Intelligent perception anchor rod system and supporting state dynamic evaluation method

Abstract

The application discloses an intelligent perception anchor rod system and a supporting state dynamic evaluation method, belongs to the technical field of underground engineering safety monitoring, and relates to the technical field of underground engineering safety monitoring.The system comprises a multi-physical perception unit, a power module, an edge processing and communication module and a data processing terminal.The multi-physical perception unit comprises a thermal pulse sensor, a three-dimensional stress perception unit and a distributed optical fiber acoustic wave sensor which are integrated in a hollow anchor rod body; the power module and the edge processing and communication module are installed in the inner cavity of the hollow anchor rod body; and the data processing terminal is installed in a roadway sidewall or a chamber and is connected with the edge processing and communication module in a wireless communication mode.The method comprises the following steps: a heterogeneous sensor array is synchronously triggered and collected; on-orbit demodulation and feature compression are carried out; a grouting body spatial form and thermal physical property evolution is reconstructed; an anchor rod axial force field and a transverse supporting stress field are reconstructed; an anchoring system integrity index is calculated; a surrounding rock and supporting structure interaction stability index is calculated; and multi-level early warning triggering and decision output are carried out.The application can realize synchronous perception of multi-source information, dynamic and accurate diagnosis of a supporting state and multi-level early warning.

Description

Technical Field

[0001] This invention belongs to the field of underground engineering support monitoring technology, specifically relating to an intelligent sensing anchor system and a method for dynamic evaluation of support status. Background Technology

[0002] Rock bolt support is one of the core technologies for controlling the stability of surrounding rock and ensuring the safety of construction and operation in underground engineering projects such as mines and tunnels. Traditional rock bolts rely solely on mechanical anchoring to achieve their support function, and their support effectiveness can only be indirectly assessed through post-construction spot checks and monitoring of surrounding rock surface displacement. They cannot directly obtain the stress state of the rock bolt body, the performance of the grout, or the true mechanical information of the interaction between the rock bolt and the surrounding rock. Especially under complex geological conditions such as deep high ground stress and weak, fractured rock strata, the quality of rock bolt grouting and the range of grout diffusion are highly concealed. Damage and deterioration of the grout-rock interface and abnormal redistribution of surrounding rock stress are difficult to identify in a timely manner, easily leading to progressive failure of the anchoring system and subsequently inducing engineering disasters such as roof falls and spalling. Therefore, achieving full-cycle, in-situ, multi-parameter intelligent perception and dynamic assessment of the rock bolt support status is crucial for effectively improving the reliability of the support system and achieving early warning of disasters. It is also a core technical challenge that urgently needs to be overcome in the field of geotechnical engineering safety.

[0003] Existing anchor bolt support condition monitoring technologies mostly focus on the detection of single physical quantities, such as using resistance strain gauges to monitor anchor bolt axial force, fiber optic grating sensors to monitor bolt strain, and point sensors to monitor local stress. These technologies generally have several limitations: First, the sensing units are mostly externally mounted or partially attached, failing to comprehensively characterize the three-dimensional stress characteristics and deformation distribution patterns along the entire length of the anchor bolt. Second, monitoring systems rely heavily on external power supply and wired data transmission, which presents numerous problems in harsh conditions such as dampness, strong vibration, and limited space in underground tunnels, including difficulties in wiring, easy equipment damage, high maintenance costs, and poor long-term power supply stability. Third, the monitoring function is limited, unable to simultaneously acquire information on the spatial distribution of grout, solidification state, and micro-fracture activity of the surrounding rock, making it difficult to achieve a comprehensive diagnosis of the anchoring system integrity from the perspective of multi-field coupling of force, heat, and sound. Furthermore, existing technologies are not adapted to the actual conditions of engineering sites without a stable power supply, thus failing to meet the needs of large-scale applications requiring unattended, long-term online monitoring. In summary, current technologies have not yet formed an embedded integrated intelligent sensing anchor bolt system that integrates energy self-sufficiency, multi-source information fusion, wireless transmission, and intelligent diagnosis.

[0004] To effectively address the shortcomings of existing technologies, there is an urgent need to provide an intelligent sensing anchor system and a dynamic assessment method for support status based on thermoelectric power generation and multi-source information fusion. Summary of the Invention

[0005] To address the problems existing in the prior art, this invention provides an intelligent sensing anchor bolt system and a method for dynamic assessment of support status. This anchor bolt system can comprehensively and synchronously sense multi-physics field information such as grouting quality, bolt stress, and surrounding rock damage. Furthermore, its self-powered operation solves the technical challenges of traditional monitoring systems, such as difficulties in downhole wiring, heavy reliance on external power sources, and difficulty in long-term stable offline operation. This method enables dynamic and accurate diagnosis and graded early warning of support status, providing advanced criteria for assessing support failure risks and enhancing the proactiveness of engineering safety assurance.

[0006] To achieve the above objectives, the present invention provides an intelligent sensing anchor bolt system, comprising a hollow anchor bolt body, a nut, a tray, a multi-physical sensing unit, a power module, an edge processing and communication module, and a data processing terminal; The nut and tray are sequentially assembled onto the exposed portion of the hollow anchor rod body; The multi-physics sensing unit includes a thermal pulse sensor, a three-dimensional stress sensing unit, and a distributed fiber optic acoustic wave sensor; the thermal pulse sensor is installed on the surface or inside the hollow anchor rod body; the three-dimensional stress sensing unit includes three three-dimensional stress sensors, which are uniformly installed circumferentially on the surface of the hollow anchor rod body; the distributed fiber optic acoustic wave sensor is installed axially along the entire length inside the hollow anchor rod body. The power module is installed inside the cavity of the hollow anchor rod body; The edge processing and communication module is connected to the multi-physical sensing unit and the power module, respectively. The data processing terminal is installed on the side wall of the tunnel or in the chamber, and is connected to the edge processing and communication module via wireless communication.

[0007] Furthermore, to effectively eliminate dependence on external power, the power module is a thermoelectric generator, which includes a semiconductor thermoelectric generator, a heat-conducting component, a heat-dissipating component, and an energy storage battery. The semiconductor thermoelectric generator has a hot end face and a cold end face and is installed inside the cavity of the hollow anchor rod body. The heat-conducting component is installed inside the cavity of the hollow anchor rod body, with one end connected to the hot end face of the semiconductor thermoelectric generator and the other end extending to the bottom of the borehole. The heat-dissipating component is installed inside the cavity of the hollow anchor rod body and makes thermal contact with the cold end face of the semiconductor thermoelectric generator. The energy storage battery is connected to the semiconductor thermoelectric generator. In this way, the natural temperature difference between the surrounding rock and the air in the tunnel can be used to achieve self-collection of energy and self-powering of the system, ensuring stability during long-term service.

[0008] Furthermore, in order to significantly increase the power generation capacity and ensure a long-term reliable power supply, the thermoelectric power generation mechanism includes multiple semiconductor thermoelectric generators connected in series, and the multiple semiconductor thermoelectric generators are distributed sequentially along the axial direction; wherein, the output power of the thermoelectric power generation mechanism is obtained according to formula (1); (1); In the formula, n is the number of thermoelectric semiconductors; To improve overall thermoelectric conversion efficiency; Let be the Seebeck coefficient of the i-th thermoelectric semiconductor; and These are the absolute temperatures of the hot and cold ends of the i-th thermoelectric semiconductor chip, respectively. Let be the internal resistance of the i-th thermoelectric semiconductor.

