficiency of emergency response, and ensures safe production in mines.
Smart Images

Figure CN121053731A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of mine early warning technology, specifically to a mine blasting safety monitoring and early warning system. Background Technology
[0002] Mining blasting is a key rock-breaking operation in mining. However, the blasting process generates a series of safety hazards, including vibrations, shock waves, and toxic gases. Therefore, constructing a comprehensive mining blasting safety monitoring and early warning system is crucial for ensuring safe mining production and reducing accident risks. Currently, mining blasting safety monitoring systems aim to monitor various parameters during the blasting process in real time, promptly identifying potential hazards and issuing warnings. Traditional mining blasting safety monitoring systems mostly rely on single vibration monitoring methods, collecting blasting vibration data by deploying vibration sensors to determine the blasting effect and safety. However, this single monitoring method has significant limitations, failing to comprehensively capture other hazardous factors generated during the blasting process.
[0003] Traditional monitoring systems lack a correlation model between blast energy and secondary disasters. The magnitude and distribution of blast energy directly affect the occurrence and development of secondary disasters, such as the diffusion of toxic gases and the expansion of rock fissures. Due to the lack of an effective correlation model, the system struggles to accurately predict the probability and extent of secondary disasters, making it impossible to take targeted preventative measures in advance.
[0004] Regarding the setting of early warning thresholds, traditional systems use a static setting method, which cannot be dynamically adjusted according to changes in actual geological conditions, blasting scale, and other factors. Mine geological conditions are complex and variable, and different rock properties and geological structures exhibit significant differences in blasting responses. Fixed early warning thresholds are prone to false alarms or missed alarms under complex geological conditions, reducing the accuracy and reliability of early warnings.
[0005] Furthermore, traditional systems suffer from deficiencies in data processing, analyzing video surveillance data and various sensor data in isolation, failing to achieve effective fusion of multimodal data. Video data can intuitively present on-site personnel behavior and environmental conditions, while sensor data can accurately reflect changes in physical parameters. The fusion of these two can provide a more comprehensive and accurate basis for safety decisions. However, due to the lack of an effective fusion mechanism, the complementary advantages of multi-source data cannot be fully utilized, limiting the functionality and decision-making capabilities of the monitoring system. Summary of the Invention
[0006] (I) Technical Issues
[0007] This invention aims to provide a mine blasting safety monitoring and early warning system that integrates multi-source sensing, intelligent analysis, and dynamic response functions, solving problems such as single monitoring, lack of correlation models, static thresholds, and isolated data analysis in traditional systems, thereby achieving accurate safety monitoring and efficient emergency management.
[0008] (II) Technical Content
[0009] To solve the above-mentioned technical problems, the technical solution of the present invention is as follows: a mine blasting safety monitoring and early warning system, comprising a multi-source sensing module, an edge computing node, a dynamic risk assessment module, an intelligent early warning decision-making module, and an emergency linkage execution module. The multi-source sensing module includes a vibration sensor array, a shock wave overpressure sensor, a toxic gas monitoring unit, a sound wave pickup, and a high-definition video monitoring unit. Each sensor is deployed in a spatial grid topology in the roadways surrounding the blasting point and in key surface areas. The edge computing node is connected to the multi-source sensing module via an industrial-grade wireless mesh network. A real-time data preprocessing unit is configured to filter out environmental noise and extract the time-frequency characteristics of the sensor data. The dynamic risk assessment module is used to input data into the edge computing node. The system includes preprocessed node data; a blasting energy inversion submodule and a secondary disaster prediction submodule; the blasting energy inversion submodule reconstructs the spatial distribution of blasting energy based on vibration and shock wave data; the secondary disaster prediction submodule combines geological models and gas concentration data to generate a toxic gas diffusion path model; the intelligent early warning decision module receives the output of the dynamic risk assessment module; it integrates a multi-level response strategy library and a deep learning early warning model; the early warning model dynamically adjusts the early warning threshold by fusing video behavior recognition and acoustic anomaly detection results; the emergency linkage execution module connects to the ventilation system, personnel positioning terminal, and audible and visual alarms via a mine-use intrinsically safe communication bus; and triggers directional area power outages, forced ventilation, or personnel evacuation commands according to the early warning level.
