Intelligent drilling and blasting monitoring method for long and large tunnel based on multi-sensor fusion

Abstract

The invention discloses an intelligent drilling and blasting monitoring method for a long and large tunnel based on multi-sensor fusion. The intelligent drilling and blasting monitoring method comprises the following steps: 1, arranging wireless sensor network nodes; 2, the drilling simulation parameters of the multiple wireless sensor network nodes are input into a tunnel intelligent drilling and blasting monitoring neural network, and a set of optimal drilling parameters under the tunnel construction safety state are obtained; 3, the optimal drilling parameters of the multiple wireless sensor network nodes are used for drilling and blasting the tunnel; fourthly, the parameters after blasting and the virtual parameters after blasting are compared, and the tunnel state under the parameters after blasting is judged; when the tunnel state is the tunnel construction safety critical state, the fifth step is executed; when the tunnel state is a tunnel construction dangerous state, an early warning system is triggered, and after the early warning system is triggered, an emergency response mechanism is automatically started; and 5, dynamically adjusting the positions and the number of the nodes of the wireless sensor network, and drilling parameters. According to the invention, active quantitative control of drilling and blasting parameters can be realized, and the construction efficiency and the quality control level are improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of drill and blast monitoring, and particularly relates to an intelligent drill and blast monitoring method for long tunnels based on multi-sensor fusion. Background Technique

[0002] During the construction of long tunnels, drill and blast operations are a key link, which involves multiple aspects such as tunnel excavation and tunneling. However, due to the complexity and uncertainty of the tunnel construction environment, the safety and efficiency of drill and blast operations are often affected by many factors. Drill and blast operations mainly include drilling and blasting, which jointly act on the excavation and tunneling of the tunnel. Intelligent drill and blast monitoring for long tunnels is a construction method based on modern information technology, aiming to improve the safety, efficiency and quality control level of tunnel drill and blast operations. The existing construction monitoring and control technologies for long tunnels have the following problems: traditional monitoring systems mostly use single sensors for data collection, resulting in ineffective information fusion between sensors, limited monitoring effects, and difficulty in achieving global optimal analysis; the monitoring system responds slowly to emergencies and cannot achieve fast and accurate early warning and control; most of the existing monitoring systems use a control method of timing plus manual operation, which not only increases energy consumption but also easily leads to a decrease in control accuracy. Summary of the Invention

[0003] The technical problem to be solved by the present invention is to provide an intelligent drill and blast monitoring method for long tunnels based on multi-sensor fusion in view of the above deficiencies in the prior art.

[0004] To solve the above technical problem, the technical solution adopted by the present invention is: an intelligent drill and blast monitoring method for long tunnels based on multi-sensor fusion, characterized in that the method includes the following steps:

[0005] Step 1: Arrange wireless sensor network nodes: Arrange a plurality of wireless sensor network nodes in the internal and surrounding areas of the tunnel. The data collected by the wireless sensor network nodes includes drilling parameters and post-blasting parameters. The drilling parameters include drill rig status, drill rig rotation speed, drill rig working pressure, drilling depth, drilling spacing, charge amount in the hole, environmental parameters, drill rig position and geological parameters; the post-blasting parameters include vibration intensity, blasting noise sound and deformation displacement;

[0006] Step 2: Input the drilling simulation parameters of multiple wireless sensor network nodes into the intelligent drill and blast monitoring neural network of the tunnel to obtain a set of optimal drilling parameters under the safe state of tunnel construction. The intelligent drill and blast monitoring neural network of the tunnel receives the drilling simulation parameters and outputs virtual post-blasting parameters;

[0007] Step 3: Use the optimal drilling parameters of multiple wireless sensor network nodes to drill and blast the tunnel. Use multiple wireless sensor network nodes to collect the parameters after blasting in real time, and perform spatio-temporal alignment and redundancy removal on the optimal drilling parameters and the parameters after blasting of multiple wireless sensor network nodes to form a unified data format and spatio-temporal reference;

[0008] Step 4: Compare the parameters after blasting with the virtual parameters after blasting to judge the state of the tunnel under the parameters after blasting. The tunnel state includes the tunnel construction safety state, the tunnel construction safety critical state, and the tunnel construction dangerous state;

[0009] When the tunnel state is the tunnel construction safety state, execute Step 2;

[0010] When the tunnel state is the tunnel construction safety critical state, execute Step 5;

[0011] When the tunnel state is the tunnel construction dangerous state, trigger the early warning system. After the early warning system is triggered, automatically start the emergency response mechanism;

[0012] Step 5: Dynamically adjust the positions and quantities of wireless sensor network nodes, as well as the drilling parameters.

