Method and system for monitoring accumulated damage of reserved rock mass after blasting of section tunnel

Abstract

The invention relates to the technical field of tunnel blasting, in particular to a method and system for monitoring accumulated damage of reserved rock mass after blasting of a section tunnel. The monitoring method comprises the steps that S1, sensing equipment is arranged and used for collecting rock mass parameter data including micro-seismic data and strain data; s2, preprocessing the data acquired by each sensing device, and mapping the rock mass parameter data acquired by each sensing device to the same three-dimensional coordinate system; and S3, calculating a damage index according to the strain energy density, the strain energy density initial value, the micro-seismic event frequency and the geological adjustment coefficient, and performing early warning according to the damage index and a preset early warning rule. According to the method, the micro-seismic data and the strain data of the reserved rock mass are collected and processed, then the damage index is calculated, and early warning is performed according to the damage index and the preset early warning rule, so that the accuracy of monitoring the blasting damage of the surrounding rock is improved, the damage condition of the surrounding rock can be rapidly evaluated, and therefore protection measures can be taken in time, and the safety of the surrounding rock is improved. The stability of the tunnel structure is ensured.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of tunnel blasting, in particular to a method and system for monitoring accumulated damage of reserved rock mass after cross-section tunnel blasting. BACKGROUND

[0002] In tunnel construction, the drill-and-blast method is a long-standing excavation method that is widely used and can adapt to different geological conditions. It is still widely used in many projects. This method achieves rock breaking and removal by drilling holes in the rock mass and loading emulsion explosives for blasting, thereby advancing tunnel excavation. Although this method has certain advantages in terms of excavation efficiency and construction flexibility, during the blasting process, the energy released by the explosives not only breaks the rock that needs to be excavated, but also affects the surrounding rock that does not need to be excavated. This impact mainly manifests as the further expansion of pre-existing rock cracks, the decrease in the strength of some rocks, the weakening of the connection between rock masses, and the overall loosening. These changes reduce the stability of the surrounding rock mass of the tunnel, which may cause problems such as falling and collapse, and adversely affect subsequent support construction and long-term operation safety. In order to ensure the safety of the tunnel, it is necessary to understand the damage to the surrounding rock after blasting in order to take timely reinforcement measures.

[0003] However, at present, the judgment of the damage degree of the surrounding rock after blasting in actual engineering mostly relies on manual observation, drilling sampling or simple vibration monitoring and other means. These methods have limited information and are difficult to fully reflect the development of internal rock cracks and the damage range, especially in the deep rock mass that cannot be seen, the damage state is more difficult to grasp. Due to the lack of effective detection and evaluation methods, construction personnel often cannot accurately determine which areas are severely damaged, leading to insufficient or excessive support measures, affecting construction efficiency and safety. Therefore, the existing technical means have obvious deficiencies in evaluating the actual damage condition of the surrounding rock after drill-and-blast construction, and it is difficult to achieve rapid, accurate and comprehensive judgment, which limits the improvement of tunnel construction quality and safety control level. SUMMARY

[0004] The purpose of the present application is to overcome the deficiency that the prior art cannot accurately evaluate the damage condition of the surrounding rock in drill-and-blast construction, and to provide a method and system for monitoring accumulated damage of reserved rock mass after cross-section tunnel blasting.

[0005] In a first aspect, the present application provides a method for monitoring accumulated damage of reserved rock mass after cross-section tunnel blasting, comprising the following steps: S1, laying out a sensing device for collecting rock mass parameter data; wherein the rock mass parameter data at least includes microseismic data and strain data; S2, pre-process the data collected by each sensing device, and map the rock mass parameter data collected by each sensing device to the same three-dimensional coordinate system; wherein the pre-processing includes noise reduction, filtering processing and data synchronization processing; S3, calculate a damage index, and perform early warning according to the damage index and a preset early warning rule; The calculation of the damage index includes: The strain data is processed by using a temperature-strain decoupling algorithm optimized by a least square method to calculate the strain energy density in real time. The damage index is calculated according to the strain energy density, an initial value of the strain energy density, a microseismic event frequency and a geological adjustment coefficient. The initial value of the strain energy density is calibrated by field acoustic testing, and the microseismic event frequency is obtained from the microseismic data.

