Micro-vibration monitoring and early warning method, device and medium for tbm crossing existing cavern group

Abstract

This application discloses a method, device, and medium for micro-vibration monitoring and early warning when a TBM traverses an existing cavern complex, relating to the field of tunnel and underground engineering construction safety monitoring. First, a multi-source sensing network is deployed on the structural surface of the existing cavern, with filtering and vibration attenuation benchmark database calculations performed locally at the edge. Then, the vibration attenuation model is dynamically updated using real-time TBM tunneling vibration source parameters, overcoming the limitations of traditional fixed models and thresholds. Next, the vibration attenuation benchmark database is compared with the dynamic vibration attenuation model to generate graded early warning commands, distinguishing different vibration risk levels and achieving refined risk assessment. Finally, simultaneous early warnings are provided through multiple terminals—cave-side audio-visual systems, portable terminals, and the TBM cab—and hardware interlocks are used to directly control the TBM to slow down or stop, constructing a closed-loop management system of monitoring, analysis, early warning, and equipment response. This suppresses vibration disturbances from the tunneling source and proactively protects the structural safety of the existing cavern.

Description

Technical Field

[0001] This application relates to the field of tunnel and underground engineering construction safety monitoring technology, specifically to a micro-vibration monitoring and early warning method, device and medium for TBMs traversing existing cavern groups. Background Technology

[0002] In the construction of underground infrastructure for large-scale water conservancy and hydropower projects, large-scale underground cavern complexes such as main powerhouses, main transformer rooms, and tailgate chambers are often constructed. These cavern structures are massive in size and have high service levels, serving as core and critical structures of the project. Their structural stability and operational safety directly determine the normal operation of the entire water conservancy hub. During the subsequent construction phases of drainage corridors, traffic tunnels, and other tunnels, due to route planning constraints, tunnel boring machines (TBMs) often need to pass close to existing cavern complexes using methods such as orthogonal underpasses, lateral detours, and small-angle oblique crossings.

[0003] During the rock-breaking and movement of the TBM cutterhead, continuous mechanical vibrations are generated. These vibrations propagate through the surrounding rock as stress waves, causing continuous disturbances to the high sidewalls, large-span arches, and supporting structures of existing caverns. These vibrations can cause minor issues such as surface cracking and support layer detachment, and in severe cases, can induce support system failure and surrounding rock instability, posing significant safety hazards to water conservancy projects.

[0004] Currently, safety management methods for this construction scenario are significantly inadequate. Traditional management methods mostly rely on construction experience to manually adjust operating parameters such as TBM tunneling speed, cutterhead rotation speed, and tunneling thrust, failing to independently, in real-time, and quantitatively monitor the actual vibration response of existing tunnels. This results in outdated management measures lacking data support. Furthermore, traditional safety inspections depend on manual on-site checks, which are not only inefficient and subjective in their judgments but also struggle to capture instantaneous dynamic vibration risks, making early risk prediction impossible. The few vibration monitoring solutions available often use single sensors for single-parameter detection, failing to achieve multi-source sensing and fusion monitoring, leading to problems such as monitoring blind spots, incomplete data, and low assessment accuracy. Moreover, the monitoring systems only have simple alarm functions and cannot be linked with the TBM main control terminal and on-site operator terminals, making it difficult to construct a complete safety closed loop of "monitoring-analysis-early warning-response."

[0005] In summary, traditional technologies are insufficient to meet the structural safety management requirements when TBMs pass close to existing hydraulic cavern groups. The industry urgently needs a monitoring and early warning method that is designed for this specific working condition, can achieve high-frequency, multi-parameter intelligent micro-vibration monitoring, and complete closed-loop control throughout the entire process, so as to proactively identify vibration risks and ensure the structural safety of existing caverns. Summary of the Invention

[0006] The purpose of this application is to provide a method, device, and medium for monitoring and early warning of micro-vibrations when a TBM passes through an existing cavern group, in order to solve the problems of low accuracy and high safety risks in traditional TBM micro-vibration monitoring.

[0007] To achieve the above objectives, the first aspect of this application provides a method for monitoring and early warning of micro-vibrations when a TBM traverses an existing cavern group, comprising: A multi-source signal sensing network is deployed on the structural surface of the existing cavern. Micro-vibration signals are collected synchronously at the edge and pre-processed with noise filtering. Real-time response data of vibration velocity peak value (PPV) at each monitoring point is calculated. A vibration attenuation benchmark database is built and updated based on the PPV response data of the monitoring points. The core tunneling vibration source parameters of the TBM host are collected in real time, a pre-built vibration attenuation benchmark database is called, and the vibration attenuation benchmark database is dynamically calibrated in combination with the collected tunneling vibration source parameters to generate a dynamic vibration attenuation model and a dynamic safety threshold system that fits the real-time tunneling conditions. The vibration attenuation benchmark database is compared with the dynamic vibration attenuation model, and the on-site construction risk level is accurately matched according to the preset dynamic risk classification rules to generate corresponding graded early warning instructions. Based on the aforementioned graded early warning instructions, early warning information is simultaneously pushed to existing cavern audio-visual devices, TBM cab early warning panels, and portable terminals to achieve multi-terminal synchronous early warning publicity. When the graded early warning level reaches the preset high-risk safety threshold, the TBM main control system is directly connected through the TBM's dedicated hard-connection control interface to actively issue forced speed reduction or shutdown instructions.

[0008] The second aspect of this application provides a micro-vibration monitoring and early warning device for TBMs traversing existing cavern groups, comprising: The acquisition module is used to deploy a multi-source signal sensing network on the structural surface of the existing cavern, synchronously acquire micro-vibration signals at the edge and perform noise filtering preprocessing, calculate the real-time response data of vibration velocity peak value (PPV) at each monitoring point, and build and update the vibration attenuation benchmark database based on the PPV response data of the monitoring points. The modeling module is used to collect core tunneling vibration source parameters in real time during the tunneling process of the TBM host, call the pre-built vibration attenuation benchmark database, and dynamically calibrate the vibration attenuation benchmark database in combination with the real-time collected tunneling vibration source parameters to generate a dynamic vibration attenuation model and a dynamic safety threshold system that fits the real-time tunneling conditions. The early warning module is used to compare the vibration attenuation benchmark database with the dynamic vibration attenuation model, accurately match the on-site construction risk level according to the preset dynamic risk classification rules, and generate corresponding graded early warning instructions. The control module is used to push warning information to the existing cavern audio-visual devices, TBM cab warning panel and portable terminal in accordance with the graded warning instructions, so as to realize the synchronous warning publicity of multiple terminals. When the graded warning level reaches the preset high-risk safety threshold, it directly connects to the TBM main control system through the TBM dedicated hard control interface and actively issues a forced speed reduction or shutdown command.

[0009] A third aspect of this application provides a computer-readable storage medium storing a program that can be loaded and executed by a processor to perform the aforementioned micro-vibration monitoring and early warning method for TBMs traversing existing cavern groups.

[0010] The beneficial effects of this application are: This application first deploys a multi-source sensing network on the structural surface of the existing cavern, and performs filtering and vibration attenuation benchmark database calculation on-site at the edge. This eliminates monitoring blind spots, improves the integrity of vibration data acquisition, reduces the amount of raw data transmission, and lowers transmission delay, enabling high-frequency, real-time, and quantitative detection of micro-vibration signals. Then, the vibration attenuation model is dynamically updated using real-time tunneling vibration source parameters of the TBM, overcoming the limitations of traditional fixed models and thresholds. This allows the vibration propagation law to adapt to real-time tunneling conditions and surrounding rock conditions, significantly improving the accuracy of vibration assessment and reducing the probability of misjudgments and omissions. Next, the vibration attenuation benchmark database is compared with the dynamic vibration attenuation model to generate graded early warning commands, which can distinguish different vibration risk levels, enabling refined risk assessment and providing a basis for differentiated treatment. Finally, multi-terminal synchronous early warning is implemented through cavern audio-visual systems, portable terminals, and the TBM cab. Hardware interlocks directly control the TBM to slow down or stop, constructing a closed-loop management system of monitoring, analysis, early warning, and equipment response. This suppresses vibration disturbances from the source of tunneling and proactively protects the structural safety of the existing cavern.