[0010] Furthermore, the vibration sensor array in the multi-source sensing module adopts a three-component MEMS accelerometer, which is arranged in three dimensions according to the radial, tangential and vertical directions of the blast point, with a sampling frequency ≥2kHz, and the mechanical vibration interference of the roadway is eliminated by an adaptive Kalman filter algorithm.
[0011] Furthermore, the edge computing node has a built-in blasting vibration signal feature extraction unit, which is used to identify the first arrival time of the P wave and the maximum amplitude of the S wave in the waveform. The extraction method includes: performing wavelet packet transform on the original signal; detecting the arrival point of the P wave through the Teager-Kaiser energy operator; and obtaining the extreme points of the S wave envelope by combining Hilbert transform, with a positioning error ≤0.1ms.
[0012] Furthermore, the secondary disaster prediction submodule of the dynamic risk assessment module operates as follows: based on the blasting energy inversion results, the discrete element method is used to simulate the rock mass fracture propagation path; the fracture network data is input into the computational fluid dynamics model to predict the diffusion rate of toxic gas along the fracture; the coordinates of the area where the gas concentration exceeds the limit are output to the early warning decision module, driving the audible and visual alarm to flash red light in the target area.
[0013] Furthermore, the deep learning early warning model of the intelligent early warning decision module has a dual-channel architecture: Channel A is used to receive the time-frequency characteristics of vibration and shock waves, and uses a 1D-CNN network to classify the blast intensity level; Channel B is used to process the RGB-D images of the video monitoring unit, and uses the YOLOv7 model to detect the lack of personnel protective equipment.
[0014] Furthermore, the emergency linkage execution module includes roadway zoning control logic: when the predicted value of blasting vibration velocity is >10mm / s, a power outage command is triggered for the target roadway; when the CO concentration is >24ppm and the diffusion velocity is >0.3m / s, the upstream fan pressurization mode is activated; the personnel positioning terminal receives evacuation path planning data, and the path algorithm prioritizes the airflow ventilation roadway and avoids the predicted fracture zone.
[0015] (III) Technical Effects
[0016] The advantages of this invention compared to the prior art are:
[0017] 1. The multi-source sensing module uses a variety of sensors, such as vibration sensor array, shock wave overpressure sensor, and toxic gas monitoring unit, deployed in a spatial grid topology structure. It can collect data such as vibration, shock wave, toxic gas concentration, sound wave and video during the blasting process in all directions. Compared with traditional single vibration monitoring, it can capture blasting safety hazards more comprehensively and improve the accuracy and comprehensiveness of monitoring.
[0018] 2. In the dynamic risk assessment module, the blasting energy inversion submodule reconstructs the spatial distribution of blasting energy based on vibration and shock wave data, and the secondary disaster prediction submodule combines geological models and gas concentration data to generate a toxic gas diffusion path model, establishing the correlation between blasting energy and secondary disasters. This allows for accurate prediction of the probability and scope of secondary disasters, facilitating the formulation of preventive measures in advance.
[0019] 3. The intelligent early warning decision module integrates a multi-level response strategy library and a deep learning early warning model. By combining video behavior recognition and acoustic anomaly detection results, it dynamically adjusts the early warning threshold, which can adapt to complex geological conditions and changes in blasting conditions, avoid false alarms and missed alarms, and achieve intelligent and flexible early warning decision-making.
[0020] 4. The emergency response module connects to the ventilation system, personnel positioning terminal and audible and visual alarms via an intrinsically safe communication bus. Based on the warning level, it triggers directional power outage, forced ventilation or personnel evacuation commands. The roadway zone control logic is scientific and reasonable, which can quickly and efficiently carry out emergency response to ensure personnel safety and stable operation of mining equipment. Attached Figure Description
[0021] Figure 1 This is a schematic diagram of a mine blasting safety monitoring and early warning system according to the present invention. Detailed Implementation
[0022] The present invention will now be described in further detail with reference to the accompanying drawings.