[0006] According to a preferred embodiment, the sensing device includes at least a plurality of microseismic sensors. The plurality of microseismic sensors are arranged along the tunnel axis to collect microseismic data in the blasting influence area and the potential damage propagation area.

[0007] According to a preferred embodiment, the sensing device further includes a plurality of strain sensors. The plurality of strain sensors are arranged at key positions of the reserved rock mass to collect strain data of the reserved rock mass. Preferably, the key positions include at least a crown, a haunch and a sidewall.

[0008] According to a preferred embodiment, S2 includes applying wavelet threshold denoising to the microseismic data, and applying sliding average filtering to the strain data; and performing data synchronization by using a dynamic time warping algorithm.

[0009] According to a preferred embodiment, the early warning rule in S3 includes comparing the damage index with a preset threshold to determine the safety state of the tunnel. The preset threshold includes a first threshold and a second threshold greater than the first threshold. In the case that the damage index is less than the first threshold, the safety state of the tunnel is safe; in the case that the damage index is greater than or equal to the first threshold and less than the second threshold, the safety state of the tunnel is early warning; and in the case that the damage index is greater than or equal to the second threshold, the safety state of the tunnel is dangerous. Preferably, the first threshold and the second threshold are set according to the surrounding rock grade and the hydrogeological condition.

[0010] According to a preferred embodiment, the sensing device further comprises a plurality of displacement sensors for monitoring displacement data of the reserved rock mass. The microseismic data, the strain data, and the displacement data are mapped to the same three-dimensional coordinate system in combination with the BIM model. Feature extraction is performed on the microseismic data, the strain data, and the displacement data, and the extracted features are input into a multi-modal deep learning model based on a Transformer network to obtain a three-dimensional damage hotspot probability map. Preferably, the feature extraction comprises: extracting a wavelet packet energy spectrum from the microseismic data; calculating a coefficient of variation from the strain data; and calculating a Lyapunov exponent from the displacement data to represent the chaotic characteristics of displacement.

[0011] According to a preferred embodiment, the method further comprises: constructing a reinforcement learning model based on DQN to establish a charge amount-damage value mapping relationship; and performing intelligent optimization of charge parameters using the reinforcement learning model. The specific way of performing intelligent optimization of charge parameters using the reinforcement learning model comprises: inputting the rock mass strength, joint density, and historical damage index into the reinforcement learning model, and outputting the adjustment range of the charge amount of the slotting hole or the peripheral hole from the reinforcement learning model.

[0012] According to a preferred embodiment, in the case of triggering an early warning, the reserved rock mass is repaired, and the repair effect is monitored according to the data collected by each sensing device. The multi-modal deep learning model is iteratively updated: the weight of the Transformer network is updated once every time the number of blasts reaches a preset value.

[0013] According to a preferred embodiment, a monitoring report is generated according to the monitoring data. The monitoring includes: the three-dimensional damage hotspot probability map and the damage index change curve, and optimization suggestions.

[0014] In a second aspect, the present application also provides a monitoring system for accumulated damage of reserved rock mass after cross-section tunnel blasting, comprising a sensing network and a data processing unit. The sensing network is configured to collect rock mass parameter data and transmit the rock mass parameter data to the data processing unit. Preferably, the rock mass parameter data at least includes microseismic data and strain data. The data processing unit is configured to process the data collected by each sensing device, including: noise reduction, filtering processing and data synchronization processing of the data collected by each sensing device; mapping the rock mass parameter data collected by each sensing device to the same three-dimensional coordinate system; calculating a damage index, and performing early warning according to the damage index and a preset early warning rule. Preferably, the calculation of the damage index comprises: processing the strain data by using a temperature-strain decoupling algorithm optimized by a least square method, and calculating strain energy density in real time; calculating the damage index according to the strain energy density, an initial value of the strain energy density, a microseismic event frequency and a geological adjustment coefficient. Preferably, the initial value of the strain energy density is calibrated by field acoustic testing; and the microseismic event frequency is obtained from the microseismic data.