[0009] This invention innovatively integrates a thermal pulse sensor, a three-dimensional stress sensing unit, and a distributed fiber optic acoustic wave sensor into a hollow anchor bolt body, forming an integrated intelligent sensing anchor bolt system. This allows for active detection monitoring of the grouting process based on the thermal pulse sensor, while also enabling long-term tracking of thermal conductivity characteristics. It allows for quantitative determination of the properties and diffusion range of the grouting medium. Simultaneously, the three-dimensional stress sensing unit uses three circumferentially evenly distributed three-dimensional stress sensors to directly measure the normal contact pressure at discrete points on the bolt surface. Furthermore, it can simultaneously sense the bolt strain field and surrounding acoustic events based on the distributed fiber optic acoustic wave sensor. The distributed fiber optic acoustic wave sensor is installed axially throughout the hollow anchor bolt body, effectively isolating it from direct friction with the surrounding rock, thus ensuring continuous and high-fidelity acquisition of axial strain and acoustic emission signals. The power module is built into the internal cavity of the bolt body, facilitating power supply to various electrical components and enabling offline operation. The inclusion of edge processing and communication modules provides basic data preprocessing capabilities on the local end, reducing data transmission volume. By placing the data processing terminal on the sidewall of the tunnel or in the chamber, it can form a layered architecture with lightweight data acquisition and processing at the pole end and centralized quantitative analysis and decision-making at the terminal, in conjunction with edge processing and communication modules. This can effectively reduce the reliance on long-distance cables and the power consumption of wireless transmission, and improve the system's deployment flexibility and long-term service reliability in complex underground environments.

[0010] This anchor bolt system is highly intelligent, enabling simultaneous perception of multi-physics field information such as grouting quality, bolt stress, and surrounding rock damage. At the same time, its self-powered approach solves the technical problems of traditional monitoring systems, such as difficulties in underground wiring, heavy dependence on external power sources, and difficulty in long-term stable offline operation, effectively improving the flexibility of underground deployment and the reliability of long-term service.

[0011] This invention also provides a method for dynamic assessment of support status, employing an intelligent sensing anchor system, comprising the following steps: S1: Synchronous acquisition of multi-source data and edge preprocessing; S11: Synchronous triggering and acquisition of heterogeneous sensor array; throughout the entire lifecycle of anchor bolt installation, grouting, and long-term service, raw signals are synchronously acquired by multiple physical sensing units according to a preset strategy to obtain multi-source heterogeneous data and send it to the edge processing and communication module; the multi-source heterogeneous data includes temperature relaxation curves. Normal contact pressure and scattering phase difference data ; S12: On-orbit demodulation and feature compression at the edge; based on scattering phase difference data Obtain the quasi-static strain field after downsampling and micro-fracture event feature vector Meanwhile, based on the temperature relaxation curve Obtain the thermal pulse temperature characteristics and downsample the quasi-static strain field. Micro-fracture event feature vector Thermal pulse temperature characteristics and normal contact pressure Send to the data processing terminal; S2: Spatiotemporal fusion of multi-source information and reconstruction of physical fields; S21: Reconstruction of the spatial morphology and thermal properties of the grouting body; the data processing terminal calculates the equivalent thermal conductivity of the medium at the location of the thermal pulse sensor based on the thermal pulse temperature characteristics and the thermal conduction inversion model. and equivalent volumetric heat capacity Simultaneously, the medium type at each measuring point is identified, and the arrival time and solidification process of the grouting front are calibrated; at the same time, a three-dimensional filling model of the grouting body in the borehole is reconstructed through a spatial interpolation algorithm. S22: Reconstruction of the axial force field and lateral support stress field of the anchor bolt; combining the bolt material parameters, the quasi-static strain field... Converted into axial force distribution field And calculate the axial force gradient ; Simultaneously, based on the normal contact pressure in at least three directions on the same cross section. The equivalent two-dimensional contact stress tensor of the rod surface at this cross section is reconstructed using a semi-inverse elasticity method or fitting algorithm. ; S3: Comprehensive diagnosis and graded early warning of support system status; S31: Calculate the integrity index of the anchoring system ; axial force gradient Three-dimensional filling model of grouting body By coupling, a normalized anchorage integrity index is obtained. ; S32: Calculate the stability index of the interaction between surrounding rock and support structure By introducing entropy theory, the stress probability entropy of the principal direction distribution of contact stress and the microfracture spatiotemporal entropy of the spatial distribution density of microfracture events are calculated separately. The two are then weighted and fused to obtain the stability index of the interaction between the surrounding rock and the support, which is negatively correlated with the total entropy of the system. ; S33: Multi-level early warning triggering and decision output; automatically executes a graded response strategy based on quantitative indicators, as follows: When the anchoring system integrity index Below the set integrity threshold or local micro-fracture event rate An abnormally high level will trigger a yellow alert, along with a warning of deterioration in anchoring performance, suggesting that attention should be paid to this issue. When the stability index of the interaction between the surrounding rock and the support When a rapid decrease or abnormal axial force is detected shifting to deeper areas, an orange alert is issued, along with a warning of the development of the plastic zone in the surrounding rock and an increase in the support load. When the contact stress tensor undergoes violent rotation or micro-fracture events exhibit spatiotemporal clustering and bursts, a red alert is issued, along with warnings of macroscopic instability and prompts for immediate action.

[0012] Furthermore, to provide intuitive and quantitative decision support for regional support optimization and disaster early warning, S3 also includes: S34: Generation and visualization output of spatiotemporal cloud map of support effectiveness; The axial force distribution field obtained by inverting each intelligent sensing anchor system Anchorage integrity index Equivalent two-dimensional contact stress tensor and the stability index of the interaction between the surrounding rock and the support structure The data is then fused and visualized in the form of a spatiotemporal cloud map within the three-dimensional digital twin model of the tunnel.

[0013] As a preferred embodiment, in S11, during the process of the multi-physical sensing units synchronously acquiring the original signal according to a preset strategy, a thermal pulse sensor actively applies thermal pulses at preset time intervals and acquires the temperature relaxation curve. Simultaneously, the normal contact pressure at discrete points on the surface of the hollow anchor rod is continuously acquired at a dynamic frequency using a three-dimensional stress sensing unit. Simultaneously, the original Rayleigh scattering phase difference signal along the entire length of the pole is continuously acquired with high spatial resolution using distributed fiber optic acoustic sensors. .

[0014] Furthermore, to achieve low-power, high-efficiency downhole wireless transmission, the on-orbit demodulation and feature compression process at the edge in S12 is as follows: S121: Demodulate scattering phase difference data Obtaining the longitudinal strain field and longitudinal strain field Separation into quasi-static strain signals With dynamic acoustic event signals ; S122: For dynamic acoustic event signals Detect micro-fracture events and extract event intensity , main frequency Duration Spatial location and event timestamp Forming micro-fracture feature vectors For dynamic acoustic event signals Downsampling is performed to obtain the quasi-static strain field after downsampling. ; S123: Based on temperature relaxation curve Extracting the cooling time constant equilibrium temperature Initial temperature rise peak This creates a thermal pulse temperature characteristic; S124: Extracted micro-fracture feature vector Quasi-static strain field after downsampling Thermal pulse temperature characteristics and normal contact pressure It is encapsulated into a lightweight data frame and sent to the data processing terminal via LPWAN.

[0015] Furthermore, in order to achieve quantitative and automated monitoring of the grouting process, in S21, the process of determining the medium type at each measuring point and calibrating the arrival time of the grouting front and the solidification process is as follows: like W / (m·K) and kJ / (m 3 ·K), determine that the medium around the sensor is air, and mark the section where the grouting front has not yet reached the location of the thermal pulse sensor; like W / (m·K) and kJ / (m 3 ·K), determine that the medium around the sensor is water, and calibrate that the grouting front uses water to displace air; like and If the medium around the sensor is in the transition zone between air / water and solidified grout, and shows an upward trend over time, it is determined that the medium around the sensor is unsolidified grout. At this point, the cross-section is marked as the grouting front has been reached. like W / (m·K) and kJ / (m 3 ·K), and the parameter value tends to stabilize in multiple consecutive measurements, indicating that the medium has been completely solidified.

[0016] Furthermore, to visualize the grouting quality and facilitate rapid defect location, the process of reconstructing the three-dimensional filling model of the grouting body inside the borehole in S21 is as follows: Based on the medium type and thermophysical parameters obtained from multiple thermal pulse sensor measuring points discretely distributed along the borehole axis, a spatial interpolation algorithm is used to extend the discrete point data to the two-dimensional or three-dimensional spatial domain of the borehole. This continuously reconstructs the grout distribution range, consolidation degree, and potential cavity location of the grouting body, resulting in a three-dimensional filling model of the grouting body. .