[0011] Other features and advantages of this application will be described in detail in the following detailed description section. Attached Figure Description

[0012] Figure 1 This is a flowchart illustrating a micro-vibration monitoring and early warning method for a TBM traversing an existing cavern group, as provided in an embodiment of this application. Figure 2 This is a general architecture block diagram provided in a specific embodiment of this application; Figure 3 This is a schematic diagram of the cross-sectional layout of a multi-source signal sensing network in an existing cavern group, provided in a specific embodiment of this application. Figure 4 This is a diagram showing a non-equidistant grid arrangement of a MEMS accelerometer in a longitudinal section of an existing cavern, provided in a specific embodiment of this application. Figure 5This is a schematic diagram of the display interface of a remote early warning panel provided in a specific embodiment of this application; Figure 6 This is a flowchart of an early warning logic and linkage control provided in a specific embodiment of this application; Figure 7 This is a schematic diagram of a micro-vibration monitoring and early warning device for a TBM traversing an existing cavern group, provided in an embodiment of this application. Detailed Implementation

[0013] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0014] In the description of this application, it should be understood that the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the stated features. In the description of this application, "a plurality of" means two or more, unless otherwise explicitly specified. Details are set forth in the following description for illustrative purposes. It should be understood that those skilled in the art will recognize that this application can be implemented without using these specific details. In other instances, well-known structures and processes will not be described in detail to avoid unnecessarily obscuring the description of this application. Therefore, this application is not intended to be limited to the embodiments shown, but rather to be consistent with the broadest scope of the principles and features disclosed herein.

[0015] Figure 1 This is a flowchart illustrating a micro-vibration monitoring and early warning method for a TBM traversing an existing cavern group, as provided in an embodiment of this application. Figure 1 As shown, the method may include steps 101-104, which will be described in detail below.

[0016] Step 101: Deploy a multi-source signal sensing network on the structural surface of the existing cavern, synchronously collect micro-vibration signals at the edge side and perform noise filtering preprocessing, calculate the real-time response data of vibration velocity peak value (PPV) at each monitoring point, and build and update the vibration attenuation benchmark database based on the PPV response data of the monitoring points.

[0017] This step involves preprocessing multi-source signals acquired at the edge and constructing a vibration benchmark database. The edge side refers to the edge computing station deployed at the monitoring site, which integrates fiber optic demodulation units, multi-channel synchronous acquisition instruments, and edge computing modules. Unlike the remote cloud, it can complete computation and processing locally at the data acquisition end.

[0018] First, a multi-source signal sensing network is deployed on the structural surface of the completed existing cavern. In this embodiment, the multi-source signal sensing network may include a distributed optical fiber subsystem and a key array-type micro-electro-mechanical system (MEMS) accelerometer subsystem. It adopts a dual-sensor combination mode of "general measurement across the entire area + precise measurement of key areas" to complete the acquisition of vibration signals from the existing cavern, forming a dual-sensor collaborative monitoring architecture with full coverage and enhanced key areas.

[0019] The distributed optical fiber subsystem employs distributed acoustic or vibration sensing optical fibers, laid in a folded-back pattern along the longitudinal and circumferential directions within or on the surface of the existing cavern support layer. The installation locations are inside or on the outer surface of the cavern support layer, closely adhering to the structural body. By demodulating the backscattered light signal of the probe pulses, vibration data at various locations along the line are calculated, enabling uninterrupted, comprehensive, and seamless monitoring of the overall vibration field of the existing cavern.

[0020] The key lattice-type MEMS accelerometer subsystem consists of triaxial MEMS accelerometers deployed in key stress-bearing and vulnerable risk areas of the cavern arch foot, arch waist, arch crown, and rock pillars. A differentiated gradient deployment strategy is adopted, with denser deployment in high-risk core areas such as the TBM crossing center projection, and sparse deployment in other low-risk areas using a non-equidistant grid.

[0021] Distributed optical fibers can be continuously deployed over long distances and large areas within the cavern, achieving full coverage of the overall vibration field and completely eliminating the monitoring blind spots present in traditional point sensors. This allows for a complete understanding of the overall vibration distribution characteristics of the cavern. Triaxial MEMS accelerometers are highly sensitive to local micro-vibration signals. By densely deploying sensors in critical stress areas and high-risk zones, they can accurately capture small disturbances caused by local abnormal vibrations and stress concentrations, precisely identifying the risk of local structural damage. A gradient-differentiated deployment approach, with denser sensors in high-risk areas and sparser sensors in low-risk areas, reduces the number of sensors, lowers equipment costs, and reduces construction and maintenance workload while maintaining monitoring capabilities. Optical fibers excel at continuous monitoring across the entire area, while MEMS accelerometers excel at high-precision detection of local points. Working together, they balance monitoring range and single-point accuracy, with data mutually corroborating each other to improve the reliability and validity of the entire sensing network. Both types of sensors are adaptable to the humid, dusty, and electromagnetically interference-prone environments of underground caverns, enabling long-term stable and uninterrupted operation to meet the monitoring needs of long-term TBM (Tunnel Boring Machine) tunneling construction.

[0022] Then, micro-vibration signals are synchronously acquired at the edge and preprocessed with noise probability waves. Noise filtering preprocessing involves using a bandpass filter to remove low-frequency background noise and high-frequency electromagnetic interference, combined with baseline correction to eliminate signal drift and purify the original micro-vibration signals. This method differs from the traditional approach of uploading massive amounts of original micro-vibration data to the cloud for computation. It effectively solves the data transmission bandwidth bottleneck and computational delay problems in long-distance tunnel construction scenarios, and calculates the real-time peak particle velocity (PPV) response data of each monitoring point. Based on the PPV response data of the monitoring points, a vibration attenuation benchmark database is built and updated. The vibration attenuation benchmark database is a static basic database established through numerical inversion based on existing tunnel rock mass mechanical parameters and on-site measured PPV data. It internally stores the inherent mapping relationship between the vibration source distance, rock mass parameters, and PPV, characterizing the original vibration propagation law of the surrounding rock in the area, and serves as the foundation for subsequent model calibration.

[0023] This step abandons the centralized cloud-based computing model for raw data, processing data locally at the edge to solve the problems of insufficient bandwidth and high latency in cloud computing during long-distance tunnel construction, ensuring real-time monitoring. Pre-processing filtering removes interference from the site environment and electromagnetic clutter, ensuring the accuracy and validity of PPV data. The benchmark library is dynamically updated based on field measurement data, which better reflects the actual vibration propagation characteristics of the surrounding rock in the tunnel compared to purely theoretical modeling. A multi-source heterogeneous sensing network combined with a sparse-density deployment strategy balances overall coverage with high-precision monitoring of key areas, eliminating blind spots associated with traditional single-point monitoring.

[0024] Step 102: Collect core tunneling vibration source parameters in real time during the TBM main engine tunneling process, call the pre-built vibration attenuation benchmark database, and dynamically calibrate the vibration attenuation benchmark database in combination with the real-time collected tunneling vibration source parameters to generate a dynamic vibration attenuation model and dynamic safety threshold system that fits the real-time tunneling conditions.

[0025] This step relies on dynamic calibration of tunneling vibration source parameters to generate an adaptive vibration attenuation model. Tunneling vibration source parameters refer to the TBM's total thrust, cutterhead torque, penetration depth, and tunneling speed. These are core parameters characterizing the intensity of rock-breaking disturbance by the TBM and also serve as inputs for calculating equivalent vibration source energy. Tunneling vibration source parameters can be synchronously acquired from the TBM main unit, achieving data linkage between vibration source status and tunnel response.