[0015] Compared with the prior art, the present application has the following beneficial effects: The present application collects microseismic data and strain data of the reserved rock mass, processes the data, calculates a damage index, and performs early warning according to the damage index and a preset early warning rule, thereby improving the accuracy of monitoring of blasting damage of surrounding rock, rapidly evaluating the damage condition of the surrounding rock, and enabling appropriate protective measures to be taken in time, thereby ensuring the stability of the tunnel structure. BRIEF DESCRIPTION OF DRAWINGS

[0016] Figure 1 FIG. 1 is a flowchart of a preferred embodiment of the method for monitoring accumulated damage of reserved rock mass after cross-section tunnel blasting according to the present application.

[0017] Figure 2 FIG. 2 is a flowchart of the method for optimizing tunnel blasting according to the present application. DETAILED DESCRIPTION

[0018] The present application will be further described in detail below with reference to specific embodiments. However, it should not be understood that the scope of the above-mentioned subject matter of the present application is limited to the following embodiments, and any technology implemented based on the content of the present application falls within the scope of the present application.

[0019] In the description of specific embodiments of the present application, the terms of orientation or positional relationship such as "upper", "lower", "left", "right", "center", "inner", "outer" and the like are expressed based on the orientation or positional relationship shown in the drawings or the orientation or positional relationship in which the product / device / apparatus of the present application is usually used, unless otherwise specified. These terms of orientation or positional relationship are only for the convenience of describing the present application or simplifying the description in specific embodiments, for the purpose of facilitating the understanding of the scheme by the skilled person, and therefore cannot be understood as indicating or implying that a specific device / component / element must have a specific orientation or be constructed and operated in a specific positional relationship, and therefore cannot be understood as limiting the present application.

[0020] In addition, the terms "horizontal", "vertical", "overhanging", "parallel" and the like do not mean that the corresponding device / component / element must be absolutely horizontal or vertical or overhanging or parallel, but can be slightly inclined or deviated. For example, "horizontal" only means that its direction is more horizontal relative to "vertical", and does not mean that the structure must be completely horizontal, but can be slightly inclined. Alternatively, it can be simplified to mean that the corresponding device / component / element is arranged in the direction of "horizontal", "vertical", "overhanging", "parallel" and the like, and can have an error / deviation of ±10% with respect to the corresponding direction, more preferably an error / deviation of ±8% or less, more preferably an error / deviation of ±6% or less, more preferably an error / deviation of ±5% or less, and more preferably an error / deviation of ±4% or less. As long as the corresponding device / component / element is within the error / deviation range, it can still achieve its role in the scheme of the present application.

[0021] In addition, the terms "first", "second", "third" and the like in the description of the present application are only used to distinguish the same or similar components for description, and should not be understood as emphasizing or implying the relative importance of the specific components.

[0022] In addition, in the description of the embodiments of the present application, "several", "a plurality of", "several" represent at least 2. It can be 2, 3, 4, 5, 6, 7, 8, 9, etc. in any case, and even more than 9.

[0023] In addition, in the description of the technical scheme of the present application, unless otherwise specified / limited / limited, the terms "arrangement", "installation", "connection", "connection", "provided with", "laid", "arrangement" should be understood broadly, for example, it can be fixedly connected, or it can be detachably connected, or it can be integrally connected, such as welding, riveting, bolting, screwing and other commonly used connection means in the art. The connection can be mechanical connection, electrical connection or communication connection; it can be directly connected or indirectly connected through an intermediate medium; it can be the communication between two elements.

[0024] Example 1 The embodiment provides a method for monitoring accumulated damage of reserved rock mass after section tunnel blasting.

[0025] Referring to Figure 1 Preferably, the method for monitoring accumulated damage of reserved rock mass after section tunnel blasting comprises the following steps: S1, laying a sensing device for collecting rock mass parameter data; S2, preprocessing data collected by each sensing device, and mapping rock mass parameter data collected by each sensing device to the same three-dimensional coordinate system; S3, calculating a damage index, and performing early warning according to the damage index and a preset early warning rule.