[0017] Furthermore, in order to achieve the entropy quantification and early warning process for instability precursors, in S32, the stability index of the interaction between the surrounding rock and the support, which is negatively correlated with the total entropy of the system, is obtained. The process is as follows: S321: Calculate the stress probability entropy of the principal directions of contact stress within the time window according to formula (2). ; (2); In the formula, The principal stress direction falls into the first The probability of a directional range; S322: Calculate the micro-fracture spatiotemporal entropy of the acoustic event distribution density in the spatial grid according to formula (3). , characterizing the discrete disorder of surrounding rock fracture; (3); In the formula, For the first time window Normalized density of acoustic events in a spatial grid; S323: According to formula (4), obtain the stability index of the interaction between the surrounding rock and the support, which is negatively correlated with the total entropy of the system. ; (4); In the formula, , These are stress probability entropy. and micro-fractured spacetime entropy The weighting coefficients.

[0018] This invention provides a dynamic assessment method for support status. First, it utilizes a thermal pulse sensor, a three-dimensional stress sensor, and a distributed fiber optic acoustic sensor integrated on the anchor bolt to collect multi-source heterogeneous data, achieving comprehensive and synchronous perception of information related to the grouting solidification process, the stress state of the bolt, and the micro-fracture activity of the surrounding rock. Second, it demodulates the DAS signal at the edge end, separates quasi-static strain from dynamic acoustic events, and extracts thermal pulse temperature characteristics. Only lightweight feature frames are uploaded via a wireless network, significantly reducing data transmission volume and computational load on the data processing terminal, while ensuring real-time monitoring data and low-power operation at the edge end. Next, using the thermal pulse temperature characteristics as input, it performs inversion calculations combined with a heat conduction inversion model, accurately calculating the equivalent thermal conductivity and volumetric heat capacity of each measuring point, and effectively identifying the medium type as air / water / uncured grout / solidified body, etc. Simultaneously, it accurately calibrates the arrival time of the grouting front and the solidification process. By reconstructing a three-dimensional filling model of the grout body through spatial interpolation, the distribution range, consolidation degree, and potential cavity locations of the grout can be visually presented, providing a direct reference for subsequent assessment of anchorage integrity and overcoming the limitation of traditional methods in visualizing grouting defects. Subsequently, the axial force distribution and its gradient are obtained based on quasi-static strain field transformation, facilitating the precise location of areas with abnormal axial force concentration. Simultaneously, an equivalent two-dimensional contact stress tensor is reconstructed using contact pressures in multiple directions on the same cross-section, achieving a quantitative characterization of the lateral compression degree experienced by the rod, providing key mechanical parameters for analyzing the interaction between the surrounding rock and the support. Furthermore, the axial force distribution gradient is coupled with the grout body filling model for analysis, enabling the identification of specific depth ranges where bond slip or debonding occurs. A normalized anchorage integrity index is calculated by comprehensively considering the entire anchorage section, upgrading anchorage performance from qualitative judgment to quantitative evaluation, and providing reliable technical indicators for graded early warning. Then, entropy theory is introduced to calculate the probability entropy of the principal direction distribution of contact stress and the information entropy of the spatial distribution of micro-fracture events, respectively. A stability index negatively correlated with the total entropy of the system is obtained through weighted fusion. Based on this stability index, the precursors of system instability are quantified from the perspective of disorder, exhibiting higher sensitivity and early warning capabilities compared to traditional single-threshold criteria. Finally, a three-level early warning mechanism is established, corresponding to anchorage performance degradation, development of the surrounding rock plastic zone, and macroscopic instability precursors, with each level of warning associated with specific quantitative indicator thresholds and handling recommendations. In particular, the identification of violent rotation of the contact stress tensor and spatiotemporal clustering of micro-fractures can capture sudden instability risks that are difficult to detect using traditional methods, achieving a shift from passive monitoring to proactive early warning and providing scientific decision support for mine roadway safety management.

[0019] This method enables dynamic and accurate diagnosis and graded early warning of support status, providing advanced criteria for support failure risk and enhancing the initiative and reliability of engineering safety assurance. Attached Figure Description

[0020] Figure 1 This is a schematic diagram of the intelligent sensing anchor system in this invention; Figure 2 This is a schematic diagram showing the distribution of the edge processing and communication module and the power module in this invention; Figure 3 yes Figure 1 Sectional view along the middle AA direction; Figure 4 This is a flowchart of the evaluation method in this invention.

[0021] In the figure: 1. Hollow anchor rod body, 2. Nut, 3. Tray, 4. Power module, 5. Thermal pulse sensor, 6. Three-dimensional stress sensor, 7. Distributed fiber optic acoustic sensor, 8. Edge processing and communication module, 9. Cold air, 10. Rock mass heat, 11. Orifice section, 12. Middle sensing section, 13. Anchoring section. Detailed Implementation

[0022] The invention will now be further described with reference to the accompanying drawings.

[0023] like Figures 1 to 3 As shown, the present invention provides an intelligent sensing anchor bolt system, including a hollow anchor bolt body 1, a nut 2, a tray 3, a multi-physical sensing unit, a power module 4, an edge processing and communication module 8, and a data processing terminal; The hollow anchor rod body 1 includes an orifice section 11, a central sensing section 12, and an anchoring section 13 in sequence. This provides a clear structural carrier for the axial partitioning of multiple physical sensing units, which facilitates the installation of different components in different partitions. The nut 2 and the tray 3 are sequentially assembled onto the exposed portion of the bore section 11 of the hollow anchor rod body 1; The multi-physics sensing unit includes a thermal pulse sensor 5, a three-dimensional stress sensing unit, and a distributed fiber optic acoustic wave sensor 7. The thermal pulse sensor 5 is installed on the surface or in the inner cavity of the hollow anchor rod body 1. In one embodiment of the present invention, it is preferably installed on the surface of the anchoring section 13 to collect thermal conductivity data of the surrounding rock. In a preferred embodiment of the present invention, the thermal pulse sensor 5 is positioned at key sections of the anchoring section 13 of the hollow anchor rod 1, for example, at the front, middle, and rear ends of the anchoring section 13. As a further preferred embodiment, the miniature thermal pulse sensor 5 includes a miniature heating element and a miniature temperature sensing element, both encapsulated on the same miniature ceramic substrate and tightly bonded to the surface of the hollow anchor rod 1 using thermally conductive insulating adhesive to ensure good thermal coupling with the surrounding medium. The heating element is preferably a miniature platinum resistance thermometer or a nickel-chromium alloy thin-film resistor, and the temperature sensing element is preferably a thermistor or a fiber Bragg grating temperature sensor. The working process of this miniature thermal pulse sensor is as follows: 1. Initial thermal equilibrium. Before grouting operations are applied, the temperature sensing element continuously monitors the initial temperature of the surrounding medium. 1. Establish a thermal balance baseline. 2. Active thermal pulse excitation. After the grouting operation begins, at preset time intervals (e.g., every 30 seconds or every minute), a thermal pulse excitation is applied to the surrounding medium by the heating element for a duration of [duration missing]. constant heating power 3. Temperature relaxation curve acquisition. After the thermal pulse ends, the temperature sensing element continuously records the change of sensor surface temperature over time at a preset sampling frequency (e.g., 10 Hz) to obtain the temperature relaxation curve. And send it to the edge processing and communication module 8, where The relaxation time is calculated from the end of the self-heating pulse. Preferably, the heat pulse excitation and temperature monitoring functions of the heat pulse sensor are intelligently scheduled by the edge processing and communication module 8: during the grouting stage after anchor installation and when grout degradation is suspected, a high-frequency activation mode (e.g., once every 30 seconds) is used; during the long-term stable monitoring stage, a low-frequency inspection mode (e.g., once per hour) is switched to effectively optimize the overall energy consumption of the system, enabling the system to operate stably for a long period.

