Excavation part surrounding rock image-based blasting parameter real-time adjustment technology

Abstract

The invention discloses an excavation part surrounding rock image-based blasting parameter real-time adjustment technology, which comprises the following working steps of: S1, equipment selection and arrangement: selecting photographic equipment with a frame rate of 100,000 frames per second and a resolution ratio of 4K or more, arranging a plurality of cameras to shoot a blasting process from different angles, and then configuring strong lighting equipment, and a strain sensor, a vibration sensor, a pressure sensor and a sonic sensor are installed, and blasting data are collected in real time. By improving the performance of the high-speed photographic equipment and applying technical means such as a three-dimensional reconstruction technology, a real-time feedback control system, AI auxiliary analysis, sensor fusion, numerical simulation, unmanned operation and the like, the precision and efficiency of researching the expansion rule of the blasting crack can be further improved. According to the improved technology, more accurate and dynamic support is provided for blasting design, blasting damage control is optimized, and the safety and efficiency of engineering application are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of surrounding rock images of excavation sites, and in particular to a real-time adjustment technology of blasting parameters based on surrounding rock images of excavation sites. Background Art

[0002] Blasting is widely used in geotechnical engineering, such as mining, tunnel construction, slope stabilization, etc. The mechanism of blasting and fragmentation involves a complex crack propagation process, and its research can help further optimize blasting design and improve operation efficiency.

[0003] However, the traditional blasting crack observation method has limitations, it is difficult to accurately record the expansion process of cracks at the moment of blasting, and lacks quantitative analysis of crack expansion. The existing technology mainly relies on static monitoring methods, which cannot fully reflect the dynamic changes in the blasting process. Therefore, a new experimental method is urgently needed to intuitively record and analyze the crack expansion law.

[0004] As mentioned above, we have designed a real-time adjustment technology of blasting parameters based on the surrounding rock image of the excavation site to solve the above problems. Summary of the invention

[0005] The purpose of the present invention is to solve the shortcomings of the prior art and to propose a real-time adjustment technology of blasting parameters based on surrounding rock images of excavation sites.

[0006] In order to achieve the above object, the present invention adopts the following technical solutions: A real-time adjustment technology of blasting parameters based on surrounding rock images of excavation sites includes the following working steps: Step S1, equipment selection and arrangement: select a photographic device with a frame rate of 100,000 frames per second and a resolution of 4K or above, and arrange several cameras to shoot the blasting process from different angles. Secondly, configure powerful lighting equipment to ensure clear images during the blasting process, and install strain, vibration, pressure and sound wave sensors to collect blasting data in real time; Step S2, data collection: calibrate the equipment in step S1, calibrate all cameras and sensors, ensure equipment time synchronization, start high-speed photography and sensors, and collect real-time data of the blasting process at the moment of blasting; Step S3: Data processing and analysis phase: Data processing includes: pre-processing of high-speed photography images such as denoising, contrast enhancement, and background removal, generating a three-dimensional reconstruction model of the crack, extracting crack features, and conducting quantitative analysis; Data analysis includes: based on historical data and experimental images, using deep learning to train AI models, learning the laws and characteristics of crack extension, using the trained deep learning model to analyze the crack extension process, automatically annotating crack characteristics, predicting crack extension trends, and providing real-time feedback on extension laws; and fusing strain, vibration, pressure and other sensor data with image data to analyze the physical mechanism of crack extension.

[0007] Step S4, real-time feedback and blasting adjustment: Based on the real-time monitored crack extension data, the blasting design is adjusted through the feedback control system to form a real-time control system control. The real-time control system control content includes one or more of the blasting delay time and the blasting charge amount. According to the experimental data and real-time feedback, the subsequent blasting design is adjusted to optimize the crack extension path and the destruction range. The blasting effect is adjusted by the real-time control system calculation; Step S5, data report and result analysis: Based on the high-speed photography and sensor data, combined with the analysis results of the AI ​​model in step S3, an experimental report is automatically generated; the experimental report includes a timing diagram of crack extension, crack feature statistics, and prediction results, and based on the data analysis results, combined with the real-time control system in step S4, analysis and adjustment are performed to obtain the final parameter adjustment results.

[0008] Preferably, in step S1, 3-4 cameras are provided and arranged at different angles and directions, and run synchronously during operation, and the site where the photographic equipment is arranged is one or more of a concrete slope or a rock slope; Powerful lighting equipment includes powerful flashlights, LED lighting or laser light sources.

[0009] Preferably, in step S1, the powerful lighting equipment includes a powerful flash lamp, LED lighting or a laser light source.