[0026] Preferably, the rock mass parameter data at least includes microseismic data and strain data.

[0027] Preferably, the preprocessing includes noise reduction, filtering processing and data synchronization processing.

[0028] Preferably, the calculation of the damage index comprises: The strain data is processed by using a temperature-strain decoupling algorithm optimized by a least square method to calculate strain energy density in real time; The damage index is calculated according to the strain energy density, an initial value of the strain energy density, a microseismic event frequency and a geological adjustment coefficient. Preferably, the initial value of the strain energy density is calibrated by field acoustic testing; and the microseismic event frequency is obtained from the microseismic data.

[0029] The embodiment collects microseismic data and strain data of the reserved rock mass, processes the data, calculates a damage index, and performs early warning according to the damage index and a preset early warning rule, thereby improving the accuracy of monitoring of blasting damage of surrounding rock, rapidly evaluating the damage condition of the surrounding rock, and thus enabling appropriate protective measures to be taken in time to ensure the stability of the tunnel structure.

[0030] Embodiment 2 The embodiment is a further improvement of the embodiment 1, and repeated contents will not be described herein.

[0031] Preferably, the calculation formula of the damage index is:

[0032] In the formula, DI is the calculated damage index; is the measured strain energy density; is an initial value of the strain energy density; is a microseismic event frequency per unit time; is a geological adjustment coefficient. Preferably, the value of the geological adjustment coefficient can be selected according to the grade of the surrounding rock, for example, 0.6 / 0.4 for V-grade surrounding rock and 0.4 / 0.6 for III-grade surrounding rock.

[0033] Preferably, the sensing device comprises at least several microseismic sensors. The several microseismic sensors are arranged along the tunnel axis to collect microseismic data of the blasting influence zone and the potential damage propagation zone. Preferably, the several microseismic sensors are arranged along the tunnel axis with intervals. Preferably, the data collected by the microseismic sensors are transmitted to the server or the data processing unit through wired or wireless means.

[0034] Preferably, the sensing device further comprises several strain sensors. The several strain sensors are arranged at key positions of the reserved rock mass to collect strain data of the reserved rock mass. Preferably, the key positions at least include: the crown, the haunch, and the sidewall.

[0035] Preferably, S2 comprises: applying wavelet threshold denoising to the microseismic data and applying moving average filtering to the strain data; and using the dynamic time warping algorithm for data synchronization.

[0036] Preferably, the early warning rule in S3 comprises: comparing the damage index with a preset threshold to determine the safety state of the tunnel. The preset threshold comprises a first threshold and a second threshold greater than the first threshold. In the case that the damage index is less than the first threshold, the safety state of the tunnel is safe; in the case that the damage index is greater than or equal to the first threshold and less than the second threshold, the safety state of the tunnel is early warning; and in the case that the damage index is greater than or equal to the second threshold, the safety state of the tunnel is dangerous. Preferably, the first threshold and the second threshold are set according to the surrounding rock grade and the hydrogeological condition.

[0037] Preferably, the sensing device further comprises several displacement sensors for monitoring displacement data of the reserved rock mass. The microseismic data, the strain data, and the displacement data are mapped to the same three-dimensional coordinate system in combination with the BIM model.

[0038] Preferably, the microseismic sensors, the strain sensors, and the displacement sensors can all realize the unification of the time base of the multiple sensors through the Beidou positioning system.

[0039] The microseismic data, the strain data, and the displacement data are subjected to feature extraction, and the extracted features are input into a multi-modal deep learning model constructed based on a Transformer network to obtain a three-dimensional damage hotspot probability map. Preferably, the feature extraction comprises: extracting a wavelet packet energy spectrum from the microseismic data; calculating a coefficient of variation from the strain data; and calculating a Lyapunov exponent from the displacement data to represent the chaotic characteristics of the displacement. Preferably, the calculation formula of the coefficient of variation is:

[0040] wherein, Coefficient of Variation (CV): A dimensionless quantity that represents the relative variability of a dataset. The CV value is calculated by dividing the standard deviation by the mean and multiplying by 100. A higher CV value indicates a higher degree of relative dispersion in the data. Standard Deviation (SD): A measure of the dispersion of a dataset, representing the average distance between each data point and the mean value. A larger SD indicates a greater degree of variability in the data. Mean (M): The arithmetic average of all values in a dataset, representing the central tendency of the data.