[0024] The equivalent thermal conductivity is obtained by performing equivalent thermophysical property inversion calculations based on the temperature characteristics of the thermal pulse at the data processing terminal. and equivalent volumetric heat capacity Based on equivalent thermal conductivity and equivalent volumetric heat capacity The system identifies the medium type at each measuring point and simultaneously calibrates the arrival time of the grouting front and the curing process. Therefore, by incorporating thermal pulse sensors, the intelligent sensing anchor system can perform real-time, quantitative, and non-destructive monitoring of the advancement position of the grouting front and the curing process of the grout along the entire length of the anchoring section during grouting construction and post-grouting curing stages. This provides direct evidence for grouting quality assessment and anchor bearing capacity prediction.

[0025] The three-dimensional stress sensing unit includes three three-dimensional stress sensors 6, which are uniformly installed circumferentially on the surface of the hollow anchor rod 1. They are used to synchronously sense the normal contact pressure of the surrounding rock on the rod at the installation position in order to obtain the three-dimensional stress state of the surrounding rock. In a preferred embodiment of the present invention, the three-dimensional stress sensor is a miniature pressure sensor chip, which is further preferably a silicon piezoresistive MEMS pressure-sensitive chip. The pressure-sensing diaphragms of the three miniature pressure sensors face radially outward, and the surfaces of the pressure-sensing diaphragms are flush with the outer surface of the anchor bolt body. A smooth transition between the two can be achieved through precision machining or local filling processes to avoid stress concentration or contact gaps between the pressure-sensing diaphragms and the surrounding medium. Each miniature pressure sensor chip is used to independently measure its azimuth angle. Normal contact pressure on the surface of the anchor bolt ,in The sensor chip number, For the first The angular coordinates of each sensor chip relative to the reference direction of the anchor bolt section. For time. Shallow grooves can be opened on the inner or outer surface of the hollow anchor rod body. The output signals of each micro pressure sensor chip are led out through the micro wires embedded in the shallow grooves and connected to the edge processing and communication module 8. In other embodiments of the present invention, multiple sets of three-dimensional stress sensors can be arranged at intervals along the axial direction of the anchor rod body. Each set of three-dimensional stress sensors is located at a different depth of characteristic section (e.g., the front end of the anchoring section, the boundary between the free section and the anchoring section, etc.) to obtain a more comprehensive three-dimensional stress distribution profile along the entire length of the anchor rod.

[0026] The distributed fiber optic acoustic sensor 7 is installed axially along the entire length of the hollow anchor rod 1 within its inner cavity to collect vibration and acoustic signals along the entire length of the rod. In a preferred embodiment of the invention, the distributed fiber optic acoustic sensor 7 includes a single-mode sensing fiber, which is laid axially along the inner wall of the hollow anchor rod 1 and continuously bonded to the inner wall of the rod using high-strength structural adhesive or epoxy resin to ensure slip-free strain transfer between the fiber and the anchor rod. The distributed fiber optic acoustic sensor 7 demodulates signals based on the principle of a phase-sensitive optical time-domain reflectometer (φ-OTDR) using coherent Rayleigh scattering. During monitoring, a highly coherent light pulse is injected into the sensing fiber by the demodulator, and Rayleigh backscattered light generated at various points in the fiber due to microscopic refractive index inhomogeneities is received. When axial strain occurs at a certain position of the hollow anchor rod 1 or micro-vibration occurs in the surrounding medium, the local optical path of the optical fiber at that position will change accordingly, thereby causing a phase change in the Rayleigh scattered light, and the original Rayleigh scattered phase difference signal can be obtained. By analyzing the phase difference signal at the same location between adjacent optical pulses, it is possible to simultaneously demodulate the longitudinal strain distribution along the entire length of the optical fiber and the acoustic signal generated by the fracture event in the medium surrounding the rod (such as grout or surrounding rock). ,in, These are the position coordinates along the axial direction of the anchor bolt. For time, The frequency is specified. Therefore, by setting up the distributed fiber optic acoustic sensor 7, synchronous, continuous, and distributed monitoring of the deformation field (strain field) of the rod itself and the micro-vibration field (acoustic field) of the surrounding rock fracture can be achieved in a single fiber optic deployment. This can provide comprehensive sensing information for the evaluation of the anchor bolt support effect and the early warning of the surrounding rock stability.

[0027] The power module 4 is installed in the inner cavity of the hollow anchor rod body 1, preferably in the inner cavity at the location of the borehole section 11, and is used to supply power to each electrical component. In a preferred embodiment of the present invention, the power module is a thermoelectric generator, which includes a semiconductor thermoelectric generator, a heat-conducting component, a heat-dissipating component, and an energy storage battery. The semiconductor thermoelectric generator has a hot end face and a cold end face and is installed in the inner cavity of the hollow anchor rod body 1. Preferably, it is installed in the inner cavity at the location of the orifice section 11, so that cold air 9 can be used to form a lower temperature on its cold end face. The heat-conducting component is installed in the inner cavity of the hollow anchor rod body 1, with one end connected to the hot end face of the semiconductor thermoelectric generator and the other end connected to the hot end face of the semiconductor thermoelectric generator. The end extends to the location of the anchoring section 13; thus, the heat 10 of the deep rock mass can be transferred to the hot end face of the semiconductor thermoelectric generator using the heat-conducting component, so as to form a large temperature difference between the hot and cold end faces; the heat dissipation component is installed in the inner cavity of the hollow anchor rod 1 and makes thermal contact with the cold end face of the semiconductor thermoelectric generator; in this way, the heat generated by the cold end face of the semiconductor thermoelectric generator can be quickly dissipated to the external environment using the excellent heat dissipation performance of the heat dissipation component, so as to ensure power generation efficiency; the energy storage battery is connected to the semiconductor thermoelectric generator. This thermoelectric generator mechanism conducts geothermal heat from the bottom of the borehole to the hot end face of the semiconductor thermoelectric generator through the heat-conducting component, and generates electrical energy using the natural temperature difference between the inner cavity of the rod and the bottom of the borehole, which is stored in the energy storage battery. It can provide a continuous and stable self-powered energy source for the anchor system, completely eliminating the dependence on external power sources and underground wiring, effectively solving the technical pain points of traditional monitoring systems that are difficult to power and difficult to operate offline for a long time, and ensuring the energy self-sufficiency and long-term service reliability of the system in harsh underground environments.

[0028] As a preferred option, the electrical energy generated by the thermoelectric power generation module is primarily used to maintain the operation of the distributed fiber optic acoustic sensor 7, and secondarily to power the edge processing and communication module 8.

[0029] As another preferred embodiment, the thermoelectric power generation mechanism includes multiple semiconductor thermoelectric generators connected in series, and the multiple semiconductor thermoelectric generators are distributed sequentially along the axial direction. This allows the temperature difference between the hot and cold ends of the multiple semiconductor thermoelectric generators to be matched along the axial attenuation gradient, which can maximize the utilization of the limited temperature difference, significantly improve the power generation, and thus achieve optimal energy harvesting efficiency.

[0030] The output power of the thermoelectric generator is obtained according to formula (1); (1); In the formula, n is the number of thermoelectric semiconductors; To consider the overall thermoelectric conversion efficiency (including circuit losses); Let V be the Seebeck coefficient of the i-th thermoelectric semiconductor, expressed in V / K. and These are the absolute temperatures of the hot and cold ends of the i-th thermoelectric semiconductor chip, respectively, in K. Let be the internal resistance of the i-th thermoelectric semiconductor, in Ω.