[0023] Combined with appendix Figure 1 A mine blasting safety monitoring and early warning system includes a multi-source sensing module, an edge computing node, a dynamic risk assessment module, an intelligent early warning decision-making module, and an emergency response execution module. The multi-source sensing module comprises a vibration sensor array, a shock wave overpressure sensor, a toxic gas monitoring unit, a sound wave pickup, and a high-definition video monitoring unit. Each sensor is deployed in a spatial grid topology in the roadways surrounding the blasting point and in key surface areas. The edge computing node is connected to the multi-source sensing module via an industrial-grade wireless mesh network. A real-time data preprocessing unit is configured to filter out environmental noise and extract the time-frequency characteristics of the sensor data. The dynamic risk assessment module is used to input the preprocessed data from the edge computing node. The system includes a blasting energy inversion submodule and a secondary disaster prediction submodule. The blasting energy inversion submodule reconstructs the spatial distribution of blasting energy based on vibration and shock wave data. The secondary disaster prediction submodule generates a toxic gas diffusion path model by combining geological models and gas concentration data. The intelligent early warning decision module receives the output of the dynamic risk assessment module. It integrates a multi-level response strategy library and a deep learning early warning model. The early warning model dynamically adjusts the early warning threshold by fusing video behavior recognition and acoustic anomaly detection results. The emergency linkage execution module connects to the ventilation system, personnel positioning terminal, and audible and visual alarms via a mine-use intrinsically safe communication bus. It triggers directional area power outages, forced ventilation, or personnel evacuation commands based on the early warning level.
[0024] The vibration sensor array in the multi-source sensing module uses a three-component MEMS accelerometer, arranged in three dimensions according to the radial, tangential, and vertical directions of the blast point, with a sampling frequency ≥2kHz, and uses an adaptive Kalman filter algorithm to eliminate mechanical vibration interference in the tunnel.
[0025] The edge computing node has a built-in blasting vibration signal feature extraction unit, which is used to identify the first arrival time of the P wave and the maximum amplitude of the S wave in the waveform. The extraction method includes: performing wavelet packet transform on the original signal; detecting the arrival point of the P wave through the Teager-Kaiser energy operator; and combining Hilbert transform to obtain the extreme points of the S wave envelope, with a positioning error ≤0.1ms.
[0026] The secondary disaster prediction submodule of the dynamic risk assessment module operates as follows: based on the blasting energy inversion results, the discrete element method is used to simulate the rock mass fracture propagation path; the fracture network data is input into the computational fluid dynamics model to predict the diffusion rate of toxic gas along the fracture; the coordinates of the area where the gas concentration exceeds the limit are output to the early warning decision module, driving the audible and visual alarm to flash red light in the target area.
[0027] The deep learning early warning model of the intelligent early warning decision module has a dual-channel architecture: Channel A is used to receive the time-frequency characteristics of vibration and shock waves and uses a 1D-CNN network to classify the blast intensity level; Channel B is used to process the RGB-D images of the video monitoring unit and uses the YOLOv7 model to detect the lack of personnel protective equipment.
[0028] The emergency linkage execution module includes roadway zoning control logic: when the predicted blasting vibration velocity is >10mm / s, a power outage command is triggered for the target roadway; when the CO concentration is >24ppm and the diffusion velocity is >0.3m / s, the upstream fan pressurization mode is activated; the personnel positioning terminal receives evacuation path planning data, and the path algorithm prioritizes the airflow ventilation roadway and avoids the predicted fracture zone.