[0041] Preferably, it further comprises: constructing a DQN-based reinforcement learning model to establish a charge quantity-damage value mapping relationship; and using the reinforcement learning model for intelligent optimization of charge parameters. The specific way of using the reinforcement learning model for intelligent optimization of charge parameters includes: inputting the rock mass strength, joint density, and historical damage index into the reinforcement learning model, and the reinforcement learning model outputs the adjustment amplitude of the slotting hole or peripheral hole charge quantity.

[0042] Preferably, in the case of triggering a warning, the reserved rock mass is repaired, and the repair effect is monitored according to the data collected by each sensing device. Preferably, the way of repairing the reserved rock mass includes grouting reinforcement of the rock mass. Preferably, after grouting reinforcement of the rock mass, the strain sensor is used to monitor the strain change of the grout solidification, and the displacement sensor is used to detect the surface closure displacement of the rock mass.

[0043] Preferably, the embodiment continuously iteratively updates the multi-modal deep learning model during the monitoring process: the Transformer network weight is updated once every time the number of blasts reaches a preset value.

[0044] Preferably, a monitoring report is generated according to the monitoring data. The monitoring includes: a three-dimensional damage hotspot probability map and a damage index change curve, as well as optimization suggestions. Preferably, the optimization suggestions include optimization of charge parameters, specifically including: blast hole spacing, row spacing, single hole charge quantity, single initiation explosive quantity, and delay time.

[0045] The embodiment can accurately evaluate the accumulated damage of the reserved rock mass after blasting in real time, significantly improving the timeliness and accuracy of surrounding rock stability monitoring. Based on the damage evaluation results, construction personnel can quickly make decisions and take targeted support measures in a timely manner, effectively ensuring the safety and structural stability of the tunnel project, and significantly reducing the risk of tunnel collapse caused by accumulated blasting damage, providing reliable technical support for the safe construction of the tunnel section.

[0046] Embodiment 3 Preferably, the embodiment is a specific application of Embodiment 1 and Embodiment 2, and the embodiment provides a method for monitoring accumulated damage of reserved rock mass after tunnel blasting.

[0047] Referring to Figure 2 , preferably, the flow of the method for monitoring accumulated damage of reserved rock mass provided by the embodiment comprises: Step one, microseismic sensor array layout and parameter configuration.

[0048] Step two, fiber Bragg grating strain sensor pre-embedding and compensation calibration.

[0049] Step three, visual displacement monitoring system installation and dynamic calibration.

[0050] Step four, multi-source data space-time alignment and preprocessing.

[0051] Step five, multi-modal damage feature extraction and fusion modeling.

[0052] Step six, dynamic calculation and verification of damage index (DI).

[0053] Step seven, dynamic threshold grading and early warning rule setting.

[0054] Step eight, intelligent optimization and feedback control of charging parameters.

[0055] Step nine, real-time monitoring and iterative updating of damage repair effect.

[0056] Step ten, comprehensive performance evaluation and report generation.

[0057] Preferably, in step one, the microseismic sensor uses a high-sensitivity microseismic sensor, and the specific model can be: SOS microseismic monitor. Preferably, the microseismic sensors are arranged in an array along the tunnel axis at intervals of 5-10 m. Preferably, the data acquisition range of the microseismic sensor covers the blasting affected area and the potential damage propagation area. Preferably, the range covering the blasting affected area and the potential damage propagation area can be set according to historical experience and existing research results.

[0058] Preferably, the blasting affected area can be the surrounding rock within 150 m of the working face. Based on the tunnel engineering geological exploration report and the surrounding rock exploration of the working face within 150 m, the basic parameters of the surrounding rock are determined. Preferably, the basic parameters of the surrounding rock include rock strength, rock mass integrity, and rock mass structure type, and the basic quality index (BQ) of the surrounding rock is obtained. The basic quality index (BQ) of the surrounding rock is an important index for evaluating the quality of the surrounding rock in underground engineering. Preferably, the sampling frequency of the microseismic sensor is ≥1 kHz, and the dynamic range is ≥120 dB. Preferably, the data collected by the microseismic sensor is uploaded to the cloud server in real time through the LoRa wireless transmission module.