[0031] The edge processing and communication module 8 is connected to the multi-physical sensing unit and the power module 4 respectively. It is responsible for data processing and analysis, and at the same time, it is used to interact and communicate with external terminals through wireless communication. In a preferred embodiment of the present invention, the edge processing and communication module 8 is disposed within the tail cavity of the anchor bolt body; alternatively, it can be externally mounted on the tray 3. The edge processing and communication module 8 includes a microcontroller, a signal conditioning and acquisition circuit, a wireless communication module, and an intelligent processing module embedded in the microcontroller. The microcontroller unit is preferably a low-power processor based on an ARM Cortex-M4 or M33 core, and the wireless communication module is preferably a low-power wide-area network communication module such as LoRa, NB-IoT, or ZigBee. The working process of the edge processing and communication module 8 is as follows: 1. Receive the temperature relaxation curve from the thermal pulse sensor 5. Receive raw Rayleigh scattering phase difference data from distributed fiber optic acoustic sensor 7. Where z is the spatial coordinate along the anchor rod axis. This is the sampling time. Simultaneously, the edge processing and communication module 8 applies normal contact pressure to each channel. Synchronous sampling is performed and transmitted to the data processing terminal. The sampling frequency is preferably 10Hz to 100Hz to capture the dynamic process of surrounding rock pressure changes. 2. On-orbit demodulation and signal separation. Based on temperature relaxation curves. Extract the thermal pulse temperature characteristics, including the cooling time constant. Stable temperature Initial temperature rise peak ; For the original scattering phase difference data Real-time demodulation is performed to convert it into a longitudinal strain field along the entire length of the optical fiber. Subsequently, digital filtering technology was used to... It is separated into two signal components: one signal component is the quasi-static strain signal. It is obtained through low-pass filtering, and its cutoff frequency is preferably between 0.1Hz and 1Hz, used to reflect the strain accumulation of the anchor rod body caused by slow effects such as creep of the surrounding rock and temperature changes; the other signal component is the dynamic acoustic event signal. It is obtained through bandpass filtering, with a preferred passband frequency of 10 Hz to 10 kHz, and is used to reflect transient elastic waves generated by micro-fractures and crack propagation in the medium surrounding the rod. Simultaneously, it is used for dynamic acoustic event signals. Event detection is performed, and a valid acoustic event is determined when the signal amplitude exceeds a preset trigger threshold. For each detected valid acoustic event, the following feature parameters are extracted in the time and frequency domains: event intensity. It can be characterized by the peak amplitude or effective value (RMS) of the signal envelope; dominant frequency distribution. The duration can be characterized by the peak frequency and -3dB bandwidth of the event signal power spectral density. It can be characterized by the duration during which the signal envelope exceeds the trigger threshold; event space location. It can be determined by the spatial coordinate z corresponding to the peak signal at the time of the event; and the real-time timestamp of the event. 3. Lightweight wireless transmission. Only the extracted micro-fracture event feature vectors are used. Quasi-static strain profile data extracted at preset time intervals (e.g., once per hour). Thermal pulse temperature characteristics and normal contact pressure The data is encapsulated into lightweight data frames and transmitted to the data processing terminal or remote monitoring center via a wireless communication module. This avoids transmitting the original dynamic waveform data, instead storing it locally in a buffer or through cyclic overlay, effectively reducing the amount of data transmitted and ensuring reliable data transmission. It also significantly reduces the occupancy of downhole communication channels. Through this edge computing architecture, the intelligent sensing anchor bolt system can complete computationally intensive tasks such as data demodulation, signal separation, and feature extraction at the local edge, while transmitting only highly compressed feature information to the data processing terminal via wireless communication. This significantly reduces the transmit power consumption and duty cycle of the wireless communication module, extending the system's continuous operating time under limited energy storage conditions, while simultaneously meeting the structural health monitoring requirements for real-time early warning of critical events.

[0032] The data processing terminal is installed on the side wall of the tunnel or in the chamber and is connected to the edge processing and communication module 8 via wireless communication. Preferably, the data processing terminal is connected to the edge processing and communication module 8 via a low-power wireless network (such as LoRa or ZigBee) to perform multi-source information fusion and dynamic assessment of support status.

[0033] In one embodiment of the present invention, a heat conduction inversion model is preset in the data processing terminal. For the temperature characteristics of a heat pulse, the heat conduction inversion model is based on the unsteady-state heat transfer analytical solution or its numerical approximation form of a point heat source in an infinite medium. It uses a least-squares fitting algorithm to simultaneously invert and calculate the equivalent thermal conductivity of the medium at the sensor location. and equivalent volumetric heat capacity Subsequently, the medium type was identified and the grouting status was calibrated. Based on the identification and calibration results, a three-dimensional filling model of the grouting body inside the borehole was reconstructed using a spatial interpolation algorithm. Simultaneously, the data processing terminal acquires normal contact pressure values ​​at at least three different azimuth angles on the same cross-section. Furthermore, the equivalent two-dimensional contact stress tensor of the rod surface at this cross-section is reconstructed using a semi-inverse elasticity method or a fitting algorithm. Finally, a comprehensive diagnosis and graded early warning of the support system status are conducted.

[0034] This invention innovatively integrates a thermal pulse sensor, a three-dimensional stress sensing unit, and a distributed fiber optic acoustic wave sensor into a hollow anchor bolt body, forming an integrated intelligent sensing anchor bolt system. This allows for active detection monitoring of the grouting process based on the thermal pulse sensor, while also enabling long-term tracking of thermal conductivity characteristics. It allows for quantitative determination of the properties and diffusion range of the grouting medium. Simultaneously, the three-dimensional stress sensing unit uses three circumferentially evenly distributed three-dimensional stress sensors to directly measure the normal contact pressure at discrete points on the bolt surface. Furthermore, it can simultaneously sense the bolt strain field and surrounding acoustic events based on the distributed fiber optic acoustic wave sensor. The distributed fiber optic acoustic wave sensor is installed axially throughout the hollow anchor bolt body, effectively isolating it from direct friction with the surrounding rock, thus ensuring continuous and high-fidelity acquisition of axial strain and acoustic emission signals. The power module is built into the internal cavity of the bolt body, facilitating power supply to various electrical components and enabling offline operation. The inclusion of edge processing and communication modules provides basic data preprocessing capabilities on the local end, reducing data transmission volume. By placing the data processing terminal on the sidewall of the tunnel or in the chamber, it can form a layered architecture with lightweight data acquisition and processing at the pole end and centralized quantitative analysis and decision-making at the terminal, in conjunction with edge processing and communication modules. This can effectively reduce the reliance on long-distance cables and the power consumption of wireless transmission, and improve the system's deployment flexibility and long-term service reliability in complex underground environments.

[0035] This anchor bolt system is highly intelligent, enabling simultaneous perception of multi-physics field information such as grouting quality, bolt stress, and surrounding rock damage. At the same time, its self-powered approach solves the technical problems of traditional monitoring systems, such as difficulties in underground wiring, heavy dependence on external power sources, and difficulty in long-term stable offline operation, effectively improving the flexibility of underground deployment and the reliability of long-term service.

[0036] like Figure 4 As shown, the present invention also provides a method for dynamic assessment of support status, employing an intelligent sensing anchor system, comprising the following steps: S1: Synchronous acquisition of multi-source data and edge preprocessing; S11: Synchronous triggering and acquisition of heterogeneous sensor array; throughout the entire lifecycle of anchor bolt installation, grouting, and long-term service, raw signals are synchronously acquired by multiple physical sensing units according to a preset strategy (according to a preset period or based on a trigger event), obtaining multi-source heterogeneous data and sending it to the edge processing and communication module 8; the multi-source heterogeneous data includes temperature relaxation curves. Normal contact pressure and scattering phase difference data ; During the process of synchronously acquiring raw signals by multiple physical sensing units according to a preset strategy, thermal pulses can be actively applied and temperature relaxation curves can be acquired by thermal pulse sensor 5 at preset time intervals (e.g., every 30 seconds during grouting and every hour during service). Simultaneously, the normal contact pressure at discrete points on the surface of the hollow anchor rod 1 is continuously acquired at a dynamic frequency (e.g., 10–100 Hz) using a three-dimensional stress sensing unit. Simultaneously, the distributed fiber optic acoustic wave sensor 7 continuously acquires the original Rayleigh scattering phase difference signal along the entire length of the pole with high spatial resolution. .