[0010] Preferably, in step S2, data collection further includes blasting execution and data collection, and the blasting execution and data collection include: Countdown and control before blasting: Through the electronic detonation system, the detonation delay time is set according to the design plan to ensure that all blasting holes are detonated in the predetermined order; Real-time monitoring and image acquisition: Start the high-speed camera to shoot at 100,000 frames per second at the moment of the explosion, and start all sensors to record real-time data; Synchronous collection of sensor data: strain gauges, vibration sensors, pressure sensors, etc. synchronously collect data at the moment of blasting and transmit them to the real-time data processing system at the same time.

[0011] Preferably, in step S3, the three-dimensional reconstruction of the model includes using computer vision technology to register images at different angles, and using three-dimensional reconstruction software including one or more of Agisoft and Photomodeler to generate three-dimensional point cloud data of crack extension; Extracting cracks includes extracting geometric features of cracks through image processing algorithms: including one or more of width, length, and expansion speed, to provide quantitative data for subsequent analysis.

[0012] Preferably, in step S3, training the AI ​​model includes: based on historical data and experimental images, using deep learning to train the AI ​​model to learn the laws and characteristics of crack extension.

[0013] Preferably, the step S3 further includes: Automatic labeling and result verification: The AI ​​model automatically labels the characteristics of the cracks and compares them with the sensor data to verify the accuracy of the AI ​​prediction; Sensor data integration: The data from strain, vibration, pressure and acoustic wave sensors are integrated with the image data, and the crack propagation path and the physical process of crack generation are obtained through data fusion algorithms such as Kalman filtering or particle filtering.

[0014] Compared with the prior art, the invention has the following beneficial effects: the invention can further improve the accuracy and efficiency of studying the expansion law of blasting cracks by improving the performance of high-speed photography equipment, applying three-dimensional reconstruction technology, real-time feedback control system, AI-assisted analysis, sensor fusion, numerical simulation and unmanned operation. The improved technology will provide more accurate and dynamic support for blasting design, optimize blasting damage control, and improve the safety and effect of engineering applications. DETAILED DESCRIPTION

[0015] The technical solutions in the embodiments of the present invention will be clearly and completely described below in combination with the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments. Example

[0016] The present invention provides a method for studying the expansion law of blasting cracks based on high-speed photography and image processing, aiming to record the dynamic expansion process of blasting cracks by using high-speed photography technology, and perform data analysis through computer image processing technology, thereby revealing the law of crack expansion and providing a scientific basis for blasting design optimization.

[0017] A real-time adjustment technology of blasting parameters based on surrounding rock images of excavation sites includes the following working steps: Step S1, equipment selection and arrangement: During operation, high-speed photography equipment with high frame rate and high resolution needs to be selected: Choose a high-speed camera that supports higher frame rates (e.g., more than 100,000 frames per second) and higher resolutions (e.g., 4K or higher) to ensure that even the smallest crack propagation details during the blasting process can be captured.

[0018] Compare the performance of different devices to ensure that the camera can capture the full picture of the blasting crack in a very short time.

[0019] Optimize lighting conditions: Design and build a high-intensity, uniform light source system, using high-power LEDs or flash lamps to ensure that the crack propagation area can receive sufficient illumination during high-speed photography.

[0020] Use light source adjustment equipment, such as spotlights, reflectors, etc., to avoid light obstruction caused by dust and smoke generated by the blasting and ensure clear images.

[0021] Performing High Dynamic Range (HDR) processing: Combined with HDR imaging technology, images under different lighting conditions are captured during the blasting process, and high dynamic range images are synthesized through software post-production to optimize the brightness and contrast of the image and reduce the loss of highlights and shadows.

[0022] Equipped with a camera sensor suitable for high-contrast scenes, it ensures a good exposure balance in momentary high light conditions.

[0023] Test and calibrate equipment: Standardized testing of high-speed photography equipment ensures that the equipment can work properly under specific environmental conditions. Calibration is performed based on factors such as lighting and climate at the experimental site to ensure high-quality images.

[0024] Expected results: Improving the resolution and clarity of images, especially in capturing the rapidly changing crack propagation process, can obtain more details and reduce image blur caused by environmental factors (such as light and dust).

[0025] Step S2: Data collection: To set up multiple camera positions: Install multiple high-speed cameras at the blasting site to simultaneously capture the crack expansion process from different angles. The cameras should be arranged to cover the main direction of crack expansion to ensure that image data can be obtained from different perspectives.

[0026] The camera should be positioned to cover the entire crack expansion process, and a tripod or stable platform should be used to avoid camera shake.

[0027] Simultaneous shooting and data recording: Ensure that all cameras are time synchronized (using a high-precision synchronizer or NTP time server) to ensure that image data from different angles can be aligned and avoid image registration errors caused by time differences.

[0028] Equipped with large-capacity storage devices to ensure that high-speed photography equipment can simultaneously record high-frame rate image data from multiple angles.