[0029] The working principle of this invention is as follows: During the blasting preparation stage, based on the mine's geological structure, tunnel layout, and blasting design plan, the workers deploy the vibration sensor array, shock wave overpressure sensor, toxic gas monitoring unit, acoustic pickup, and high-definition video monitoring unit from the multi-source sensing module in a spatial grid topology structure around the blasting point in key areas of the tunnels and surface. The vibration sensor array uses a three-component MEMS accelerometer, arranged in three dimensions (radial, tangential, and vertical) around the blasting point, with a sampling frequency set to 2kHz. An adaptive Kalman filter algorithm effectively eliminates mechanical vibration interference from the tunnels, ensuring the accuracy of the collected data.
[0030] Once the blasting operation begins, multi-source sensing modules collect various data in real time. A vibration sensor array acquires vibration signals generated by the blast, a shock wave overpressure sensor records shock wave pressure changes, a toxic gas monitoring unit monitors the concentration of toxic gases such as CO, a sound wave pickup captures sound wave information from the blast and the surrounding environment, and a high-definition video monitoring unit captures on-site personnel behavior and environmental conditions. This data is transmitted to edge computing nodes via an industrial-grade wireless mesh network.
[0031] After receiving data, the edge computing node's built-in real-time data preprocessing unit first processes the data to filter out environmental noise. Next, the vibration signal is processed using the blasting vibration signal feature extraction unit. Specifically, the original signal undergoes wavelet packet transform, then the arrival point of the P-wave is detected using the Teager-Kaiser energy operator, and the extreme points of the S-wave envelope are obtained using Hilbert transform. This identifies the initial arrival time of the P-wave and the maximum amplitude of the S-wave within the waveform, with the positioning error controlled within 0.1ms, extracting effective time-frequency features. The processed data is then transmitted to the dynamic risk assessment module.
[0032] The dynamic risk assessment module receives preprocessed data from edge computing nodes. Among them, the blasting energy inversion submodule reconstructs the spatial distribution of blasting energy based on vibration and shock wave data using a specific algorithm. The secondary disaster prediction submodule simulates the rock mass fracture propagation path using the discrete element method based on the blasting energy inversion results. Then, it inputs the fracture network data into the computational fluid dynamics model and combines it with the gas concentration data obtained by the toxic gas monitoring unit to predict the diffusion rate of toxic gas along the fracture. Finally, it outputs the coordinates of the area where the gas concentration exceeds the limit.
[0033] The intelligent early warning decision module receives the output from the dynamic risk assessment module, and its built-in deep learning early warning model comes into play. In the dual-channel architecture, channel A receives the time-frequency characteristics of vibration and shock waves, and classifies the blast intensity level using a 1D-CNN network; channel B processes RGB-D images from the video surveillance unit, using a YOLOv7 model to detect missing personal protective equipment. Simultaneously, the model integrates video behavior recognition and acoustic anomaly detection results, dynamically adjusts the early warning threshold, and combines it with a multi-level response strategy library to determine whether an early warning needs to be issued and to define the warning level.
[0034] Once the intelligent early warning decision module determines the warning level, the emergency response execution module performs corresponding operations via the intrinsically safe communication bus used in mining. For example, when the predicted blasting vibration velocity is >10mm / s, a power outage command is triggered for the target roadway; when the CO concentration is >24ppm and the diffusion velocity is >0.3m / s, the upstream fan pressurization mode is activated; simultaneously, the personnel positioning terminal receives evacuation path planning data, which is generated by a path algorithm, prioritizing ventilation roadways and avoiding predicted fracture zones to guide personnel to evacuate safely, while the audible and visual alarm flashes red light in the target area to remind personnel to pay attention to safety.
[0035] Through the coordinated work of the above modules, this mine blasting safety monitoring and early warning system achieves full-process and all-round monitoring of blasting operations. From data collection, analysis and evaluation to early warning and emergency response, all links work closely together to effectively ensure the safe conduct of mine blasting operations.
[0036] The present invention and its embodiments have been described above. This description is not restrictive, and the accompanying drawings are only one embodiment of the present invention; the actual structure is not limited thereto. In conclusion, if those skilled in the art are inspired by this description and design similar structures and embodiments without departing from the spirit of the invention, such designs should fall within the protection scope of the present invention.