[0059] Preferably, in step two, the strain data of the reserved rock mass is collected by using the fiber Bragg grating strain sensor. Preferably, the distributed fiber Bragg grating sensors are embedded at the key positions of the reserved rock mass, such as the arch crown, arch shoulder and side wall, and are arranged in a spiral winding manner (with a spacing of 10 cm), and the spatial resolution reaches 1 cm. Preferably, the data collected by the fiber Bragg grating strain sensor is also uploaded to the cloud server in real time through the LoRa wireless transmission module. Preferably, when the cloud server receives the data collected by the fiber Bragg grating strain sensor, a temperature-strain decoupling algorithm optimized by the least square method is used to eliminate the interference of high humidity (> 85%) environment, compensate and calibrate the fiber Bragg grating strain sensor, and ensure the strain measurement accuracy of the fiber Bragg grating strain sensor ± 5με.

[0060] Preferably, in step three, an XDYG-EC type visual displacement measuring instrument (resolution 2448×2048 pixels, frame rate ≥ 30 Hz) is used as a visual sensor to build a visual displacement monitoring system, and an infrared reflective target (spacing 2 m) is installed. Preferably, the data collected by the visual sensor is also uploaded to the cloud server in real time. Preferably, the cloud server eliminates the image jitter caused by blasting vibration by using an improved ORB-SLAM algorithm, so that the displacement measurement accuracy reaches ± 0.1 mm.

[0061] Preferably, in step four, the BIM model is used to map the microseismic, strain and displacement data to the same three-dimensional coordinate system (XYZ accuracy ± 2 cm); the dynamic time warping algorithm (DTW) is used to solve the data synchronization problem of different sampling rate devices (such as microseismic data 1000 Hz, strain data 10 Hz); the wavelet threshold denoising is applied to the microseismic data, and the moving average filtering is applied to the fiber data, etc.

[0062] Preferably, in step five, the wavelet packet energy spectrum (frequency band 0.1-100 Hz) is extracted from the microseismic data; the coefficient of variation is calculated according to the strain data; the Lyapunov index is calculated according to the displacement data, representing the displacement chaotic characteristics. Preferably, a multi-modal deep learning model is constructed based on the Transformer network, the aforementioned extracted features are input into the multi-modal deep learning model, and a three-dimensional damage hotspot probability map is obtained.

[0063] Preferably, in step six, the damage index is calculated according to the strain energy density, the initial value of the strain energy density, the microseismic event frequency and the geological adjustment coefficient. Preferably, the initial value of the strain energy density is calibrated by the field acoustic test; the microseismic event frequency is obtained from the microseismic data. Preferably, the calculation formula of the damage index is:

[0064] In the formula, DI is the calculated damage index; is the measured strain energy density; is the initial value of the strain energy density; is the frequency of microseismic events per unit time; is the geological adjustment coefficient. Preferably, the value of the geological adjustment coefficient can be selected according to the surrounding rock grade, for example, 0.6 / 0.4 for V-grade surrounding rock and 0.4 / 0.6 for III-grade. Preferably, the cloud server also calculates the strain energy density change rate in real time according to the strain data and verifies the damage trend in combination with ANSYS numerical simulation.

[0065] Preferably, in step seven, the threshold values in the alarm rule are dynamically set according to the surrounding rock grade and hydrogeological conditions. For example, the first threshold value is set to 0.3 and the second threshold value is set to 0.6. In the case of a damage index less than 0.3, the tunnel safety state is safe; in the case of a damage index greater than or equal to 0.3 and less than 0.6, the tunnel safety state is pre-warning; and in the case of a damage index greater than or equal to 0.6, the tunnel safety state is dangerous.