[0026] Dynamic calibration is the process of converting tunneling parameters into equivalent source energy, combining measured PPV with the least squares method to invert site coefficients, attenuation exponents, and rock mass parameters in the benchmark database, and correcting the database parameters in real time. For example, an attenuation model can be constructed based on the Sadovsky formula, converting tunneling parameters into equivalent source energy through energy mapping relationships. Parameter inversion is completed using time window sampling and least squares fitting, with continuous iterative calibration of the benchmark database. The model adaptively changes with tunneling location and state. The generated dynamic vibration attenuation model is a working condition-adaptive model obtained after real-time TBM tunneling parameter calibration based on the vibration attenuation benchmark database, accurately reflecting the vibration propagation law of the surrounding rock under the current tunneling state. The dynamic safety threshold system is a graded safety threshold corrected based on real-time vibration propagation characteristics, distinct from fixed thresholds, and changes synchronously with tunneling conditions and surrounding rock state.

[0027] This step involves real-time acquisition of core vibration source parameters during the TBM's tunneling process, calling upon a pre-built vibration attenuation benchmark database, and dynamically calibrating the database using the real-time acquired vibration source parameters. This overcomes the technical shortcomings of traditional static attenuation models, such as poor adaptability and large prediction errors in complex and variable geological conditions, significantly reducing vibration prediction errors. The model is bound to the real-time tunneling status of the TBM, accurately matching various crossing conditions such as orthogonal underpasses, lateral detours, and oblique crossings. The dynamic safety threshold system aligns with the current vibration propagation patterns, providing a scientific basis for subsequent risk assessment. By integrating TBM vibration source and tunnel vibration data, a full-link correlation analysis of "vibration source-propagation-response" is achieved, generating a dynamic vibration attenuation model and a dynamic safety threshold system that fits the real-time tunneling conditions.

[0028] Step 103: Compare the vibration attenuation benchmark database with the dynamic vibration attenuation model, accurately match the on-site construction risk level according to the preset dynamic risk classification rules, and generate corresponding graded early warning instructions.

[0029] This step involves using dynamic threshold differentiation to achieve accurate risk classification and early warning. The vibration attenuation benchmark database in step 101 represents the inherent vibration patterns of the surrounding rock in the region and serves as a reference. The dynamic vibration attenuation model in step 102 represents the real-time vibration propagation patterns under the current tunneling conditions and serves as the core evaluation benchmark. By comparing the vibration attenuation benchmark database with the dynamic vibration attenuation model, the traditional fixed static evaluation threshold is abandoned. For example, using the static benchmark database as a reference, the offset of the dynamic model is checked, and the risk range of the measured PPV is determined by combining the dynamic safety threshold.

[0030] Then, accurately match the on-site construction risk level according to the preset dynamic risk division rules, and generate corresponding hierarchical warning instructions. The dynamic risk division rules are based on the safety allowable vibration velocity of chamber blasting vibration, divide three-level risk thresholds, and superimpose the abnormal duration window auxiliary criterion, and judge the risk by combining the double conditions of amplitude and duration. The hierarchical warning instructions are instructions generated corresponding to different risk levels, which can be, in sequence, the instruction to strengthen monitoring, the instruction to adjust tunneling parameters, and the instruction to stop the machine. The false alarm rate and missed alarm rate of vibration risk monitoring are greatly reduced through the dynamic adaptive threshold discrimination mechanism.

[0031] The dynamic discrimination mechanism in this step combines the duration auxiliary criterion to effectively distinguish real structural risks from instantaneous interference and solve the problem of false triggering of traditional single-threshold warnings. The multi-level risk division corresponds to different disposal strategies to avoid "one-size-fits-all" control. By comprehensively judging based on the inherent laws of the surrounding rock and the real-time working conditions, the accuracy of risk identification is significantly improved. The hierarchical warning instructions provide a standardized output for subsequent multi-terminal prompts and equipment joint control.

[0032] Step 104: According to the hierarchical warning instructions, synchronously push warning information to the existing chamber sound and light device, the warning panel in the TBM cab, and the portable terminal, so as to achieve multi-terminal synchronous warning publicity. When the hierarchical warning level reaches the preset high-risk safety threshold, directly connect to the TBM main control system through the exclusive hard joint control interface of the TBM, and actively issue instructions to forcibly reduce the speed or stop the machine.

[0033] This step is for multi-terminal linked warning and hardware hard joint control active safety defense. The multi-terminal warning system includes three types of terminals: the existing chamber sound and light device (on-site local warning), the warning panel in the TBM cab (operated personnel can view in real time), and the portable terminal (field inspection personnel receive information), to achieve full-post information synchronization. According to the hierarchical warning instructions, synchronously push warning information to the existing chamber sound and light device, the warning panel in the TBM cab, and the portable terminal, so as to achieve multi-terminal synchronous warning publicity. The warning information is pushed synchronously across the whole area, covering multiple operation scenarios such as inside the tunnel, in the cab, and inspection personnel, ensuring that all personnel can timely learn about the risks. When the graded warning level reaches the preset high-risk safety threshold, the TBM's main control system is directly connected via a dedicated hardware control interface to proactively issue forced speed reduction or shutdown commands. This hardware control interface is independent of the software system's hardware communication interface, enabling hardware-level signal transmission unaffected by software crashes or network interruptions. For example, at a medium-risk level, speed reduction and parameter adjustment commands are issued, and the hydraulic proportional valve is adjusted via a programmable logic controller (PLC) to achieve flexible vibration control; at a high-risk level, a hardware emergency stop is triggered, cutting off the cutterhead motor power and resetting the hydraulic valves to lock the equipment's posture. This pure hardware interlocking circuit offers a higher safety level than software control, adapting to the complex electromagnetic and network environment of tunnels and overcoming the limitations of traditional warning systems that only provide early warnings without proactive control. By directly connecting to the TBM's main control system via the dedicated hardware control interface, forced speed reduction or shutdown commands are proactively issued, cutting off high-risk tunneling power at the hardware level and reducing tunneling vibration disturbances at the source, achieving proactive safety defense for TBM construction through existing tunnel complexes.

[0034] This step addresses the issues of untimely and incomplete information delivery caused by single alarm methods by synchronously pushing information across multiple terminals. It breaks through the traditional passive mode of "only alarming, not responding," automatically intervening in TBM operation to reduce vibration disturbance at the source. The hardware-based control is unaffected by software or network failures, reliably executing shutdown operations under high-risk conditions, thus strengthening the final safety line. For medium-risk situations, only tunneling parameters are adjusted; shutdown is only implemented for high-risk situations, ensuring the safety of the tunnel structure while minimizing the impact on construction progress.

[0035] This application first deploys a multi-source sensing network on the structural surface of the existing cavern, and performs filtering and vibration attenuation benchmark database calculation on-site at the edge. This eliminates monitoring blind spots, improves the integrity of vibration data acquisition, reduces the amount of raw data transmission, and lowers transmission delay, enabling high-frequency, real-time, and quantitative detection of micro-vibration signals. Then, the vibration attenuation model is dynamically updated using real-time tunneling vibration source parameters of the TBM, overcoming the limitations of traditional fixed models and thresholds. This allows the vibration propagation law to adapt to real-time tunneling conditions and surrounding rock conditions, significantly improving the accuracy of vibration assessment and reducing the probability of misjudgments and omissions. Next, the vibration attenuation benchmark database is compared with the dynamic vibration attenuation model to generate graded early warning commands, which can distinguish different vibration risk levels, enabling refined risk assessment and providing a basis for differentiated handling. Finally, multi-terminal synchronous early warning is implemented through cavern audio-visual systems, portable terminals, and the TBM cab. Hardware interlocking directly controls the TBM to slow down or stop, constructing a closed-loop management system of monitoring, analysis, early warning, and equipment response. This suppresses vibration disturbances from the source of tunneling and proactively protects the structural safety of the existing cavern.

[0036] In step 101, the synchronous acquisition of micro-vibration signals and noise filtering at the edge side may include the following steps. First, a front-end distributed edge processing architecture is adopted. The distributed edge processing architecture deploys data acquisition, signal processing, feature analysis and other functions on the edge computing nodes near the monitoring site. Unlike centralized processing in the remote cloud, it can complete the entire signal processing process on-site without transmitting massive amounts of raw waveform data.

