TBM construction risk real-time early warning method based on multi-information fusion

Abstract

The invention discloses a TBM construction risk real-time early warning method based on multi-information fusion. The method comprises the steps of S1, geological forecast detection, S2, surrounding rock integrity and surrounding rock strength detection in the excavation process, S3, construction jamming risk evaluation in the excavation process, S4, geological interpretation result correction and S5, stratum risk early warning in front of equipment construction. The method mainly focuses on TBM jamming caused by collapse of the fault fracture zone, the relationship between the collapse body height and the cutterhead torque is established according to the soil pressure theory, evaluation of the collapse body height is converted into monitoring of the cutterhead idling torque, and evaluation of the collapse body height is achieved; the landslide distribution range is obtained through advanced geological forecast data, and advanced forecast of the TBM construction risk is achieved in combination with a theoretical calculation method; for the advanced forecasting result, combining slag piece parameters and tunneling parameter identification in the TBM tunneling approaching collapse body process, and dynamically updating a geological forecasting result and a collapse result; the evaluation result of the construction risk is of great significance to selection of TBM anti-jamming and de-trapping measures, and the safety of TBM construction can be guaranteed.

Description

Technical Field

[0001] This invention relates to the field of TBM construction technology, and in particular to a construction risk early warning method. Background Technology

[0002] During the construction of deep-buried TBM tunnels, the complex geological conditions pose risks of tunnel face collapse, large deformation, and sudden water inrush. Severe incidents can lead to equipment jamming and malfunctions, significantly impacting the safety and efficiency of tunnel construction. Furthermore, the enclosed construction environment and harsh working conditions at the tunnel face make effective real-time monitoring impossible. TBMs are equipped with various information sensing terminals. Utilizing the diverse sensor data collected, geological risk monitoring and early warning are conducted through data fusion, improving the accuracy of early warnings and effectively addressing TBM construction risks while increasing construction efficiency.

[0003] The interaction between the surrounding rock and the TBM (Tunnel Boring Machine) is a significant factor leading to TBM jamming. Existing technologies include Chinese patent application number 202311243340.2, which proposes a method for predicting TBM cutterhead jamming based on the collapse height of the surrounding rock. This method utilizes advanced geological prediction to obtain the height of the plastic zone of the surrounding rock and assesses the jamming risk. Chinese patent application number 202110245177.8 proposes a method for detecting the collapse of the surrounding rock behind the TBM shield. This method calculates the collapse volume based on the theoretical excavation volume and the actual excavation profile surface, and assesses the shield jamming risk based on the collapse volume. Furthermore, Chinese patent application number 202410213640.4 proposes a method and system for predicting TBM jamming risk, which uses rock mass strength parameters to establish a cluster analysis model and obtain jamming risk analysis results. Furthermore, Chinese patent application number 202310784434.4 proposes a high-risk early warning method for TBM jamming, which achieves jamming risk warning based on advanced geological forecasting and thrust and torque threshold indicators. From the above research and analysis, the key factor affecting jamming is the load of the surrounding rock on the TBM shield and cutterhead, and rock geological parameters such as plastic zone, collapse volume, surrounding rock strength, and fracture condition have a crucial impact on jamming. Due to the high randomness of rock geological parameters, the risk assessment of TBM jamming in fault fracture zones needs to consider multi-scale influences: it is necessary to accurately assess the jamming risk of the TBM in the current strata, reasonably judge the jamming situation, and provide a decision-making basis for tunneling parameter adjustment; it is also necessary to grasp the jamming risk of the strata ahead of the tunnel face, providing a decision-making basis for the selection of advanced support measures and the implementation of anti-jamming and escape measures. Given the limited range of early warning methods in current technologies, it is worthwhile to explore a comprehensive assessment method for equipment jamming risk that utilizes monitoring parameters of the cutterhead and shield to evaluate the risk of jamming in the preceding strata, and to use advanced geological forecasting to evaluate the risk of jamming in the preceding strata. Summary of the Invention

[0004] To address the shortcomings in the aforementioned background technology, this invention proposes a real-time early warning method for TBM construction risks based on multi-information fusion, which solves the problem of how to use advanced geological forecast data to obtain the distribution range of landslide bodies and combine it with theoretical calculation methods to achieve advanced prediction of TBM construction risks.