[0066] Preferably, in step eight, a charge amount-damage value mapping relationship is established based on a DQN reinforcement learning model. The rock mass strength, joint density, and historical damage index are input into the reinforcement learning model, and the reinforcement learning model outputs the adjustment amplitude of the charge amount of the slotting hole or the peripheral hole, thereby realizing intelligent optimization and feedback control of the charge parameters.

[0067] Preferably, in step nine, the strain change (threshold value ± 20με) of the slurry solidification is monitored by the optical fiber sensor, and the surface closure displacement (accuracy 0.2 mm) is detected by the visual system, thereby realizing real-time monitoring of the damage repair effect after grouting reinforcement. Preferably, the multi-modal deep learning model is iteratively updated, and the specific way is that the Transformer network weight is updated once every 10 blasting cycles, thereby improving the geological adaptability of the multi-modal deep learning model, and finally the fault zone recognition accuracy is improved from 82% to 89%.

[0068] Preferably, in step ten, the monitoring method is comprehensively evaluated and a report is generated according to the overbreak, support cost, and pre-warning response time. Preferably, the overbreak can be determined by laser scanning, and the support cost can be represented by the amount of concrete. Preferably, the generated report includes a damage distribution thermal map, a DI change curve, and optimization suggestions.

[0069] Preferably, the embodiment realizes real-time and accurate evaluation of accumulated damage of reserved rock after blasting, significantly improving the timeliness and accuracy of surrounding rock stability monitoring. According to the obtained damage evaluation results, the construction personnel can quickly make decisions and take targeted support measures in a timely manner, effectively ensuring the safety and structural stability of the tunnel engineering, and can significantly reduce the risk of tunnel collapse caused by accumulated blasting damage, thereby providing reliable technical support for the safe construction of the cross-section tunnel.

[0070] Embodiment 4 The embodiment provides a cross-section tunnel blasting post-reserved rock mass accumulated damage monitoring system. Preferably, the cross-section tunnel blasting post-reserved rock mass accumulated damage monitoring system provided by the embodiment performs the steps of the cross-section tunnel blasting post-reserved rock mass accumulated damage monitoring method in the use.

[0071] Preferably, the cross-section tunnel blasting post-reserved rock mass accumulated damage monitoring system provided by the embodiment comprises a sensing network and a data processing unit. The sensing network is used to collect rock mass parameter data and transmit the rock mass parameter data to the data processing unit. Preferably, the rock mass parameter data at least includes microseismic data and strain data. The data processing unit is used to process the data collected by each sensing device, including: noise reduction, filtering processing and data synchronization processing of the data collected by each sensing device; mapping the rock mass parameter data collected by each sensing device to the same three-dimensional coordinate system; calculating a damage index, and performing early warning according to the damage index and a preset early warning rule. Preferably, the calculation of the damage index comprises: processing the strain data by using a temperature-strain decoupling algorithm optimized by a least square method, and calculating strain energy density in real time; and calculating the damage index according to the strain energy density, an initial value of the strain energy density, a microseismic event frequency and a geological adjustment coefficient. Preferably, the initial value of the strain energy density is calibrated by field acoustic testing; and the microseismic event frequency is obtained from the microseismic data.

[0072] The above merely describes the preferred embodiments of the present application and is not used to limit the present application, and any modification, equivalent replacement and improvement made within the spirit and principle of the present application should be included in the protection scope of the present application.

Claims

1. A method for monitoring accumulated damage of a reserved rock volume after cross-section tunnel blasting, characterized in that, The method comprises the following steps: S1, laying out a sensing device for collecting rock mass parameter data; wherein the rock mass parameter data at least includes microseismic data and strain data; S2, pre-processing the data collected by each sensing device, and mapping the rock mass parameter data collected by each sensing device to the same three-dimensional coordinate system; wherein the pre-processing includes noise reduction, filtering processing and data synchronization processing; S3, calculating a damage index, and performing early warning according to the damage index and a preset early warning rule; The calculation of the damage index comprises: processing the strain data by using a temperature-strain decoupling algorithm optimized by a least square method to calculate strain energy density in real time; calculating the damage index according to the strain energy density, an initial value of the strain energy density, a microseismic event frequency and a geological adjustment coefficient; The initial value of the strain energy density is calibrated by field acoustic testing; and the microseismic event frequency is obtained from the microseismic data.