[0013] The above intelligent drilling and blasting monitoring method for long tunnels based on multi-sensor fusion is characterized in that: in Step 1, the specific steps of arranging wireless sensor network nodes include:

[0014] Determine the parameters and environmental factors that need to be monitored during the tunnel drilling and blasting operation;

[0015] Considering factors such as the length, shape, and geological conditions of the tunnel, determine the layout requirements and coverage range of the sensors;

[0016] Carry out layout planning in the internal and surrounding areas of the tunnel to determine the installation positions of the sensors. Considering distributed layout, make the sensors cover all key parts of the tunnel, including the drilling and blasting operation area, tunnel wall, vault, and arch bottom.

[0017] The above intelligent drilling and blasting monitoring method for long tunnels based on multi-sensor fusion is characterized in that: before obtaining the tunnel intelligent drilling and blasting monitoring neural network, it further includes:

[0018] Use the tunnel drilling and blasting historical data set to train the pre-constructed drilling and blasting monitoring neural network to obtain a trained drilling and blasting monitoring neural network;

[0019] Transfer the shared parameters of the drilling and blasting monitoring neural network to the tunnel intelligent drilling and blasting monitoring neural network through transfer learning;

[0020] The intelligent tunneling drilling and blasting monitoring neural network described above includes a VGG convolutional neural network. The VGG convolutional neural network includes an input layer, a feature extraction module, a fully connected layer, a Softmax classifier, and an output layer. The input of the input layer is a grayscale image after fusing drilling simulation parameters, and the output layer is the expected vector of the tunnel state corresponding to the parameters after blasting.

[0021] The above-mentioned intelligent tunneling drilling and blasting monitoring method based on multi-sensor fusion is characterized in that: in step two, the specific steps of spatio-temporal alignment and redundancy removal for the optimal drilling parameters and the parameters after blasting of multiple wireless sensor network nodes include:

[0022] Perform time alignment on the collected data to determine a unified time reference, align the data from different sensors according to the timestamps, and according to the formula t' = t i -t 0 , perform time alignment on the timestamp t i of sensor i, where t 0 is the unified time reference and t' is the aligned time of sensor i;

[0023] Perform spatial alignment on the collected data. According to the position information of the sensors and the geometric model of the tunnel, map the data into a unified spatial coordinate system. The position coordinates of sensor i are determined, and the geometric model of the tunnel is a known function. Map the data of sensor i into the unified spatial coordinate system;

[0024] Compare the aligned data, identify duplicate and similar data, delete redundant data, and retain representative data;

[0025] Convert the data from different sensors into a unified data format and establish a spatio-temporal reference for the data in the unified format.

[0026] The above-mentioned intelligent tunneling drilling and blasting monitoring method based on multi-sensor fusion is characterized in that: in step five, the specific steps of dynamically adjusting the positions and quantities of wireless sensor network nodes and the drilling parameters include:

[0027] Increase the layout density of wireless sensor network nodes adjusted dynamically;

[0028] Reduce the drilling speed, the working pressure of the drill, the drilling depth, the drilling spacing, and the charge amount in the hole.

[0029] The above-mentioned intelligent tunneling drilling and blasting monitoring method based on multi-sensor fusion is characterized in that: in step four, the emergency response mechanism is to notify the personnel in the tunnel to evacuate according to the predetermined evacuation route and start the ventilation system and the safety facilities of the protective doors.

[0030] The present invention uses multi-sensor fusion technology, including drilling parameters and post-blasting parameters. The drilling parameters include the drill rig status, drill rig rotation speed, drill rig working pressure, drilling depth, drilling spacing, charge amount in the hole, environmental parameters, drill rig position, and geological parameters; the post-blasting parameters include vibration intensity, blasting noise sound, and deformation displacement; the drilling parameters and post-blasting parameters work together to improve the comprehensiveness and accuracy of the monitoring system, achieve precise monitoring of the tunnel drilling and blasting operation, combine real-time monitoring with intelligent analysis, quickly identify potential hazards, and achieve rapid early warning and emergency response, improve construction safety, can achieve active quantitative control of drilling and blasting parameters, improve construction efficiency and quality control level, effectively reduce the energy consumption of the monitoring system, meet the requirements of energy conservation and environmental protection, and are convenient for popularization and use.