[0005] The technical solution of this invention is implemented as follows: A real-time early warning method for TBM construction risks based on multi-information fusion, the steps of which are: S1 Geological prediction and detection: advance geological prediction is carried out at the location of the tunneling machine to predict the spatial range of the collapse body in front, and the theoretical force on the cutterhead during tunneling is calculated using the formula of the collapse body height and the friction torque of the cutterhead during idling. The theoretical force on the cutterhead is compared with the cutterhead escape torque to give the geological prediction result of whether the risk of the stratum is machine jamming.

[0006] S2 rock integrity and strength detection during excavation: Rock integrity detection is performed as follows: During tunneling machine construction, the video monitoring system monitors the muck discharge video stream of the conveyor belt, and the muck particle size is identified by image recognition method. The particle size distribution is statistically analyzed, and a relationship model between muck particle size and rock integrity is established through statistical analysis. Finally, the rock integrity of the tunnel face is identified based on muck images.

[0007] The surrounding rock strength detection process involves: acquiring the tunneling parameters of the tunneling machine during construction, establishing a comprehensive rock breaking index k using these parameters, creating a database of the comprehensive rock breaking index k and the surrounding rock strength, constructing a surrounding rock strength prediction model based on the comprehensive rock breaking index k through data statistics, and evaluating the strength of the surrounding rock under tunneling using the tunneling parameters to achieve rapid perception of the surrounding rock strength.

[0008] S3 Evaluation of the risk of equipment jamming during excavation: The integrity and strength of the surrounding rock are analyzed together to obtain a comprehensive evaluation result of the surrounding rock quality at the excavation face, and to establish the evaluation result of the risk of equipment jamming during the excavation process.

[0009] S4 Correction of geological interpretation results: Using the risk assessment results of the machine in step S3, the geological prediction results in step S1 are corrected.

[0010] S5. Risk warning of strata ahead of equipment construction: Based on the revised geological forecast results in step S4, the risk of the strata ahead is classified, and corresponding measures are taken according to the risk classification.

[0011] The calculation process of the formula for the height of the collapsed body and the friction torque of the cutterhead during idle rotation in step S1 is as follows: Under the stress state of the cutterhead when the machine is stopped, based on the Coulomb earth pressure theory, when the cutterhead is subjected to a certain degree of collapsed body, each point on the cutterhead is subjected to the passive earth pressure of the collapsed body, and the torque dT at the discrete points is... f for:

[0012] ……①

[0013] ……②

[0014] In the formula, K p is the coefficient of passive earth pressure; φ is the friction angle of the broken soil mass in the landslide body, and the internal friction angle of the crushed stone is 30°; ρ is the density of the broken soil mass kg / m 3 ; g is the acceleration of gravity m / s 2 ; y is the vertical height from the discrete point to the center of the cutter head, is the distance from the discrete point to the center of the circle, and dxdy is the integral of the area;

[0015] Integrating formula ②, we get:

[0016] ……③

[0017] The integration domain is: ……④

[0018] where r is the distance from the discrete point to the center of the cutter head, and θ is the rotation angle of the cutter head;

[0019] Then, solving formula ③ gives: ……⑤

[0020] T f is the frictional torque generated by the height of the broken body on the idling of the cutter head, μ is the friction coefficient between the surrounding rock and the cutter head, R is the radius of the cutter head, and H is the height of the landslide body.

[0021] In step S1, the spatial range of the length, width, and height of the front landslide body is predicted by the seismic wave method, and the theoretical force T on the cutter head during tunneling is calculated using the height parameter H1 and formula ⑤ f1 , and the breakout torque Tmax of the cutter head is a design parameter of the equipment performance. When T f1 < Tmax, the equipment can pass through this formation without the risk of jamming, and the geological prediction result is that the risk of this formation is no jamming; when T f1 > Tmax, the equipment has the risk of jamming; the geological prediction result is that the risk of this formation is jamming.

[0022] In step S2, the CNN or YOLO image recognition method is used to identify the particle size of the slag chips. The abnormal large particle size and the large particle size correspond to the broken integrity of the surrounding rock, and the medium particle size corresponds to the intact integrity of the surrounding rock.

[0023] It is defined that the particle size of the slag chip greater than 1.5 times the spacing between the hob cutters is an abnormal large particle size, the particle size of the slag chip greater than 1 and less than or equal to 1.5 times the spacing between the hob cutters is a large particle size, and the particle size of the slag chip less than or equal to 1 times the spacing between the hob cutters is a medium particle size.