Claims
1. A mine blasting safety monitoring and early warning system, comprising a multi-source sensing module, an edge computing node, a dynamic risk assessment module, an intelligent early warning decision-making module, and an emergency response execution module, characterized in that: The multi-source sensing module includes a vibration sensor array, a shock wave overpressure sensor, a toxic gas monitoring unit, a sound wave pickup, and a high-definition video monitoring unit; each sensor is deployed in a spatial grid topology in the surrounding tunnels and key areas of the surface of the blasting point. The edge computing node is connected to the multi-source sensing module through an industrial-grade wireless mesh network; it is equipped with a real-time data preprocessing unit to filter out environmental noise and extract the time-frequency characteristics of sensor data. The dynamic risk assessment module is used to input preprocessed data from edge computing nodes; it includes a blasting energy inversion submodule and a secondary disaster prediction submodule; the blasting energy inversion submodule reconstructs the spatial distribution of blasting energy based on vibration and shock wave data; the secondary disaster prediction submodule combines geological models and gas concentration data to generate a toxic gas diffusion path model; The intelligent early warning decision module is used to receive the output of the dynamic risk assessment module; it integrates a multi-level response strategy library and a deep learning early warning model; the early warning model dynamically adjusts the early warning threshold by fusing video behavior recognition and sound wave anomaly detection results. The emergency response module is connected to the ventilation system, personnel positioning terminal and audible and visual alarm via an intrinsically safe communication bus for mining; it triggers directional area power outage, forced ventilation or personnel evacuation commands according to the warning level.
2. The mine blasting safety monitoring and early warning system according to claim 1, characterized in that, The vibration sensor array in the multi-source sensing module uses a three-component MEMS accelerometer, arranged in three dimensions according to the radial, tangential, and vertical directions of the blast point, with a sampling frequency ≥2kHz, and uses an adaptive Kalman filter algorithm to eliminate mechanical vibration interference in the tunnel.
3. The mine blasting safety monitoring and early warning system according to claim 1, characterized in that, The edge computing node has a built-in blasting vibration signal feature extraction unit, which is used to identify the first arrival time of the P wave and the maximum amplitude of the S wave in the waveform. The extraction method includes: performing wavelet packet transform on the original signal; detecting the arrival point of the P wave through the Teager-Kaiser energy operator; and combining Hilbert transform to obtain the extreme points of the S wave envelope, with a positioning error ≤0.1ms.
4. The mine blasting safety monitoring and early warning system according to claim 1, characterized in that, The secondary disaster prediction submodule of the dynamic risk assessment module operates as follows: based on the blasting energy inversion results, the discrete element method is used to simulate the rock mass fracture propagation path; the fracture network data is input into the computational fluid dynamics model to predict the diffusion rate of toxic gas along the fracture; the coordinates of the area where the gas concentration exceeds the limit are output to the early warning decision module, driving the audible and visual alarm to flash red light in the target area.
5. A mine blasting safety monitoring and early warning system according to claim 1, characterized in that, The deep learning early warning model of the intelligent early warning decision module has a dual-channel architecture: Channel A is used to receive the time-frequency characteristics of vibration and shock waves and uses a 1D-CNN network to classify the blast intensity level; Channel B is used to process the RGB-D images of the video monitoring unit and uses the YOLOv7 model to detect the lack of personnel protective equipment.
6. The mine blasting safety monitoring and early warning system according to claim 1, characterized in that, The emergency linkage execution module includes roadway zoning control logic: when the predicted blasting vibration velocity is >10mm / s, a power outage command is triggered for the target roadway; when the CO concentration is >24ppm and the diffusion velocity is >0.3m / s, the upstream fan pressurization mode is activated; the personnel positioning terminal receives evacuation path planning data, and the path algorithm prioritizes the airflow ventilation roadway and avoids the predicted fracture zone.
Citation Information
Cited By
Copper ore blasting construction monitoring system based on cloud platform
CN121935715A
Real-time monitoring and early warning device for multiple gases in closed space and early warning method thereof
CN122238424A