[0037] Utilizing dedicated hardware in a converged data acquisition and edge computing station, high-sampling-rate synchronous acquisition and analog-to-digital conversion of distributed fiber optic signals and MEMS accelerometer signals are achieved through an on-site fiber optic demodulation unit and a multi-channel synchronous vibration acquisition instrument. The converged data acquisition and edge computing station is an integrated field hardware device that combines signal acquisition, demodulation, computation, and storage functions, serving as the core carrier for edge-side data processing. The fiber optic demodulation unit contains dedicated hardware modules used to analyze the optical signals transmitted from the distributed fiber optic cables, converting them into recognizable vibration electrical signals. The multi-channel synchronous vibration acquisition instrument can simultaneously connect to multiple sensors, ensuring complete consistency of the time axes of each signal, enabling synchronous sampling of multiple measurement points and multiple signal types. Dedicated acquisition equipment achieves high-sampling-rate synchronous acquisition and time-series alignment of both fiber optic and MEMS signals, avoiding time misalignment of multi-source data and ensuring the accuracy of subsequent vibration law analysis and parameter inversion.

[0038] Raw vibration time-history signals are acquired using a triaxial accelerometer. A bandpass filter, built into the edge acquisition node, is used to filter out background environmental noise with frequencies below the first preset threshold and electromagnetic interference (EMI) noise with frequencies above the second preset threshold. The first preset threshold is the low-frequency threshold, which is the lower cutoff frequency of the filter. Signals with frequencies below this value are considered low-frequency background noise and are directly filtered out. The second preset threshold is the high-frequency threshold, which is the upper cutoff frequency of the filter. Signals with frequencies above this value are considered high-frequency EMI and are directly filtered out. By using a bandpass filter to selectively remove environmental noise and EMI, combined with baseline correction to eliminate system errors, the true vibration state of the surrounding rock is restored to the greatest extent possible, preventing interference signals from affecting subsequent analysis at the source.

[0039] The filtered signal is then subjected to baseline correction and integration. Baseline correction is a signal processing technique that corrects for zero-point offset and baseline drift in vibration signals, eliminating systematic errors caused by equipment and environment, as well as systematic errors caused by zero-point drift. Vibration parameters are then calculated through integration, extracting the dominant frequency component and energy characteristic value of the peak particle vibration velocity. The dominant frequency component is the frequency component with the highest energy proportion in the micro-vibration signal and is an important characteristic parameter characterizing the vibration properties of the surrounding rock. The energy characteristic value is a quantitative indicator reflecting the overall energy magnitude of the vibration signal, which can intuitively reflect the disturbance intensity of TBM tunneling on the existing tunnel. Finally, standardized high-precision monitoring data is compiled as a vibration attenuation benchmark database.

[0040] Employing a distributed edge architecture, signal conversion, filtering, and feature extraction are completed on-site, with only standardized feature data uploaded, significantly reducing data transmission volume and solving the problems of insufficient bandwidth and high transmission latency in long-distance tunnel construction, ensuring real-time monitoring. Core features such as peak vibration velocity, dominant frequency, and energy are uniformly extracted to form standardized monitoring data, which can be directly used for the construction and iterative updates of a vibration attenuation benchmark database, achieving seamless integration of acquisition, processing, and database construction processes. All processing functions are integrated within the on-site edge hardware, adapting to the complex working conditions of underground caverns with high dust levels and strong electromagnetic interference. The equipment boasts high operational stability and can meet the needs of long-term continuous monitoring.

[0041] In this embodiment, the tunneling vibration source parameters may include the TBM's total thrust, cutterhead torque, penetration depth, and tunneling speed. These parameters comprehensively characterize the actual working conditions of the tunneling operation, directly determining the rock-breaking strength and the magnitude of vibration disturbance. They serve as the fundamental basis for determining the vibration source strength and provide comprehensive input data for the tunneling working conditions to the model's dynamic calibration. Penetration depth, the depth to which the TBM cutterhead cuts into the surrounding rock per revolution, is a key indicator reflecting tunneling efficiency and rock-breaking load. By using multi-dimensional vibration source parameters, the actual tunneling state of the TBM is fully restored, avoiding the problem of a single parameter providing a one-sided representation of the working conditions and ensuring the integrity of the model's input data.

[0042] In step 102, a vibration attenuation benchmark database is established based on the rock mass mechanical parameters of the cavern and previous measured data using a numerical inversion method. For example, data analysis methods that rely on known measured results and monitoring data to inversely deduce the inherent mechanical parameters of the rock mass and the vibration propagation law are common methods for establishing basic databases in geotechnical engineering. The vibration attenuation benchmark database characterizes the mapping relationship between the vibration source distance, rock mass parameters, and peak vibration velocity of the response particles, providing a static benchmark for subsequent model calibration.

[0043] This system acquires multi-dimensional vibration source condition data in real time, comprising total thrust, cutterhead torque, penetration depth, and tunneling speed during the TBM tunneling process. Based on a pre-defined energy mapping relationship, it calculates the equivalent vibration source energy at the current moment, obtaining key intermediate variables required for model calibration. The energy mapping relationship is a pre-defined conversion rule used to transform TBM mechanical operating parameters into equivalent vibration source energy, achieving a quantitative conversion between mechanical conditions and vibration energy. The equivalent vibration source energy is a unified quantitative index that converts the mechanical disturbances generated by TBM rock breaking and movement into an equivalent value, representing the total energy intensity of the vibration source under the current conditions. It is the core intermediate variable for vibration propagation calculation.

[0044] By utilizing the equivalent source energy and the measured peak velocity of the mass particles, the rock mass parameters in the vibration attenuation benchmark database are updated in real time to adapt to the dynamic changes in geological conditions and tunneling status during the tunneling process. This results in a dynamically updated vibration attenuation benchmark database, ultimately forming a dynamic vibration attenuation model that fits the actual working conditions on site. This solves the shortcomings of traditional static models that cannot follow changes in working conditions and have large prediction deviations.

[0045] As an example, multiple sets of equivalent source energies and corresponding measured peak particle velocities within the data sampling time window can be selected as model fitting sample data. The data sampling time window is a manually set continuous data acquisition duration used to extract a segment of historical data under stable operating conditions as fitting samples, avoiding interference from transient abnormal data.

[0046] A vibration attenuation prediction model adapted to the tunneling conditions of cavern groups is built based on the Sadovsky formula. The site coefficient and attenuation index, characterizing the propagation characteristics of the surrounding rock, are set as unknown rock mass parameters to be optimized. The site coefficient is an empirical parameter representing the influence of the overall geological and topographical conditions of the region on vibration propagation, while the attenuation index reflects the characteristic parameter of the rate of vibration attenuation in the surrounding rock medium; the larger the index value, the more significant the attenuation of vibration energy with propagation distance. Based on this, a quantitative mapping relationship between source energy, propagation distance, and particle vibration velocity is established.

[0047] Subsequently, the least squares method was used to fit and analyze the sample data, iteratively solving for the optimal site coefficient and attenuation index that minimized the model's prediction error. The latest rock mass parameters obtained from the solution were then updated in real time to the vibration attenuation benchmark database, achieving dynamic adaptive correction of the rock mass vibration attenuation characteristics. Based on the Sadovsky formula combined with engineering condition optimization modeling, and using the least squares iterative optimization method, the optimal site coefficient and attenuation index were obtained, minimizing the model's prediction error and improving the accuracy of vibration law calculation.

[0048] In step 103, the permissible vibration velocity of the mass points corresponding to the existing cavern group during blasting vibration is first obtained. This permissible vibration velocity is a common safety indicator in the fields of water conservancy and underground engineering. It is the ultimate safe vibration velocity determined by combining the cavern structural form, support type, and surrounding rock grade, representing the maximum vibration limit that the existing cavern and support structure can withstand over a long period. It serves as the legal and technical benchmark for setting the thresholds in this scheme. Using this as the core benchmark for risk assessment, a three-tiered risk threshold—the first threshold, the second threshold, and the third threshold—is set according to the structural safety tolerance standard. These are three levels of critical values ​​based on the permissible vibration velocity, from low to high: the first threshold, the second threshold, and the third threshold, corresponding to low, medium, and high risk levels. This is used to quantitatively distinguish the severity of vibration disturbances and achieve a refined gradient division of risk levels.