[0031] The following is a further detailed description of the technical solution of the present invention through the accompanying drawings and embodiments. Description of the Drawings

[0032] Figure 1 It is a flow block diagram of the present invention. Detailed Embodiments

[0033] As Figure 1 shown, an intelligent drilling and blasting monitoring method for long tunnels based on multi-sensor fusion of the present invention includes the following steps:

[0034] Step 1: Arrange wireless sensor network nodes: Arrange multiple wireless sensor network nodes inside and around the tunnel. The data collected by the wireless sensor network nodes includes drilling parameters and post-blasting parameters. The drilling parameters include the drill rig status, drill rig rotation speed, drill rig working pressure, drilling depth, drilling spacing, charge amount in the hole, environmental parameters, drill rig position, and geological parameters; the post-blasting parameters include vibration intensity, blasting noise sound, and deformation displacement; the environmental parameters include temperature and humidity parameters.

[0035] Step 2: Input the drilling simulation parameters of multiple wireless sensor network nodes into the tunnel intelligent drilling and blasting monitoring neural network to obtain a set of optimal drilling parameters under the safe state of tunnel construction. The tunnel intelligent drilling and blasting monitoring neural network receives the drilling simulation parameters and outputs virtual post-blasting parameters;

[0036] Step 3: Use the optimal drilling parameters of multiple wireless sensor network nodes to perform drilling and blasting on the tunnel, use multiple wireless sensor network nodes to collect post-blasting parameters in real time, and perform spatio-temporal alignment and redundancy removal processing on the optimal drilling parameters and post-blasting parameters of multiple wireless sensor network nodes to form a unified data format and spatio-temporal reference;

[0037] Step 4: Compare the parameters after blasting with the virtual parameters after blasting, and judge the state of the tunnel under the parameters after blasting. The tunnel state includes the tunnel construction safety state, the tunnel construction safety critical state, and the tunnel construction danger state;

[0038] When the tunnel state is the tunnel construction safety state, execute Step 2;

[0039] When the tunnel state is the tunnel construction safety critical state, execute Step 5;

[0040] When the tunnel state is the tunnel construction danger state, trigger the early warning system. After the early warning system is triggered, automatically start the emergency response mechanism;

[0041] Step 5: Dynamically adjust the positions and quantities of the wireless sensor network nodes, as well as the drilling parameters.

[0042] In this embodiment, in Step 1, the specific steps of arranging the wireless sensor network nodes include:

[0043] Determine the parameters and environmental factors that need to be monitored during the tunnel drilling and blasting operation;

[0044] Consider the length, shape, and geological condition factors of the tunnel to determine the arrangement requirements and coverage range of the sensors;

[0045] Conduct layout planning in the internal and peripheral areas of the tunnel to determine the installation positions of the sensors. Considering the distributed layout, make the sensors cover each key part of the tunnel, including the drilling and blasting operation area, the tunnel wall, the vault, and the arch bottom.

[0046] In this embodiment, before obtaining the tunnel intelligent drilling and blasting monitoring neural network, it further includes:

[0047] Use the tunnel drilling and blasting historical data set to train the pre-constructed drilling and blasting monitoring neural network to obtain a trained drilling and blasting monitoring neural network;

[0048] Transfer the shared parameters of the drilling and blasting monitoring neural network to the tunnel intelligent drilling and blasting monitoring neural network through transfer learning;

[0049] The tunnel intelligent drilling and blasting monitoring neural network includes a VGG convolutional neural network. The VGG convolutional neural network includes an input layer, a feature extraction module, a fully connected layer, a Softmax classifier, and an output layer. The input of the input layer is the grayscale image after the fusion of the drilling simulation parameters, and the output layer is the expected vector of the tunnel state corresponding to the parameters after blasting.

[0050] In this embodiment, in Step 2, the specific steps of performing spatio-temporal alignment and redundancy removal processing on the optimal drilling parameters and the parameters after blasting of multiple wireless sensor network nodes include:

[0051] Perform time alignment on the collected data to determine a unified time reference, align the data from different sensors according to the timestamps, and according to the formula t' = t i - t 0 , perform time alignment on the timestamp t i of sensor i, where t 0 is the unified time reference and t' is the aligned time of sensor i;

[0052] Perform spatial alignment on the collected data. According to the position information of the sensors and the geometric model of the tunnel, map the data to a unified spatial coordinate system. The position coordinates of sensor i are determined, and the geometric model of the tunnel is a known function. Map the data of sensor i to the unified spatial coordinate system;

[0053] Compare the aligned data, identify duplicate and similar data, delete redundant data, and retain representative data;

[0054] Convert the data from different sensors into a unified data format and establish a spatio-temporal reference for the data in the unified format.