[0024] In step S2, when testing the surrounding rock strength, the tunneling parameters used are thrust F, torque T, rotation speed n, and propulsion speed v. Then, the comprehensive rock breaking index k = Fv / Tn. The larger the value of k, the greater the surrounding rock strength.

[0025] In step S3, when the integrity and strength of the surrounding rock are analyzed together, statistical analysis and mechanism analysis of big data show that when the integrity of the surrounding rock is broken and the strength of the surrounding rock is low, the comprehensive evaluation result of the surrounding rock quality at the tunnel face is that the quality of the surrounding rock formation is poor and the risk of jamming is high.

[0026] When correcting the geological prediction results in step S4, if the risk assessment result of machine jamming during excavation is high and the geological prediction result indicates that the risk of machine jamming in that stratum is high, the geological interpretation result at that location is accurate. If the risk assessment result of machine jamming during excavation is high and the geological prediction result indicates that the risk of machine jamming in that stratum is low, it indicates that there is a deviation in the geological prediction result. The geological interpretation result at that location is then corrected, and the geological interpretation result is corrected to indicate that there is a risk of machine jamming.

[0027] Based on the revised geological forecast results in step S4, the risk level of the preceding strata is classified as either no jamming or jamming risk. When the risk level is jamming risk, a geological risk warning is issued, and necessary safety measures are taken.

[0028] The beneficial effects of this invention are as follows: This invention verifies TBM tunnel collapse through theoretical analysis, geological monitoring, tunneling parameters, and slag flake analysis; by fusing multi-source data during TBM construction, it improves the reliability of construction risk early warning and enhances the safety and efficiency of TBM construction. This method, through multi-source information fusion and dynamic feedback correction, forms a closed loop of "advanced detection—in-excavation perception—risk assessment—forecast correction—tiered early warning," significantly improving the real-time performance, accuracy, and engineering operability of TBM jam risk identification.

[0029] This invention assesses the risks of TBM construction, primarily focusing on TBM jamming caused by landslides in fault fracture zones. It utilizes earth pressure theory to establish the relationship between landslide height and cutterhead torque, transforming the assessment of landslide height into monitoring cutterhead idling torque. Furthermore, it uses advanced geological forecast data to obtain the distribution range of landslides and combines this with theoretical calculations to predict TBM construction risks in advance. For the forecast results, it dynamically updates the geological forecast and landslide outcomes by combining the analysis of muck parameters and tunneling parameters during the TBM's approach to the landslide. The assessment results of these construction risks are crucial for selecting TBM anti-jamming and escape measures, ensuring the safety of TBM construction in hazardous geological conditions.

[0030] Based on multi-source data from tunnel construction, this invention analyzes and mines excavation parameters, slag parameters, and advanced geological prediction data to achieve comprehensive evaluation of the surrounding rock quality at the tunnel face, evaluation of the current equipment jamming situation, and evaluation of the surrounding rock quality ahead of the tunnel face. This allows for a comprehensive evaluation of the TBM jamming risk in the strata ahead, and the jamming risk is dynamically updated as the TBM excavation progresses, effectively achieving a comprehensive evaluation of jamming risk. Attached Figure Description

[0031] To more clearly illustrate the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0032] Figure 1 This is a schematic diagram of the modeling process of the present invention;

[0033] Figure 2 This is a schematic diagram of the force on the cutterhead under the action of a collapsed body according to the present invention. Detailed Implementation

[0034] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0035] Example 1, as Figure 1 As shown, a real-time early warning method for TBM construction risks based on multi-information fusion includes the following steps: S1 Geological prediction and detection: Advanced geological prediction is performed at the location of the tunnel boring machine (TBM) to predict the spatial extent of the landslide ahead. The theoretical force on the cutterhead during tunneling is calculated using the formula relating the landslide height to the cutterhead's idling friction torque. The theoretical force on the cutterhead is compared with the cutterhead's freeing torque to provide a geological prediction result indicating whether the geological risk is a machine jamming issue. The geological prediction and detection uses a seismic wave prediction system to perform advanced geological prediction ahead of the tunnel face, obtaining comprehensive information on the spatial length, width, and height of abnormal geological bodies ahead of the tunnel face, which is used for subsequent calculations of the relationship between the landslide height and the cutterhead's rotation torque.