[0037] S12: On-orbit demodulation and feature compression at the edge; at edge processing and communication module 8, based on scattering phase difference data. Obtain the quasi-static strain field after downsampling and micro-fracture event feature vector , ,in, For the intensity of the event, Main frequency, For duration, For the event space location, This is the timestamp of the event, and also based on the temperature relaxation curve. Obtain the thermal pulse temperature characteristics, including the cooling time constant. Stable temperature Initial temperature rise peak and the downsampled quasi-static strain field Micro-fracture event feature vector Thermal pulse temperature characteristics and normal contact pressure Send to the data processing terminal; Specifically, the on-orbit demodulation and feature compression process at the edge is as follows: S121: Demodulate scattering phase difference data Obtaining the longitudinal strain field and longitudinal strain field Separation into quasi-static strain signals With dynamic acoustic event signals ; S122: For dynamic acoustic event signals Detect micro-fracture events and extract event intensity , main frequency Duration Spatial location and event timestamp Forming micro-fracture feature vectors For dynamic acoustic event signals Downsampling is performed to obtain the quasi-static strain field after downsampling. ; S123: Based on temperature relaxation curve Extracting the cooling time constant equilibrium temperature Initial temperature rise peak This process generates thermal pulse temperature characteristics. During this process, only the characteristics are extracted, without performing a complete inversion of thermal properties. This effectively reduces the computational load on the edge and also helps to reduce the computing power requirements of the edge modules, thereby reducing the manufacturing cost of a single intelligent sensing anchor system and facilitating its wider application. S124: Extracted micro-fracture feature vector Quasi-static strain field after downsampling Thermal pulse temperature characteristics and normal contact pressure It is encapsulated into lightweight data frames and sent to the data processing terminal via LPWAN (such as LoRa / NB-IoT).

[0038] In this technical solution, the original scattering phase difference data is decoupled into a quasi-static strain field and a dynamic acoustic event signal. Key parameters of the micro-fracture event are extracted from these signals, and the quasi-static strain is downsampled. Simultaneously, the thermal pulse temperature characteristics and normal contact pressure are combined, and the data is encapsulated into a lightweight data frame before being transmitted via LPWAN. This process completes core feature extraction and data compression locally, significantly reducing the amount of data and power consumption during wireless transmission, while retaining key feature information for subsequent physical field reconstruction and diagnosis. It effectively adapts to the stringent requirements of low-power, high-real-time data transmission in complex downhole environments.

[0039] S2: Spatiotemporal fusion of multi-source information and reconstruction of physical fields; S21: Reconstruction of the spatial morphology and thermal properties of the grouting body; at the data processing terminal, using the thermal pulse temperature characteristics as input data, the equivalent thermal conductivity of the medium at the location of the thermal pulse sensor 5 is calculated through the built-in thermal conduction inversion model. and equivalent volumetric heat capacity When some data of the thermal pulse temperature characteristics is missing, the original temperature relaxation curve can be requested from the edge processing and communication module 8. The thermal pulse temperature features are then re-extracted and used as input data for inversion calculation to ensure the accuracy of the calculation. Simultaneously, the medium type at each measuring point was identified, and the arrival time and solidification process of the grouting front were calibrated. The medium type included air / water / uncured grout / solidified body, etc. Furthermore, a three-dimensional filling model of the grout within the borehole was reconstructed using a spatial interpolation algorithm. As a preferred option, the spatial interpolation algorithm is preferably radial basis function interpolation or kriging interpolation. In the embodiments of the present invention, kriging interpolation is preferred. In a specific embodiment of the present invention, the process of determining the medium type (air / water / uncured grout / solidified body) at each measuring point and calibrating the arrival time of the grouting front and the solidification process is as follows: like W / (m·K) and kJ / (m 3 ·K), determine that the medium around the sensor is air, and mark the section where the grouting front has not yet reached the location of the thermal pulse sensor 5; like W / (m·K) and kJ / (m 3 ·K), determine that the medium around the sensor is water, and calibrate that the grouting front uses water to displace air; like and If the medium around the sensor is in the transition zone between air / water and solidified grout, and shows an upward trend over time, it is determined that the medium around the sensor is unsolidified grout. At this point, the cross-section is marked as the grouting front has been reached. like W / (m·K) and kJ / (m 3 ·K), and the parameter value tends to stabilize in multiple consecutive measurements, indicating that the medium has been completely solidified.

[0040] By pre-calibrated equivalent thermal conductivity With volumetric heat capacity The system automatically identifies the type of medium surrounding the sensor—whether it be air, water, uncured grout, or solidified material—within a defined threshold range. Based on this identification, it accurately calibrates the arrival time of the grouting front and the solidification process of the grout. This process cleverly transforms the qualitative judgment of grouting quality into quantitative automatic identification based on thermophysical parameters. It eliminates the need for human experience in judgment and calibration, avoiding errors caused by subjective factors. This enables seamless and continuous monitoring of the grouting process, providing reliable temporal and spatial information for anchorage integrity analysis and evaluation.

[0041] As a preferred option, the process of reconstructing the three-dimensional filling model of the grouting body inside the borehole is as follows: Based on the medium type and thermophysical parameters obtained from five measuring points of multiple thermal pulse sensors discretely distributed along the borehole axis, a spatial interpolation algorithm is used to extend the discrete point data to the two-dimensional or three-dimensional spatial domain of the borehole. This continuously reconstructs the grout distribution range, consolidation degree, and potential cavity location of the grouting body, resulting in a three-dimensional filling model of the grouting body. This process can transform point information from a limited number of measuring points into field information of the grout body, realizing the visualization and quantification of the grout filling morphology. It can intuitively identify poorly consolidated areas or cavity defects, and can provide spatial distribution basis for anchorage integrity assessment.

[0042] S22: Reconstruction of the anchor bolt axial force field and the transverse support stress field; combined with bolt material parameters (including elastic modulus) Cross-sectional area ), quasi-static strain field Converted into axial force distribution field , And calculate the axial force gradient ; Simultaneously, based on the normal contact pressure in at least three directions on the same cross section. The equivalent two-dimensional contact stress tensor of the rod surface at this cross section is reconstructed using a semi-inverse elasticity method or fitting algorithm. Among them, the equivalent two-dimensional contact stress tensor Including maximum principal stress minimum principal stress and main direction In the transverse compressive stress field of the anchor bolt, they reflect the magnitude of the stress in the two principal directions of the surrounding rock pressure on the bolt surface, and are used to quantify the degree of transverse compression of the anchor bolt by the surrounding rock and determine the direction of the principal stress.

[0043] S3: Comprehensive diagnosis and graded early warning of support system status; establishing integrated evaluation indicators to grade and assess the health status of the support system, specifically: S31: Calculate the integrity index of the anchoring system ; axial force gradient Three-dimensional filling model of grouting body By coupling, a normalized anchorage integrity index is obtained. ; If the axial force decreases gradually within a certain depth range And this interval corresponds to the three-dimensional filling model of the grouting body. If a poorly consolidated area or a void is found in the anchorage, it is determined that bond slip or debonding has occurred at that interface. Based on the overall situation of the entire anchorage section, a normalized anchorage integrity index is calculated. Anchorage integrity index It can be used to evaluate the bonding state of multiple interfaces of anchoring agent, rod, grout and rock mass; S32: Calculate the stability index of the interaction between surrounding rock and support structure Combining the evolution of contact stress tensor with microfracture activity, entropy theory is introduced to quantify the disorder of the system, specifically: By introducing entropy theory, the stress probability entropy of the principal direction distribution of contact stress is calculated to effectively characterize the deflection disorder of the load transfer path, and the micro-fracture spatiotemporal entropy of the spatial distribution density of micro-fracture events is calculated to effectively characterize the discrete disorder of surrounding rock fracture. The two are then weighted and fused to obtain the stability index of the interaction between surrounding rock and support, which is negatively correlated with the total entropy of the system. The specific process is as follows: S321: Calculate the stress probability entropy of the principal directions of contact stress within the time window according to formula (2). This characterizes the degree of deflection disorder in the load transfer path; (2); In the formula, The principal stress direction falls within the historical time window of the first The probability of a directional range; S322: Calculate the micro-fracture spatiotemporal entropy of the acoustic event distribution density in the spatial grid according to formula (3). , characterizing the discrete disorder of surrounding rock fracture; (3); In the formula, For the first time window Normalized density of acoustic events in a spatial lattice; S323: According to formula (4), obtain the stability index of the interaction between the surrounding rock and the support, which is negatively correlated with the total entropy of the system. ; (4); In the formula, , These are stress probability entropy. and micro-fractured spacetime entropy The weighting coefficients.