[0055] In this embodiment, in step five, the specific steps for dynamically adjusting the positions and quantities of the wireless sensor network nodes and the drilling parameters include:

[0056] Increase the layout density of the dynamically adjustable wireless sensor network nodes;

[0057] Reduce the drilling speed, drilling working pressure, drilling depth, drilling spacing, and charge amount in the hole.

[0058] In this embodiment, in step four, the emergency response mechanism is to notify the personnel in the tunnel to evacuate according to the predetermined evacuation route and activate the safety facilities of the ventilation system and the protective door.

[0059] When implementing the present invention, the wireless sensor network technology is adopted to realize the real-time collection of sensor data and perform preliminary processing. Through the wireless transmission module, the preprocessed data is transmitted to the monitoring center and the cloud server in real time.

[0060] The specific steps for the layout of the wireless sensor network nodes further include:

[0061] Install the sensors and perform debugging according to the layout plan in the above steps, calibrate and test the sensors, and establish the connection between the sensors and the data acquisition system;

[0062] During the tunnel drilling and blasting operation, monitor the data of the sensors in real time, analyze and verify the monitored data, and adjust and optimize the layout of the sensors according to the monitoring results.

[0063] Deploy wireless sensor network nodes inside and around the tunnel. Each node contains multiple sensors, and the distribution of the wireless sensor network nodes covers the entire monitoring area;

[0064] The sensors convert the collected data into electrical signals and digital signals;

[0065] The wireless sensor network nodes perform preliminary processing on the collected data, remove noise and interference signals in the data, improve the quality of the data, and compress the data to reduce the data volume.

[0066] Configure a wireless transmission module in the wireless sensor network node for sending the preprocessed data;

[0067] Set the parameters of the wireless transmission module, including transmission frequency, power, and communication protocol;

[0068] The wireless sensor network nodes transmit the preprocessed data to nearby gateways and relay nodes in real time through the wireless transmission module;

[0069] The gateways and relay nodes receive data from multiple wireless sensor network nodes, summarize it, and transmit the data to the monitoring center through the wireless network. At the same time, the data is also uploaded to the cloud server in real time;

[0070] During the data transmission process, encrypt the data using encryption technology, and at the same time, use a verification mechanism to verify the data;

[0071] The monitoring center and the cloud server receive data from the sensor network and perform further processing and analysis on the data.

[0072] The tunnel intelligent drilling and blasting monitoring neural network can also be replaced by using Bayesian networks or fuzzy logic.

[0073] The results after data fusion of the data collected by the wireless sensor network nodes can generate corresponding charts, including line charts, bar charts, and pie charts, intuitively showing the key parameters and potential risks;

[0074] Write the analysis results in the form of a report, detailing the status of the tunnel drilling and blasting operation, existing problems, and recommended measures;

[0075] Present the generated charts and reports to the monitoring personnel through the interface of the monitoring system.

[0076] Through multi-sensor fusion technology, including drilling parameters and post-blasting parameters, the drilling parameters include drill rig status, drill rig rotation speed, drill rig working pressure, drilling depth, drilling spacing, charge amount in the hole, environmental parameters, drill rig position, and geological parameters; the post-blasting parameters include vibration intensity, blasting noise sound, and deformation displacement; the drilling parameters and post-blasting parameters work together to improve the comprehensiveness and accuracy of the monitoring system, achieve precise monitoring of tunnel drilling and blasting operations, combine real-time monitoring with intelligent analysis, quickly identify potential hazards, and achieve rapid early warning and emergency response, improve construction safety, enable active quantitative control of drilling and blasting parameters, improve construction efficiency and quality control level, effectively reduce the energy consumption of the monitoring system, and meet the requirements of energy conservation and environmental protection.

[0077] The above are only the preferred embodiments of the present invention, and do not impose any limitations on the present invention. Any simple modifications, changes, and equivalent structural changes made to the above embodiments based on the technical essence of the present invention still fall within the protection scope of the technical solution of the present invention.