[0049] The output of the dynamically calibrated vibration attenuation model is compared and judged with the three preset thresholds step by step to match different tunneling vibration risk conditions and form a differentiated hierarchical control strategy.

[0050] If the monitored vibration response value is greater than the first threshold and less than or equal to the second threshold, it is determined to be a low-risk working condition, and a first warning instruction is generated. The first warning instruction is the control instruction corresponding to the low-risk working condition. The core execution content is to increase the frequency of data collection and inspection, continuously track the vibration change trend, and implement the corresponding control strategy of strengthening the frequency of on-site monitoring.

[0051] If the monitored vibration response value is greater than the second threshold and less than or equal to the third threshold, it is determined to be a medium-risk working condition, and a second early warning instruction is generated. The second early warning instruction is a control instruction corresponding to the medium-risk working condition, which is used to guide the on-site adjustment of tunneling parameters such as TBM thrust, tunneling speed, and cutterhead rotation speed to reduce the intensity of vibration disturbance from the source and implement a control strategy of dynamically adjusting TBM tunneling parameters.

[0052] If the monitored vibration response value is greater than the third threshold, it is determined to be a high-risk and high-danger working condition, and a shutdown warning command is generated. The shutdown warning command corresponds to the control command for high-risk working conditions and serves as the highest-level handling command. It is used to issue a control command to stop the TBM tunneling operation and immediately trigger the tunneling termination control mechanism.

[0053] The system adopts the industry-standard permissible particle vibration velocity for blasting vibration as a benchmark. Threshold settings comply with engineering safety standards, ensuring the scientific and compliant nature of risk assessment results and avoiding subjective biases arising from human experience. A three-tiered threshold system enables hierarchical and graded assessment, moving away from the traditional single-threshold approach that simply classifies safety and danger, and accurately distinguishing risk levels corresponding to different disturbance intensities. For low, medium, and high risks, three corresponding response methods—enhanced monitoring, adjustment of tunneling parameters, and emergency shutdown—are applied, forming a tiered control system. This ensures the safety of the tunnel structure while minimizing unnecessary downtime, balancing construction safety and project progress. The system is used in conjunction with a dynamically calibrated vibration attenuation model, using real-time vibration calculations as the basis for judgment. Combined with the dynamic threshold system, this further improves the accuracy of risk identification and effectively reduces the probability of misjudgment. Risk levels, early warning commands, and on-site control actions are all correlated, with standardized command outputs facilitating automatic system execution and enabling on-site personnel to quickly understand and implement response measures, thus improving emergency response efficiency.

[0054] In step 104, if a second early warning command is generated, a soft-adjustment active control mode is adopted. This mode is a flexible control approach for medium-risk conditions, relying on PLC output control signals to adjust hydraulic components and gradually change the TBM's operating parameters without directly cutting off power; it is a gradual risk management method. A speed control signal is sent to the TBM's main control PLC via a hard-line interface to dynamically adjust the opening of the proportional valve in the hydraulic propulsion system, flexibly reducing the TBM's tunneling speed.

[0055] If a shutdown command is generated, the hardware-level forced shutdown protection mode is activated. This mode represents the highest level of safety protection for high-risk operating conditions, relying on a hardware circuit to achieve forced shutdown. It is unaffected by software programs or network status, offering higher safety redundancy. The hard-wired relay connected in series in the TBM emergency stop safety circuit is directly driven via a hard-wired control interface. Hard-wired relays are commonly used industrial electrical switching components, controlling contact opening and closing by switching a coil on and off. In this embodiment, they are connected to the emergency stop circuit in a normally closed manner, conforming to fail-safe design principles and forcibly opening the normally closed contacts of the relay.

[0056] Upon detecting the level change signal in the emergency stop safety circuit, the TBM main control PLC immediately and forcefully cuts off the power supply to the cutterhead drive motor. Simultaneously, it forcibly resets the proportional valve of the hydraulic propulsion system to the neutral position, quickly locking the TBM's tunneling posture and preventing abnormal actions such as slippage or deviation. The proportional valve of the hydraulic propulsion system is a core control component of the TBM's hydraulic system. By changing the valve opening, it regulates the flow and pressure of hydraulic oil, thereby controlling the overall propulsion speed.

[0057] Medium-risk areas employ soft-adjustment speed reduction, eliminating the need to interrupt tunneling operations and ensuring construction continuity while mitigating vibration disturbances. High-risk areas utilize hardware-based forced shutdown to completely eliminate disturbance sources, forming a comprehensive control system that balances high and low risks. Flexible speed regulation relies on the electrical control system for precise adjustment of operating conditions. The hardware emergency stop circuit is independent of the software system and can operate normally even in abnormal situations such as program crashes or network interruptions, significantly improving the reliability of the entire control system. The emergency stop circuit uses a normally closed contact design, adhering to industrial fail-safe principles. Even in the event of line breaks or relay failures, it will automatically trigger a shutdown, eliminating safety blind spots. After shutdown, the power supply to the cutterhead motor is simultaneously cut off, and the hydraulic valves are reset to the neutral position, doubly locking the equipment status and effectively preventing the main unit from slipping or malfunctioning after shutdown, protecting the equipment and surrounding structures. Commands travel directly from the early warning system to the TBM main control unit via a hardware-based control interface. The short signal transmission link and low action execution delay allow for rapid adjustment of operating conditions or emergency shutdown, quickly mitigating safety risks.

[0058] In this embodiment, a step based on duration-assisted determination may also be included. Specifically, real-time micro-vibration signals during TBM tunneling are acquired, and the peak particle velocity and dominant frequency core characteristic parameters of the measuring points are extracted simultaneously to track the changes in tunneling vibration response in real time. The dominant frequency is the frequency component with the highest energy proportion in the micro-vibration signal and is an important characteristic parameter reflecting the inherent characteristics of surrounding rock vibration and distinguishing vibration sources.

[0059] Based on the vibration amplitude threshold classification, and using the first, second, and third threshold risk standards as a foundation, an auxiliary judgment condition of abnormal duration is added to construct a dual early warning discrimination mechanism based on amplitude threshold and duration. Abnormal duration is the length of time that the peak vibration velocity of the measuring particle exceeds the corresponding risk threshold, and this over-limit state is maintained continuously. When the peak vibration velocity of a particle exceeds the corresponding risk threshold, the duration of the abnormal vibration condition is continuously monitored. Only when the duration of the abnormal vibration condition exceeds the anomaly judgment time window is the corresponding level of warning command triggered, completing risk identification and command output. The anomaly judgment time window is a pre-set time threshold value based on the engineering conditions and the safety level of the cavern structure. It serves as the boundary for distinguishing between instantaneous disturbances and persistent real risks.

[0060] During TBM tunneling, short-term over-limit phenomena such as mechanical impact and electromagnetic disturbances are prone to occur. Duration verification can eliminate these non-structural, instantaneous anomalies, avoiding meaningless alarms. Early warnings are only triggered for prolonged, continuous vibration exceeding limits, accurately identifying the real risks threatening the tunnel structure and significantly reducing the probability of false alarms and false triggers. This auxiliary judgment step is superimposed on the original three-level threshold classification rules, requiring no modification to the basic risk classification logic and seamlessly integrating with the aforementioned monitoring, judgment, and control processes. Combining peak mass velocity and dominant frequency synchronous monitoring with amplitude and duration dual criteria, a multi-dimensional assessment of vibration status is conducted, making risk assessment results more rigorous and reliable. Furthermore, it can reduce frequent speed reductions and shutdowns caused by short-term disturbances, strengthening safety defenses while minimizing the impact of control actions on normal construction progress.