[0029] Image registration and 3D reconstruction: Use image processing algorithms (such as feature point-based registration methods or multi-view stereo matching algorithms) to register images taken at different angles to generate three-dimensional point cloud data of crack extension.

[0030] Use 3D reconstruction software (such as Photomodeler, Agisoft Metashape, etc.) to synthesize multi-angle images into a 3D model to show the 3D expansion process of the crack in detail.

[0031] Data Analysis and Visualization: The three-dimensional reconstruction results are imported into the data analysis software to analyze the spatial characteristics of crack extension, including crack depth, length, and extension speed.

[0032] Using three-dimensional visualization tools, the crack propagation process is dynamically displayed for further quantitative analysis.

[0033] Expected results: Through simultaneous shooting at multiple angles combined with 3D reconstruction technology, more comprehensive and accurate crack extension data can be obtained, revealing the expansion characteristics of cracks in three-dimensional space.

[0034] Step S3: Data processing and analysis phase: Data collection and processing: During the blasting process, image data captured by high-speed cameras, data from dynamic strain sensors (such as strain changes when cracks appear), and other real-time monitoring data (such as pressure and vibration) are collected in real time.

[0035] All collected data are synchronously analyzed through a real-time data processing system to display the crack expansion and related physical quantities (such as crack width, velocity, strain distribution, etc.) in real time.

[0036] Real-time feedback control system: Based on the results of image processing and strain data analysis, a feedback control mechanism is established. For example, if the crack extension speed is too fast or exceeds the predetermined range, the system can automatically adjust the detonation delay or blasting charge.

[0037] Through the electronic detonation system, real-time data is fed back to the detonation device, and the blasting parameters are adjusted in real time to optimize the crack expansion process.

[0038] Dynamically optimized blasting design: During the experiment or actual blasting process, the blasting plan is dynamically adjusted using the real-time monitoring data analysis results. According to the expansion of the cracks, the detonation delay time of other blasting holes is adjusted in time to ensure the maximum blasting effect.

[0039] Expected results: By achieving dynamic adjustment and optimization of the blasting process and through real-time data feedback, the crack expansion path and damage range can be controlled more accurately, thereby improving blasting efficiency and safety.

[0040] Step S4: Real-time feedback and blasting adjustment: Data preprocessing: Preprocess the images acquired by high-speed photography, including denoising, contrast enhancement, and background interference removal, to ensure that the input data is suitable for AI analysis.

[0041] Perform edge detection and feature extraction on the crack image to clarify the specific location and shape of the crack.

[0042] AI model training: Deep learning techniques (such as convolutional neural networks (CNNs)) are used to train the crack propagation process, and the training dataset contains image data of crack propagation under different blasting conditions.

[0043] During the training process, the network learns the crack expansion pattern, expansion speed, crack morphology changes and other laws, and performs classification and prediction.

[0044] Real-time crack identification and propagation prediction: Use the trained deep learning model to automatically analyze the high-speed photography images acquired in real time, identify the expansion state of the cracks, and predict the expansion trend of the cracks in the next milliseconds or seconds.

[0045] For complex crack shapes, AI can help automatically label crack features, such as crack width, length, shape and other parameters, and perform statistical analysis.

[0046] Automated report generation and optimization suggestions: The experimental report is automatically generated based on the AI ​​analysis results, including detailed data of the crack propagation process and optimization suggestions.

[0047] Utilize the feedback provided by the AI ​​model to adjust blasting design parameters and optimize future blasting plans.

[0048] Expected results: AI-assisted analysis can improve the automated identification and prediction accuracy of the crack propagation process, reduce manual intervention, and improve analysis efficiency. Especially when faced with complex cracks, AI's predictive capabilities can identify potential problems in advance.

[0049] Step S5, data reporting and result analysis: Select and install a variety of sensors: install strain gauges, pressure sensors, accelerometers, acoustic wave sensors, etc., and automatically generate an experimental report based on high-speed photography and sensor data, combined with the analysis results of the AI ​​model in step S3; the experimental report includes a timing diagram of crack expansion, crack feature statistics, and prediction results, and based on the data analysis results, combined with the real-time control system in step S4, analyze and adjust, and then obtain the final parameter adjustment results.

[0050] The workflow of this invention demonstrates how to combine multiple advanced technologies (high-speed photography, AI-assisted analysis, real-time feedback control, sensor fusion, etc.) to achieve efficient and accurate research on the expansion law of blasting cracks. Through this process, not only can the expansion process of cracks be accurately captured, but also the blasting design can be dynamically optimized, the blasting efficiency and safety can be improved, and strong theoretical support and technical guarantee can be provided for future engineering blasting.