[0044] A sharp increase in entropy indicates the system's evolution towards instability; the stability index of the interaction between the surrounding rock and the support. A value close to 1 indicates that the system is currently in a stable state, while a value close to 0 indicates that the system is currently in an unstable state.

[0045] By calculating the stress probability entropy of the principal directions of contact stress distribution. To quantify the deflection disorder of the load transfer path, and simultaneously calculate the spatiotemporal entropy of the spatial distribution density of micro-fracture events. This is used to characterize the discrete disorder of surrounding rock fracture, and the two are weighted and fused to obtain the stability index of the interaction between surrounding rock and support, which is negatively correlated with the total entropy of the system. This process quantitatively characterizes the degree of evolution of the support system from order to disorder from an information theory perspective, transforming the traditionally difficult-to-quantify instability precursors into calculable entropy indicators. This allows for the effective and sensitive capture of early abnormal signals such as principal stress direction deflection and micro-fracture spatial accumulation, providing scientific and accurate criteria for graded early warning.

[0046] S33: Multi-level early warning triggering and decision output; automatically executes a graded response strategy based on quantitative indicators, as follows: When the anchoring system integrity index Below the set integrity threshold or local micro-fracture event rate An abnormally high level will trigger a yellow alert, along with a warning of deterioration in anchoring performance, suggesting that attention should be paid to this issue. When the stability index of the interaction between the surrounding rock and the support When a rapid decrease or abnormal axial force is detected shifting to deeper areas, an orange alert is issued, along with a warning of the development of the plastic zone in the surrounding rock and an increase in the support load. When the contact stress tensor undergoes a violent rotation ( When an event exceeding the limit or a micro-rupture event exhibits a spatiotemporal clustering and eruption, a red alert will be issued, along with warnings of macroscopic instability and prompts for immediate action.

[0047] In another embodiment of the present invention, in order to provide intuitive and quantitative decision support for regional support optimization and disaster early warning, based on S31 to S33 in S3, the following is also included: S34: Generation and visualization output of spatiotemporal cloud map of support effectiveness; The axial force distribution field obtained by inverting each intelligent sensing anchor system Anchorage integrity index Equivalent two-dimensional contact stress tensor and the stability index of the interaction between the surrounding rock and the support structure The data is then fused and visualized in the 3D digital twin model of the tunnel as a spatiotemporal cloud map. For ease of viewing, the spatiotemporal cloud map supports playback along the timeline and slice analysis by spatial location, thus providing intuitive and quantitative decision-making support for regional support scheme optimization and disaster early warning.

[0048] Traditional monitoring methods typically rely on external power supply and wired data transmission, which are difficult to implement in harsh roadway environments, suffer from low reliability, and are unsuitable for long-term operation. Sensing units are often limited to localized point measurements, failing to acquire multi-dimensional information such as the stress distribution along the entire length of the anchor bolt, the spatial diffusion of the grout, and its solidification state. Furthermore, the monitoring function is limited, lacking comprehensive perception and diagnostic capabilities for the overall mechanical behavior and interface damage processes of the bolt-grout-surrounding rock system. To address these shortcomings, this invention proposes an intelligent sensing anchor bolt system and a dynamic assessment method for support status. Its core lies in achieving energy self-sufficiency through a built-in thermoelectric power generation module utilizing environmental temperature differences. It integrates thermal pulses, distributed optical fibers, and three-dimensional stress sensors to achieve synchronous in-situ sensing of grouting quality, bolt strain field, and surrounding rock contact stress. Furthermore, relying on wireless transmission and information fusion algorithms, it constructs an integrated solution that is self-powered, wireless, multi-dimensionally sensing, and intelligently assessing, thereby achieving full-cycle, refined dynamic monitoring and safety early warning of the anchor bolt support status from construction to long-term service. First, multi-source heterogeneous data was collected using thermal pulse sensors, three-dimensional stress sensors, and distributed fiber optic acoustic sensors integrated on the anchor bolt, enabling comprehensive and synchronous sensing of information related to the grouting solidification process, the stress state of the bolt, and the micro-fracture activity of the surrounding rock. Second, the DAS signal was demodulated at the edge, separating quasi-static strain from dynamic acoustic events, and thermal pulse temperature characteristics were extracted. Only lightweight feature frames were uploaded via a wireless network, significantly reducing data transmission volume and computational load on the data processing terminal, while ensuring real-time monitoring and low-power operation at the edge. Third, using the thermal pulse temperature characteristics as input, inversion calculations were performed using a heat conduction inversion model. This accurately calculated the equivalent thermal conductivity and volumetric heat capacity of each measuring point and effectively identified the medium type as air / water / uncured grout / solidified body, etc. Simultaneously, the arrival time of the grouting front and the solidification process could be accurately determined. By reconstructing a three-dimensional filling model of the grout body through spatial interpolation, the distribution range, consolidation degree, and potential cavity locations of the grout can be visually presented, providing a direct reference for subsequent assessment of anchorage integrity and overcoming the limitation of traditional methods in visualizing grouting defects. Subsequently, the axial force distribution and its gradient are obtained based on quasi-static strain field transformation, facilitating the precise location of areas with abnormal axial force concentration. Simultaneously, an equivalent two-dimensional contact stress tensor is reconstructed using contact pressures in multiple directions on the same cross-section, achieving a quantitative characterization of the lateral compression degree experienced by the rod, providing key mechanical parameters for analyzing the interaction between the surrounding rock and the support. Furthermore, the axial force distribution gradient is coupled with the grout body filling model for analysis, enabling the identification of specific depth ranges where bond slip or debonding occurs. A normalized anchorage integrity index is calculated by comprehensively considering the entire anchorage section, upgrading anchorage performance from qualitative judgment to quantitative evaluation, and providing reliable technical indicators for graded early warning. Then, entropy theory is introduced to calculate the probability entropy of the principal direction distribution of contact stress and the information entropy of the spatial distribution of micro-fracture events, respectively. A stability index negatively correlated with the total entropy of the system is obtained through weighted fusion.Based on this stability index, the precursors of system instability are quantified from the perspective of disorder, exhibiting higher sensitivity and early warning capabilities compared to traditional single-threshold criteria. Finally, a three-level early warning mechanism is established, corresponding to anchorage performance degradation, development of the surrounding rock plastic zone, and macroscopic instability precursors, with each level of warning associated with specific quantitative indicator thresholds and handling recommendations. In particular, the identification of violent rotation of the contact stress tensor and spatiotemporal clustering of micro-fractures can capture sudden instability risks that are difficult to detect using traditional methods, achieving a shift from passive monitoring to proactive early warning and providing scientific decision support for mine roadway safety management.

[0049] This method enables dynamic and accurate diagnosis and graded early warning of support status, providing advanced criteria for support failure risk and enhancing the initiative and reliability of engineering safety assurance.

Claims

1. An intelligent sensing anchor bolt system, comprising a hollow anchor bolt body, a nut, and a tray, characterized in that, It also includes a multi-physical sensing unit, a power module, an edge processing and communication module, and a data processing terminal; The nut and tray are sequentially assembled onto the exposed portion of the hollow anchor rod body; The multi-physics sensing unit includes a thermal pulse sensor, a three-dimensional stress sensing unit, and a distributed fiber optic acoustic wave sensor; the thermal pulse sensor is installed on the surface or inside the hollow anchor rod body; the three-dimensional stress sensing unit includes three three-dimensional stress sensors, which are uniformly installed circumferentially on the surface of the hollow anchor rod body; the distributed fiber optic acoustic wave sensor is installed axially along the entire length inside the hollow anchor rod body. The power module is installed inside the cavity of the hollow anchor rod body; The edge processing and communication module is connected to the multi-physical sensing unit and the power module, respectively. The data processing terminal is installed on the side wall of the tunnel or in the chamber, and is connected to the edge processing and communication module via wireless communication.