[0061] The following is an example of a specific implementation. The application scenario is as follows: During the construction of an underground powerhouse in a water conservancy project, a TBM needs to pass through the rock column between the existing main transformer room and the main powerhouse in a spiral path over a short distance.

[0062] Figure 2 This is a general architecture block diagram provided in a specific embodiment of this application. See also... Figure 2 The system in this embodiment can be composed of four main parts: a multi-source signal sensing network 1, a fusion data acquisition and edge computing station 2, a central intelligent analysis and early warning platform 3, and a multi-linkage alarm terminal 4, supplemented by a vibration source synchronous acquisition device 5 installed on the TBM main beam, forming a complete "sensing-analysis-early warning-joint control" closed-loop system. The multi-source signal sensing network 1 is used to comprehensively capture micro-vibration information induced by TBM tunneling from both "surface" and "point" dimensions.

[0063] Figure 3 This is a schematic diagram of the cross-sectional layout of a multi-source signal sensing network in an existing cavern group, provided in a specific embodiment of this application. See also... Figure 3 For the distributed optical fiber subsystem 11, this embodiment of the application uses a distributed acoustic wave sensing optical fiber approximately 800 meters long. This optical fiber is laid along the support sprayed layer surface of the main transformer room and the side walls and roof arch of the main plant, bending in a zigzag path with a vertical turning interval of 3 meters. It is tightly fixed to the lining surface with special anchors and adhesives to form a continuous vibration sensing surface covering the entire monitoring area.

[0064] Its working principle is as follows: the fiber demodulation unit 21 in the integrated data acquisition and edge computing station 2 sends a probe light pulse to the fiber and receives and analyzes the phase change of the Rayleigh backscattered light signal in real time, thereby demodulating the vibration waveform at each meter along the fiber length direction, which is equivalent to deploying tens of thousands of virtual vibration sensors on the inner wall of the cave, and can draw a vibration distribution heat map of the entire monitored area of ​​the cave.

[0065] For the critical lattice-type MEMS accelerometer subsystem 12, this embodiment of the application installs high-sensitivity, low-noise triaxial MEMS accelerometers at key stress-bearing locations such as the arch foot, arch waist, arch crown, and rock column centerline of the high sidewalls of the main transformer room and main plant. Each accelerometer is firmly fixed to the bedrock or sprayed layer surface by expansion bolts and rigid brackets to ensure synchronous vibration with the structural substrate, and is protected by a protective cover to avoid the influence of construction dust and splashing water.

[0066] Figure 4 This is a unfolded diagram of a non-equidistant grid arrangement in the longitudinal section of an existing cavern, as provided in a specific embodiment of this application for a MEMS accelerometer. See also... Figure 4 The MEMS accelerometers are arranged using a non-equidistant grid strategy in the longitudinal direction. Based on the results of previous finite element numerical simulation analysis, the core vibration influence zone is defined as a 15-meter area on each side of the expected orthogonal underpass centerline of the TBM drainage corridor. Within this core zone, the spacing of the sensors along the tunnel axis is increased to 1.5 meters. In the edge influence zone outside the core zone, the spacing is increased to 10 meters. Through this gradient arrangement, the system can concentrate monitoring resources and achieve high spatial resolution vibration data acquisition in the highest risk area, accurately capturing key dynamic parameters such as peak particle velocity, acceleration, and dominant frequency. The integrated data acquisition and edge computing station 2 is placed in a safe corner of the main transformer room, responsible for localized high-speed signal acquisition and front-end intelligent processing.

[0067] It mainly contains three units: fiber optic demodulation unit 21, multi-channel synchronous vibration acquisition instrument 22, and edge computing module 23.

[0068] The fiber optic demodulation unit 21 is responsible for transmitting optical signals and receiving and demodulating scattered signals, outputting vibration waveform data in real time. The multi-channel synchronous vibration acquisition instrument 22 is a high-precision synchronous data acquisition device with one channel, connected to all MEMS accelerometer subsystems 12 through shielded cables, performing synchronous analog-to-digital conversion on all channels, and completely preserving the characteristics of TBM rock-breaking vibration signals from low frequency to high frequency.

[0069] The edge computing module 23 embeds a high-performance FPGA processor. This module is configured to execute filtering algorithms locally in real time to remove non-TBM vibration source noise from the background, such as construction machinery and water pump operation. After filtering, the module calculates the peak particle velocity (PPV) and dominant vibration frequency of all measuring points online once per second, and packages these key characteristic parameters, sending them in real time to the central intelligent analysis and early warning platform 3 via the underground industrial communication network.

[0070] The Central Intelligent Analysis and Early Warning Platform 3 is located in the ground monitoring center and consists of an industrial-grade server, serving as the intelligent core of the entire system.

[0071] The adaptive vibration risk classification model 31, its construction and operation include the following steps: First, before the system is deployed, based on the rock mechanics parameters of the tunnel section and the previous measured data, the numerical inversion method is used to establish the "vibration source distance-rock parameters-response PPV" attenuation benchmark database 32 for the tunnel.

[0072] Secondly, during system operation, the model acquires real-time cutterhead thrust and torque data from the vibration source synchronous acquisition device 5 installed on the TBM main beam via a wired data link. Using these real-time tunneling parameters, the model dynamically calibrates the baseline attenuation curve to accurately reproduce the actual vibration propagation pattern under the current construction conditions.

[0073] Based on this, the model sets four threshold logic units 33, which are based on the allowable PPV value determined by the evaluation of the existing cavern structure: the green safety zone is 40% below the allowable value; the blue caution zone is 40% to 70% of the allowable value; the yellow warning zone is 70% to 90% of the allowable value; and the red danger zone is 90% and above the allowable value. Each threshold level can be parametrically adjusted in the model according to engineering needs.

[0074] The multi-level linkage alarm terminal 4 is used to accurately push the hierarchical early warning instructions generated by the central platform to all relevant responsible parties in the most direct and effective way.

[0075] In this embodiment, multiple sets of explosion-proof and waterproof three-color alarm devices 41 are installed in the main transformer room and main powerhouse. Different warning levels correspond to different alarm modes, allowing personnel inside the tunnel to identify the current risk level without having to check a screen.

[0076] Figure 5 This is a schematic diagram of the display interface of a remote early warning panel provided in a specific embodiment of this application. See also... Figure 5 The TBM cab remote early warning panel 42 is the core interactive interface between the system and the TBM operator. It is an industrial-grade explosion-proof tablet computer installed on the TBM main control console in the drainage gallery. The panel interface is divided into four main areas: the top status bar displays the current warning level in real time; the left chart area dynamically plots the real-time PPV time-series curves of key measuring points and marks them with yellow and red threshold reference lines; the right heat map area displays the vibration response distribution of the existing cavern with color gradients. Figure 5(For grayscale illustration, the actual control interface uses a color gradient display) to help operators intuitively understand the specific locations of concentrated vibration. The bottom instruction suggestion area automatically generates and clearly displays control suggestions in text form based on the current warning level, such as "It is recommended to immediately reduce the tunneling speed to below 20% of the current value and reduce the cutterhead thrust." This panel can also be connected to the TBM control system via a data interface to issue automatic speed reduction or shutdown hard-linked control commands under the highest danger level.

[0077] The portable mobile early warning terminal 43 is an industrial-grade handheld device equipped for on-site safety management personnel. Through a wireless network covering the entire underground cavern complex, it can simultaneously receive real-time early warning information and key measurement point data pushed by the central platform, ensuring that management personnel can grasp the safety status of the caverns at any time from any location.

[0078] Figure 6 This is a flowchart illustrating an early warning logic and linkage control system provided in a specific embodiment of this application. The following is in conjunction with... Figure 6 The process described herein is a complete overview of the system's operation.

[0079] When the TBM drainage tunnel reaches the boundary of the preset influence zone below the main transformer room, the system switches from sleep monitoring mode to full-power operation. The multi-source signal sensing network 1 continuously detects vibration signals, and the fusion data acquisition and edge computing station 2 continuously completes acquisition, filtering, and online PPV calculation, and uploads the feature data in real time.