[0051] The basic principles, main features and advantages of the present invention are shown and described above. It should be understood by those skilled in the art that the present invention is not limited to the above embodiments. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, which fall within the scope of the present invention. The scope of the present invention is defined by the attached claims and their equivalents.

Claims

1. A real-time adjustment technology for blasting parameters based on surrounding rock images of excavation sites, characterized in that: The following steps are included: Step S1, equipment selection and arrangement: select a photographic device with a frame rate of 100,000 frames per second and a resolution of 4K or above, and arrange several cameras to shoot the blasting process from different angles. Secondly, configure powerful lighting equipment to ensure clear images during the blasting process, and install strain, vibration, pressure and sound wave sensors to collect blasting data in real time; Step S2, data collection: calibrate the equipment in step S1, calibrate all cameras and sensors, ensure equipment time synchronization, start high-speed photography and sensors, and collect real-time data of the blasting process at the moment of blasting; Step S3: Data processing and analysis phase: Data processing includes: pre-processing of high-speed photography images such as denoising, contrast enhancement, and background removal, generating a three-dimensional reconstruction model of the crack, extracting crack features, and conducting quantitative analysis; Data analysis includes: based on historical data and experimental images, using deep learning to train AI models, learning the laws and characteristics of crack expansion, using the trained deep learning model to analyze the crack expansion process, automatically annotating crack characteristics, predicting crack expansion trends, and providing real-time feedback on expansion laws; and fusing strain, vibration, pressure and other sensor data with image data to analyze the physical mechanism of crack expansion; Step S4, real-time feedback and blasting adjustment: Based on the real-time monitored crack extension data, the blasting design is adjusted through the feedback control system to form a real-time control system control. The real-time control system control content includes one or more of the blasting delay time and the blasting charge amount. According to the experimental data and real-time feedback, the subsequent blasting design is adjusted to optimize the crack extension path and the destruction range. The blasting effect is adjusted by the real-time control system calculation; Step S5, data report and result analysis: Based on the high-speed photography and sensor data, combined with the analysis results of the AI ​​model in step S3, an experimental report is automatically generated; the experimental report includes a timing diagram of crack extension, crack feature statistics, and prediction results, and based on the data analysis results, combined with the real-time control system in step S4, analysis and adjustment are performed to obtain the final parameter adjustment results.

2. The real-time adjustment technology of blasting parameters based on surrounding rock images of excavation sites according to claim 1 is characterized in that: In step S1, 3-4 cameras are provided and arranged at different angles and directions, and run synchronously during operation, and the site where the photographic equipment is arranged is one or more of a concrete slope or a rock slope; Powerful lighting equipment includes powerful flashlights, LED lighting or laser light sources.

3. The real-time adjustment technology of blasting parameters based on surrounding rock images of excavation sites according to claim 1 is characterized in that: In step S1, the powerful lighting equipment includes a powerful flash lamp, LED lighting or a laser light source.

4. The real-time adjustment technology of blasting parameters based on surrounding rock images of excavation sites according to claim 1 is characterized in that: In step S2, data collection also includes blasting execution and data collection, and the blasting execution and data collection include: Countdown and control before blasting: Through the electronic detonation system, the detonation delay time is set according to the design plan to ensure that all blasting holes are detonated in the predetermined order; Real-time monitoring and image acquisition: Start the high-speed camera to shoot at 100,000 frames per second at the moment of the explosion, and start all sensors to record real-time data; Synchronous collection of sensor data: strain gauges, vibration sensors, pressure sensors, etc. synchronously collect data at the moment of blasting and transmit them to the real-time data processing system at the same time.

5. The real-time adjustment technology of blasting parameters based on surrounding rock images of excavation sites according to claim 1 is characterized in that: In step S3, the three-dimensional reconstruction model includes using computer vision technology to register images at different angles, and using three-dimensional reconstruction software including one or more of Agisoft and Photomodeler to generate three-dimensional point cloud data of crack extension; Extracting cracks includes extracting geometric features of cracks through image processing algorithms: including one or more of width, length, and expansion speed, to provide quantitative data for subsequent analysis.

6. The real-time adjustment technology of blasting parameters based on surrounding rock images of excavation sites according to claim 1 is characterized in that: In step S3, training the AI ​​model includes: based on historical data and experimental images, using deep learning to train the AI ​​model to learn the laws and characteristics of crack expansion.

7. The real-time adjustment technology of blasting parameters based on surrounding rock images of excavation sites according to claim 1 is characterized in that: The step S3 further includes: Automatic labeling and result verification: The AI ​​model automatically labels the characteristics of the cracks and compares them with the sensor data to verify the accuracy of the AI ​​prediction; Sensor data integration: The data from strain, vibration, pressure and acoustic wave sensors are integrated with the image data, and the crack propagation path and the physical process of crack generation are obtained through data fusion algorithms such as Kalman filtering or particle filtering.