2. The method according to claim 1, characterized in that, The sensing device at least includes a plurality of microseismic sensors; The plurality of microseismic sensors are laid out along a tunnel axis to collect microseismic data of a blasting influence area and a potential damage propagation area.

3. The method according to claim 2, characterized in that, The sensing device further includes a plurality of strain sensors; The plurality of strain sensors are laid out at key positions of a reserved rock mass to collect strain data of the reserved rock mass; The key positions at least include a vault, a spandrel and a side wall.

4. The method according to claim 3, wherein S2 The method comprises: applying wavelet threshold noise reduction to the microseismic data, and applying sliding average filtering to the strain data; performing data synchronization by using a dynamic time warping algorithm.

5. The method according to claim 4, characterized in that, The early warning rule in S3 comprises: comparing the damage index with a preset threshold value to determine a tunnel safety state; the preset threshold value includes a first threshold value and a second threshold value greater than the first threshold value; in a case where the damage index is less than the first threshold value, the tunnel safety state is safe; in a case where the damage index is greater than or equal to the first threshold value and less than the second threshold value, the tunnel safety state is early warning; in a case where the damage index is greater than or equal to the second threshold value, the tunnel safety state is dangerous; The first threshold value and the second threshold value are set according to the surrounding rock grade and hydrogeological conditions.

6. The method according to claim 5, characterized in that, characterized in that, The sensing device further includes a plurality of displacement sensors for monitoring displacement data of the reserved rock mass; The microseismic data, the strain data and the displacement data are mapped to the same three-dimensional coordinate system in combination with a BIM model; characteristic extraction is performed on the microseismic data, the strain data and the displacement data, the extracted characteristics are input into a multi-modal deep learning model based on a Transformer network to obtain a three-dimensional damage hotspot probability map; The characteristic extraction comprises: extracting a wavelet packet energy spectrum from the microseismic data; calculating a coefficient of variation from the strain data; calculating a Lyapunov index from the displacement data to represent a displacement chaotic characteristic.

7. The method according to claim 6, characterized in that, The method further comprises: constructing a reinforcement learning model based on DQN to establish a charge amount-damage value mapping relationship; performing intelligent optimization of charge parameters by using the reinforcement learning model, including: Inputting the rock mass strength, joint density, and historical damage index into the reinforcement learning model, the reinforcement learning model outputs the adjustment range of the charge amount of the slot hole or the peripheral hole.

8. The method according to claim 7, wherein, in the case of triggering a warning, the reserved rock mass is repaired, and the repair effect is monitored according to the data collected by the sensing devices. The multi-modal deep learning model is iteratively updated: the Transformer network weight is updated once every preset number of times of blasting. A monitoring report is generated according to the monitoring data.

9. The method according to claim 8, characterized in that, The monitoring includes the three-dimensional damage hotspot probability map and the damage index change curve, and optimization suggestions. The sensing network and the data processing unit are included.

10. A system for monitoring the accumulation of damage in a pre- reserved rock volume after cross-sectional tunnel blasting, characterized by, The sensing network is configured to collect rock mass parameter data and transmit the rock mass parameter data to the data processing unit, wherein the rock mass parameter data at least includes microseismic data and strain data. The data processing unit is configured to process the data collected by the sensing devices, including: The data collected by the sensing devices is subjected to noise reduction, filtering processing, and data synchronization processing. The rock mass parameter data collected by the sensing devices is mapped to the same three-dimensional coordinate system. The damage index is calculated, and a warning is given according to the damage index and a preset warning rule. The calculation of the damage index includes: The strain data is processed by using a temperature-strain decoupling algorithm optimized by a least square method to calculate the strain energy density in real time. The damage index is calculated according to the strain energy density, an initial value of the strain energy density, a microseismic event frequency, and a geological adjustment coefficient. The initial value of the strain energy density is calibrated by field acoustic testing, and the microseismic event frequency is obtained from the microseismic data. ​

Images