[0080] When the TBM tunnels to a certain mileage directly below the main transformer room, the central intelligent analysis and early warning system 3 receives a data packet uploaded by the edge computing station 2. After comparison by the adaptive risk classification model 31 and the vibration source strength calibration, it is determined that the real-time PPV value of the MEMS accelerometer measuring point PPV-14 located in the core encryption area has exceeded the preset yellow warning threshold of the measuring point, but has not reached the red danger threshold.

[0081] Based on this judgment, the central intelligent analysis and early warning system 3 immediately generates a yellow warning command and triggers the multi-linked alarm terminal 4 to initiate the yellow warning response procedure: Inside the main transformer room and main plant, the yellow rotating light of the on-site multi-level audible and visual alarm 41 began to rotate and emitted intermittent buzzing sounds, reminding all workers inside the tunnel that they were in a warning state, to pay attention to safety, and to follow up on subsequent instructions.

[0082] The TBM cab remote early warning panel 42 immediately pops up an alarm window with a yellow border. The peak value of the PPV-14 measuring point in the left time-series curve area has exceeded the yellow threshold line. The heat map on the right shows that the arch foot area of ​​the main transformer room is in a high-risk state. The bottom instruction suggestion area pushes out specific operation instructions that suggest immediately reducing the tunneling speed to less than 20% of the current value and reducing the cutterhead thrust.

[0083] Security personnel conducting patrols in the nearby area had their portable mobile early warning terminals vibrate and display the same yellow warning message.

[0084] After receiving the panel instructions, the TBM operator immediately performed the deceleration and thrust reduction operations as suggested, and the real-time PPV value at the PPV-14 measuring point returned to the green safe range.

[0085] Once the central intelligent analysis and early warning platform 3 detects that the risk status has been cleared, it automatically cancels the yellow warning command, the audible and visual alarms stop sounding, the panel alarm window closes, and the system returns to normal monitoring status.

[0086] The entire process, from the triggering of the warning to the safe lifting of the alert, was completed in a very short time, ensuring the continuity of tunneling construction and the structural safety of the existing caverns.

[0087] Figure 7 This is a schematic diagram of a micro-vibration monitoring and early warning device for a TBM traversing an existing cavern group, provided in an embodiment of this application. Figure 7 As shown, the micro-vibration monitoring and early warning device 700 for TBMs traversing existing cavern groups may include a data acquisition module 701, a modeling module 702, an early warning module 703, and a control module 704.

[0088] The acquisition module 701 is used to deploy a multi-source signal sensing network on the structural surface of the existing cavern, synchronously acquire micro-vibration signals on the edge side and perform noise filtering preprocessing, calculate the real-time response data of vibration velocity peak value (PPV) at each monitoring point, and build and update the vibration attenuation benchmark database based on the PPV response data of the monitoring points.

[0089] The modeling module 702 is used to collect core tunneling vibration source parameters in real time during the tunneling process of the TBM host, call the pre-built vibration attenuation benchmark database, and dynamically calibrate the vibration attenuation benchmark database in combination with the real-time collected tunneling vibration source parameters to generate a dynamic vibration attenuation model and a dynamic safety threshold system that fits the real-time tunneling conditions.

[0090] The early warning module 703 is used to compare the vibration attenuation benchmark database with the dynamic vibration attenuation model, accurately match the on-site construction risk level according to the preset dynamic risk classification rules, and generate corresponding graded early warning instructions.

[0091] The control module 704 is used to push warning information to the existing cavern sound and light devices, the TBM cab warning panel and portable terminal in accordance with the graded warning instructions, so as to realize the synchronous warning publicity of multiple terminals. When the graded warning level reaches the preset high-risk safety threshold, it directly connects to the TBM main control system through the TBM dedicated hard control interface and actively issues a forced speed reduction or shutdown command.

[0092] Among them, the acquisition module 701, modeling module 702, early warning module 703 and control module 704 can be used to execute steps 101-104 in the embodiment of the above-mentioned micro-vibration monitoring and early warning method for TBM crossing existing cavern groups. For the specific implementation of these modules and more details, please refer to the corresponding method section, which will not be elaborated here.

[0093] This application also provides a computer-readable storage medium storing a program that can be loaded and executed by a processor, which is a micro-vibration monitoring and early warning method for TBMs traversing existing cavern groups according to any of the embodiments of this application.

[0094] Those skilled in the art will understand that all or part of the functions of the various methods in the above embodiments can be implemented by hardware or by computer programs. When all or part of the functions in the above embodiments are implemented by computer programs, the program can be stored in a computer-readable storage medium, which may include: read-only memory, random access memory, disk, optical disk, hard disk, etc., and the program is executed by a computer to achieve the above functions. For example, the program can be stored in the memory of a device, and when the program in the memory is executed by the processor, all or part of the above functions can be achieved. In addition, when all or part of the functions in the above embodiments are implemented by computer programs, the program can also be stored in a server, another computer, disk, optical disk, flash drive, or external hard drive, etc., and can be downloaded or copied to the memory of a local device, or the system of the local device can be updated. When the program in the memory is executed by the processor, all or part of the functions in the above embodiments can be achieved.

[0095] The above examples illustrate this application only to aid understanding and are not intended to limit its scope. Those skilled in the art to which this application pertains can make various simple deductions, modifications, or substitutions based on the ideas presented.

Claims

1. A micro-vibration monitoring and early warning method for TBM crossing existing cavern groups, characterized in that, The application relates to a tunnel construction safety monitoring and early warning method based on vibration attenuation. A multi-source signal sensing network is arranged on the structure surface of an existing chamber, micro-vibration signals are synchronously collected on the edge side and subjected to noise filtering pretreatment, vibration velocity peak value (PPV) real-time response data of each monitoring point are solved in real time, and a vibration attenuation benchmark database is built and updated based on the PPV response data of the monitoring points; Real-time tunneling vibration source parameters in the TBM main machine tunneling process are collected, the pre-constructed vibration attenuation benchmark database is called, and the vibration attenuation benchmark database is dynamically calibrated in combination with the real-time collected tunneling vibration source parameters, a dynamic vibration attenuation model and a dynamic safety threshold system are generated, and the core tunneling vibration source parameters include tunneling thrust and torque; The vibration attenuation benchmark database and the dynamic vibration attenuation model are compared, a field construction risk grade is accurately matched according to a preset dynamic risk division rule, and corresponding grading early warning instructions are generated; According to the grading early warning instructions, early warning information is synchronously pushed to an existing chamber sound and light device, a TBM driver's cab early warning panel and a portable terminal, multi-terminal synchronous early warning publicity is realized, when the grading early warning grade reaches a preset high-risk safety threshold, a TBM special hard connection control interface is directly connected to a TBM main control system, and a forced speed reduction or shutdown instruction is actively issued.

2. The micro-vibration monitoring and early warning method for TBM crossing existing cavern groups according to claim 1, characterized in that, The multi-source signal sensing network comprises a distributed optical fiber subsystem and a key point array MEMS accelerometer subsystem, and forms a double-sensor collaborative monitoring architecture with global coverage and key reinforcement; The distributed optical fiber subsystem adopts distributed acoustic or vibration sensing optical fibers, is arranged in a meander shape in a support layer or on a surface along a longitudinal direction and a ring direction of the existing chamber, backscattering light signals of demodulated probe light pulses are detected, and uninterrupted, dead-angle-free and global continuous monitoring of the overall vibration field of the existing chamber is realized; The key point array MEMS accelerometer subsystem is composed of three-axis MEMS accelerometers, is arranged in key stress and easy-to-damage risk areas of arch feet, arch waists, arch tops and rock columns of the chamber, adopts a differential gradient arrangement strategy, and is densely arranged in a high-risk core area of a TBM passing-through center projection and sparsely arranged in other low-risk areas in a non-equidistant grid manner.