2. The intelligent sensing anchor system according to claim 1, characterized in that, The power module is a thermoelectric generator, which includes a semiconductor thermoelectric generator, a heat-conducting component, a heat dissipation component, and an energy storage battery. The semiconductor thermoelectric generator is provided with a hot end face and a cold end face, and is installed in the inner cavity of the hollow anchor rod body; The heat-conducting component is installed in the inner cavity of the hollow anchor rod body, with one end connected to the hot end face of the semiconductor thermoelectric generator and the other end extending to the bottom of the borehole. The heat dissipation component is installed in the inner cavity of the hollow anchor rod body and makes thermal contact with the cold end face of the semiconductor thermoelectric generator. The energy storage battery is connected to a semiconductor thermoelectric generator.

3. The intelligent sensing anchor system according to claim 2, characterized in that, The thermoelectric power generation mechanism includes multiple semiconductor thermoelectric generators connected in series, and the multiple semiconductor thermoelectric generators are distributed sequentially along the axial direction; wherein, the output power of the thermoelectric power generation mechanism is obtained according to formula (1); （1）； In the formula, n is the number of thermoelectric semiconductors; To improve overall thermoelectric conversion efficiency; Let be the Seebeck coefficient of the i-th thermoelectric semiconductor; and These are the absolute temperatures of the hot and cold ends of the i-th thermoelectric semiconductor chip, respectively. Let be the internal resistance of the i-th thermoelectric semiconductor.

4. A method for dynamic assessment of support status, employing an intelligent sensing anchor system as described in any one of claims 1 to 3, characterized in that, Includes the following steps: S1: Synchronous acquisition of multi-source data and edge preprocessing; S11: Throughout the entire lifecycle of anchor installation, grouting, and long-term service, raw signals are synchronously acquired by multi-physical sensing units according to a preset strategy to obtain multi-source heterogeneous data, which is then sent to the edge processing and communication module; the multi-source heterogeneous data includes temperature relaxation curves. Normal contact pressure and scattering phase difference data ; S12: Based on scattering phase difference data Obtain the quasi-static strain field after downsampling and micro-fracture event feature vector Meanwhile, based on the temperature relaxation curve The temperature characteristics of the thermal pulse are obtained and sent to the data processing terminal; S2: Spatiotemporal fusion of multi-source information and reconstruction of physical fields; S21: Based on the temperature characteristics of the thermal pulse, the equivalent thermal conductivity of the medium at the location of the thermal pulse sensor is calculated using a heat conduction inversion model. and equivalent volumetric heat capacity Simultaneously, the medium type at each measuring point is identified, and the arrival time and solidification process of the grouting front are calibrated; at the same time, the three-dimensional filling model of the grouting body in the borehole is reconstructed. S22: Combining the rod material parameters, the quasi-static strain field Converted into axial force distribution field And calculate the axial force gradient ; Simultaneously, based on the normal contact pressure in at least three directions on the same cross section. The equivalent two-dimensional contact stress tensor of the rod surface at this cross section is reconstructed using a semi-inverse elasticity method or fitting algorithm. ; S3: Comprehensive diagnosis and graded early warning of support system status; S31: Axial force gradient Three-dimensional filling model of grouting body By coupling, a normalized anchorage integrity index is obtained. ; S32: Calculate the stress probability entropy of the principal direction distribution of contact stress and the micro-fracture spatiotemporal entropy of the spatial distribution density of micro-fracture events, and then weight and fuse the two to obtain the stability index of the interaction between the surrounding rock and the support. ; S33: Automatically execute a tiered response strategy based on quantitative indicators, as follows: When the anchoring system integrity index Below the set integrity threshold or local micro-fracture event rate An abnormally high level triggers a yellow alert, indicating a degradation in anchoring performance and suggesting continued monitoring; this is when the stability index of the interaction between the surrounding rock and the support... When a rapid decrease or abnormal axial force is detected shifting to deeper areas, an orange alert is issued, along with a warning of the development of the plastic zone in the surrounding rock and an increase in the support load. When the contact stress tensor undergoes violent rotation or micro-fracture events exhibit spatiotemporal clustering and bursts, a red alert is issued, along with warnings of macroscopic instability and prompts for immediate action.

5. The method for dynamic assessment of support status according to claim 4, characterized in that, S3 also includes: S34: Generation and visualization output of spatiotemporal cloud map of support effectiveness; The axial force distribution field obtained by inverting each intelligent sensing anchor system Anchorage integrity index Equivalent two-dimensional contact stress tensor and the stability index of the interaction between the surrounding rock and the support structure The data is then fused and visualized in the form of a spatiotemporal cloud map within the three-dimensional digital twin model of the tunnel.

6. The method for dynamic assessment of support status according to claim 4, characterized in that, In S11, a thermal pulse is actively applied at preset time intervals using a thermal pulse sensor, and temperature relaxation curves are collected. Simultaneously, the normal contact pressure at discrete points on the surface of the hollow anchor rod is continuously acquired at a dynamic frequency using a three-dimensional stress sensing unit. Simultaneously, the original Rayleigh scattering phase difference signal along the entire length of the pole is continuously acquired with high spatial resolution using distributed fiber optic acoustic sensors. .

7. The method for dynamic assessment of support status according to claim 4, characterized in that, In S12, the on-orbit demodulation and feature compression process at the edge is as follows: S121: Demodulate scattering phase difference data Obtaining the longitudinal strain field and longitudinal strain field Separation into quasi-static strain signals With dynamic acoustic event signals ; S122: For dynamic acoustic event signals Detect microfracture events and extract event intensity , main frequency Duration Spatial location and event timestamp Forming micro-fracture feature vectors For dynamic acoustic event signals Downsampling is performed to obtain the quasi-static strain field after downsampling. ; S123: Based on temperature relaxation curve Extracting the cooling time constant Equilibrium temperature Initial temperature rise peak This creates a thermal pulse temperature characteristic; S124: Extracted micro-fracture feature vector Quasi-static strain field after downsampling Thermal pulse temperature characteristics and normal contact pressure It is encapsulated into a lightweight data frame and sent to the data processing terminal via LPWAN.

8. The method for dynamic assessment of support status according to claim 4, characterized in that, In S21, the process of identifying the medium type at each measuring point and calibrating the arrival time of the grouting front and the curing process is as follows: like W / (m·K) and kJ / (m 3 ·K), determine that the medium around the sensor is air, and mark the section where the grouting front has not yet reached the location of the thermal pulse sensor; like W / (m·K) and kJ / (m 3 ·K), determine that the medium around the sensor is water, and calibrate that the grouting front uses water to displace air; like and If the medium around the sensor is in the transition zone between air / water and solidified grout, and shows an upward trend over time, it is determined that the medium around the sensor is unsolidified grout. At this point, the cross-section is marked as the grouting front has been reached. like W / (m·K) and kJ / (m 3 ·K), and the parameter value tends to stabilize in multiple consecutive measurements, indicating that the medium has been completely solidified.

9. The method for dynamic assessment of support status according to claim 8, characterized in that, In S21, the process of reconstructing the three-dimensional filling model of the grouting body inside the borehole is as follows: Based on the medium type and thermophysical parameters obtained from multiple thermal pulse sensor measuring points discretely distributed along the borehole axis, a spatial interpolation algorithm is used to extend the discrete point data to the two-dimensional or three-dimensional spatial domain of the borehole. This continuously reconstructs the grout distribution range, consolidation degree, and potential cavity location of the grouting body, resulting in a three-dimensional filling model of the grouting body. .

10. The method for dynamic assessment of support status according to claim 4, characterized in that, In S32, the stability index of the interaction between the surrounding rock and the support is obtained. The process is as follows: S321: Calculate the stress probability entropy of the principal directions of contact stress within the time window according to formula (2). ; （2）； In the formula, The principal stress direction falls into the first The probability of a directional range; S322: Calculate the micro-fracture spatiotemporal entropy of the acoustic event distribution density in the spatial grid according to formula (3). , characterizing the discrete disorder of surrounding rock fracture; （3）； In the formula, For the first time window Normalized density of acoustic events in a spatial grid; S323: According to formula (4), obtain the stability index of the interaction between the surrounding rock and the support, which is negatively correlated with the total entropy of the system. ; （4）； In the formula, , These are stress probability entropy. and micro-fractured spacetime entropy The weighting coefficients.

Images