Claims

1. A method for intelligent drilling and blasting monitoring of long and large tunnels based on multi-sensor fusion, characterized in that: The method comprises the following steps: Step 1: Arrange wireless sensor network nodes: Arrange multiple wireless sensor network nodes inside and around the tunnel. The data collected by the wireless sensor network nodes include drilling parameters and post-blasting parameters. The drilling parameters include drilling rig status, drilling rig speed, drilling rig working pressure, drilling depth, drilling spacing, charge in the hole, environmental parameters, drilling rig position and geological parameters; the post-blasting parameters include vibration intensity, blasting noise and deformation displacement; Step 2: input the drilling simulation parameters of multiple wireless sensor network nodes into the tunnel intelligent drilling and blasting monitoring neural network to obtain a set of optimal drilling parameters under the safe state of tunnel construction. The tunnel intelligent drilling and blasting monitoring neural network receives the drilling simulation parameters and outputs post-blasting virtual parameters; Step 3: Drill and blast the tunnel using the optimal drilling parameters of multiple wireless sensor network nodes, use multiple wireless sensor network nodes to collect post-blasting parameters in real time, and perform spatiotemporal alignment and de-redundancy processing on the optimal drilling parameters and post-blasting parameters of multiple wireless sensor network nodes to form a unified data format and spatiotemporal reference; Step 4: Compare the post-blasting parameters with the post-blasting virtual parameters to determine the tunnel state under the post-blasting parameters, wherein the tunnel state includes a tunnel construction safety state, a tunnel construction safety critical state, and a tunnel construction dangerous state; When the tunnel state is a safe state for tunnel construction, execute step 2; When the tunnel state is a critical state for tunnel construction safety, execute step 5; When the tunnel state is dangerous for tunnel construction, the early warning system is triggered, and after the early warning system is triggered, the emergency response mechanism is automatically started; Step 5: Dynamically adjust the location and quantity of wireless sensor network nodes and drilling parameters.

2. The method for intelligent drilling and blasting monitoring of long tunnels based on multi-sensor fusion according to claim 1, characterized in that: In step 1, the specific steps of deploying wireless sensor network nodes include: Identify parameters and environmental factors that need to be monitored during tunnel drilling and blasting operations; Consider the length, shape and geological conditions of the tunnel to determine the sensor layout requirements and coverage; Carry out layout planning inside and around the tunnel, determine the installation location of the sensors, and consider distributed layout so that the sensors cover all key parts of the tunnel, including the drilling and blasting operation area, tunnel wall, vault and vault bottom.

3. The method for intelligent drilling and blasting monitoring of long and large tunnels based on multi-sensor fusion according to claim 1, characterized in that: Before obtaining the tunnel intelligent drilling and blasting monitoring neural network, it also includes: The trained drilling and blasting monitoring neural network is obtained by using the tunnel drilling and blasting historical data set training and the pre-built drilling and blasting monitoring neural network; The shared parameters of the drilling and blasting monitoring neural network are transferred to the tunnel intelligent drilling and blasting monitoring neural network through transfer learning; The tunnel intelligent drilling and blasting monitoring neural network includes a VGG convolutional neural network, which includes an input layer, a feature extraction module, a fully connected layer, a Softmax classifier and an output layer. The input layer input is a grayscale image after the drilling simulation parameters are fused, and the output layer is an expected vector of the tunnel state corresponding to the parameters after blasting.

4. The method for intelligent drilling and blasting monitoring of long tunnels based on multi-sensor fusion according to claim 1 is characterized in that: In step 2, the specific steps of performing spatiotemporal alignment and redundancy removal processing on the optimal drilling parameters and post-blasting parameters of multiple wireless sensor network nodes include: Time align the collected data, determine a unified time base, align the data from different sensors according to the timestamp, and use the formula t'=t i -t0, timestamp t for sensor i i Perform time alignment, where t0 is the unified time reference and t' is the alignment time of sensor i; Perform spatial alignment on the collected data. According to the position information of the sensor and the geometric model of the tunnel, the data is mapped to a unified spatial coordinate system. The position coordinates of sensor i are determined, and the geometric model of the tunnel is a known function. The data of sensor i is mapped to the unified spatial coordinate system. Compare the aligned data, identify duplicate and similar data, delete redundant data, and retain representative data; Convert data from different sensors into a unified data format and establish a spatiotemporal reference for the data in the unified format.

5. The method for intelligent drilling and blasting monitoring of long tunnels based on multi-sensor fusion according to claim 1, characterized in that: In step 5, the specific steps of dynamically adjusting the location and number of wireless sensor network nodes and drilling parameters include: Increase dynamic adjustment of wireless sensor network node deployment density; Reduce the drilling rig speed, drilling rig working pressure, drilling depth, drilling spacing and the amount of charge in the hole.

6. The method for intelligent drilling and blasting monitoring of long tunnels based on multi-sensor fusion according to claim 1, characterized in that: In step 4, the emergency response mechanism is to notify the personnel in the tunnel to evacuate according to the predetermined evacuation route and activate the ventilation system and protective door safety facilities.

Images