3. The micro-vibration monitoring and early warning method for TBM crossing existing cavern groups according to claim 1, characterized in that, The micro-vibration signals are synchronously collected on the edge side and subjected to noise filtering processing, comprising the following steps: A front-end distributed edge processing architecture is adopted, special hardware of a fusion data acquisition and edge computing station is relied on, distributed optical fiber signals and MEMS accelerometer signals are synchronously collected and analog-digital converted at a high sampling rate through an in-station optical fiber demodulation unit and a multi-channel synchronous vibration collector; Original vibration time history signals are collected through three-axis acceleration sensors, background environmental noise with a frequency lower than a first preset threshold and electromagnetic interference clutter with a frequency higher than a second preset threshold are removed through a band-pass filter built in an edge collection node; Baseline correction and integration processing are conducted on the filtered signals, main frequency components and energy eigenvalues of particle peak vibration velocities are extracted, standardized high-precision monitoring data are formed, and the data are used as the vibration attenuation benchmark database.

4. The micro-vibration monitoring and early warning method for TBM crossing existing cavern groups according to claim 1, characterized in that, The tunneling vibration source parameters include total thrust of the TBM, cutter head torque, penetration and tunneling speed. The dynamic calibration of the vibration attenuation benchmark database based on the real-time acquired tunneling vibration source parameters, generating a dynamic vibration attenuation model, includes: Based on the rock mechanics parameters of the cavern and the previous measured data, the vibration attenuation benchmark database was established using the numerical inversion method. The vibration attenuation benchmark database characterizes the mapping relationship between the vibration source distance, rock parameters and the peak vibration velocity of the response particles, providing a static benchmark for subsequent model calibration. The system acquires multi-dimensional vibration source condition data, including total thrust, cutterhead torque, penetration depth, and tunneling speed, during the real-time TBM tunneling process. Based on the preset energy mapping relationship, it calculates the equivalent vibration source energy at the current moment and obtains the key intermediate variables required for model calibration. Using the equivalent source energy and the measured peak particle vibration velocity, the rock mass parameters in the vibration attenuation benchmark database are updated in real time to adapt to the dynamic changes in geological conditions and tunneling status during the tunneling process, resulting in the dynamically updated vibration attenuation benchmark database, and finally forming a dynamic vibration attenuation model that fits the actual working conditions on site.

5. The micro-vibration monitoring and early warning method for TBM crossing existing cavern groups according to claim 4, characterized in that, The process of using the equivalent source energy and measured peak particle velocity to perform real-time inversion and updating of rock mass parameters in the vibration attenuation benchmark database includes: Multiple sets of equivalent source energies and corresponding measured peak particle vibration velocities within the data sampling time window are selected as model fitting sample data; Based on the Sadovsky formula, a vibration attenuation prediction model adapted to the tunneling conditions of cavern groups is built. The site coefficient and attenuation index, which characterize the propagation characteristics of the surrounding rock, are set as unknown rock mass parameters to be optimized, and a quantitative mapping relationship between source energy, propagation distance and particle vibration velocity is established. The least squares method is used to fit and analyze the sample data, and the optimal site coefficient and attenuation index that minimize the model prediction error are solved iteratively. The latest rock mass parameters obtained from the solution are then covered and updated in real time to the vibration attenuation benchmark database to realize dynamic adaptive correction of the rock mass vibration attenuation characteristics.

6. The micro-vibration monitoring and early warning method for TBM crossing existing cavern groups according to claim 1, characterized in that, The process involves comparing the vibration attenuation benchmark database with the dynamic vibration attenuation model, accurately matching the on-site construction risk level according to preset dynamic risk classification rules, and generating corresponding graded early warning instructions, including: Obtain the permissible mass vibration velocity for blasting vibration corresponding to the existing cavern group, and use it as the core benchmark for risk assessment. Based on the structural safety tolerance standard, set a three-level gradient risk threshold of first threshold, second threshold and third threshold to achieve a refined gradient division of risk level. The output of the dynamically calibrated vibration attenuation model is compared and judged with the three preset thresholds step by step to match different tunneling vibration risk conditions and form a differentiated hierarchical control strategy. If the monitored vibration response value is greater than the first threshold and less than or equal to the second threshold, it is determined to be a low-risk working condition, a first early warning instruction is generated, and a corresponding control strategy to strengthen the frequency of on-site monitoring is executed. If the monitored vibration response value is greater than the second threshold and less than or equal to the third threshold, it is determined to be a medium-risk working condition, a second early warning instruction is generated, and a control strategy for dynamically adjusting the TBM tunneling parameters is executed accordingly. If the monitored vibration response value is greater than the third threshold, it is determined to be a high-risk and high-danger working condition, a shutdown warning command is generated, and the excavation and termination control mechanism is immediately triggered.

7. The TBM micro-vibration monitoring and early warning method through an existing cavern group according to claim 6, characterized in that, The method of directly connecting to the TBM main control system via the TBM-specific hardware interface to actively issue forced speed reduction or shutdown commands includes: If the second early warning command is generated, the soft adjustment active prevention and control mode is adopted. The speed control signal is sent to the TBM main control PLC through the hard control interface to dynamically adjust the opening of the proportional valve of the hydraulic propulsion system and flexibly reduce the TBM tunneling propulsion speed. If the shutdown command is generated, the hardware-level forced disconnection protection mode is activated, and the hard-wired relay connected in series in the TBM emergency stop safety circuit is directly driven through the hard-wired control interface to force the normally closed contacts of the relay to open. After the TBM main control PLC detects the level change signal of the emergency stop safety circuit in real time, it immediately cuts off the power supply to the cutterhead drive motor and forces the proportional valve of the hydraulic propulsion system to the neutral position, quickly locking the TBM host tunneling posture.

8. The micro-vibration monitoring and early warning method for TBM crossing existing cavern groups according to claim 6, characterized in that, It also includes a step for determining based on duration, which includes: Real-time micro-vibration signals are collected during the TBM tunneling process, and the peak vibration velocity and main vibration frequency of the mass points are extracted simultaneously to track the changes in the tunneling vibration response in real time. Based on the classification of vibration amplitude thresholds, and using the classification risk standards of the first, second, and third thresholds as a basis, an auxiliary judgment condition for abnormal duration is added to construct a dual early warning and discrimination mechanism based on amplitude threshold and duration. When the peak vibration velocity of the particle is detected to exceed the risk threshold of the corresponding level, the duration of the abnormal vibration condition is continuously monitored. Only when the duration of the abnormal vibration condition exceeds the abnormal judgment time window is the corresponding level of warning instruction triggered.

9. A TBM crossing existing cavern group micro-vibration monitoring and early warning device, characterized in that, include: The acquisition module is used to deploy a multi-source signal sensing network on the structural surface of the existing cavern, synchronously acquire micro-vibration signals at the edge and perform noise filtering preprocessing, calculate the real-time response data of vibration velocity peak value (PPV) at each monitoring point, and build and update the vibration attenuation benchmark database based on the PPV response data of the monitoring points. The modeling module is used to collect core tunneling vibration source parameters in real time during the tunneling process of the TBM host, call the pre-built vibration attenuation benchmark database, and dynamically calibrate the vibration attenuation benchmark database in combination with the real-time collected tunneling vibration source parameters to generate a dynamic vibration attenuation model and a dynamic safety threshold system that fits the real-time tunneling conditions. The early warning module is used to compare the vibration attenuation benchmark database with the dynamic vibration attenuation model, accurately match the on-site construction risk level according to the preset dynamic risk classification rules, and generate corresponding graded early warning instructions. The control module is used to push warning information to the existing cavern audio-visual devices, TBM cab warning panel and portable terminal in accordance with the graded warning instructions, so as to realize the synchronous warning publicity of multiple terminals. When the graded warning level reaches the preset high-risk safety threshold, it directly connects to the TBM main control system through the TBM dedicated hard control interface and actively issues a forced speed reduction or shutdown command.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a program that can be loaded by a processor and executed as described in any one of claims 1 to 8, a method for monitoring and early warning of micro-vibrations of a TBM traversing an existing cavern group.

Images