[0036] S2 rock integrity and strength testing during excavation: Rock integrity testing involves monitoring the muck discharge video stream from the conveyor belt during tunneling machine operation using a video monitoring system. Image recognition methods are used to identify muck particle size, and the particle size distribution is statistically analyzed. A model relating muck particle size to rock integrity is then established based on this statistical analysis, ultimately enabling the identification of rock integrity at the tunnel face based on muck particle images. This step utilizes muck particle identification to establish the relationship between muck particle size and rock integrity.

[0037] The surrounding rock strength detection process involves: acquiring the tunneling parameters during tunnel boring machine (TBM) construction; establishing a comprehensive rock-breaking index (k) using these parameters; creating a database of the comprehensive rock-breaking index (k) and surrounding rock strength; and constructing a surrounding rock strength prediction model based on the comprehensive rock-breaking index (k) through data statistics. The TBM parameters are then used to evaluate the strength of the surrounding rock during tunneling, enabling rapid perception of its strength. This step utilizes the TBM parameters to establish a tunnel surrounding rock collapse prediction model.

[0038] S3's risk assessment of equipment jamming during excavation involves a fusion analysis of surrounding rock integrity and strength to obtain a comprehensive evaluation of the surrounding rock quality at the excavation face, thus establishing a risk assessment result for equipment jamming during the excavation process. This step is based on the fusion of information on surrounding rock integrity and strength, and real-time geological monitoring during excavation.

[0039] S4 Correction of Geological Interpretation Results: The geological prediction results in step S1 are corrected using the risk assessment results of the stuck machine in step S3. This step uses the geological information obtained from the surrounding rock integrity and strength information to verify the geological prediction detection results and update the geological prediction results. When the prediction results are consistent with the collapse body identification results, the original interpretation results are maintained; when there is a discrepancy between the two results, the geological interpretation results ahead are dynamically corrected in conjunction with the construction identification results.

[0040] S5. Risk warning of strata ahead of equipment construction: Based on the revised geological forecast results in step S4, the risk of the strata ahead is classified, and corresponding measures are taken according to the risk classification to improve the safety of TBM construction.

[0041] This invention verifies the analysis of TBM tunnel collapses through theoretical analysis, geological monitoring, tunneling parameters, and slag flake data. By fusing multi-source data from TBM construction, the reliability of construction risk early warning is improved, enhancing the safety and efficiency of TBM construction. This method, through multi-source information fusion and dynamic feedback correction, forms a closed loop of "advanced detection—in-excavation perception—risk assessment—prediction correction—tiered early warning," significantly improving the real-time performance, accuracy, and engineering operability of TBM jam risk identification. This scheme couples "long-distance macroscopic detection" with "short-distance microscopic perception," upgrading the "static geological model" to a "dynamic digital twin." Without increasing additional construction time, it achieves tiered, quantifiable, and real-time early warning of TBM jam risks, possessing the triple value of safety, economy, and intelligence.

[0042] Example 2: A real-time early warning method for TBM construction risks based on multi-information fusion, comprising the following steps: S1 Geological prediction and detection: Advanced geological prediction is performed at the location of the tunneling machine to predict the spatial extent of the landslide ahead. The theoretical force on the cutterhead during tunneling is calculated using the formula for the height of the landslide and the friction torque of the cutterhead during idle movement. The theoretical force on the cutterhead is compared with the cutterhead escape torque to provide a geological prediction result indicating whether the geological risk is a machine jamming issue. Specifically, as follows... Figure 2 As shown:

[0043] The calculation process of the formula for the height of the landslide and the friction torque of the cutterhead during idle is as follows: Under the stress state of the cutterhead when the machine is stopped, based on Coulomb's earth pressure theory, when the cutterhead is subjected to a certain degree of landslide, each point on the cutterhead is subjected to the passive earth pressure of the landslide, and the torque dT at the discrete points is... f for:

[0044] ……①

[0045] ……②

[0046] In the formula, K p ρ is the passive earth pressure coefficient; φ is the friction angle of the broken soil within the collapse body, with the internal friction angle of the crushed stone being 30°; ρ is the density of the broken soil (kg / m³). 3 g is the acceleration due to gravity in m / s². 2 y represents the vertical height from the discrete point to the center of the cutter head. Let dxdy be the distance from a discrete point to the center of the circle, and dxdy be the integral over the area.

[0047] Integrating formula ②, we get:

[0048] ……③

[0049] The integration domain is: ……④

[0050] where, r is the distance from the discrete point to the center of the cutter head, and θ is the rotation angle of the cutter head;

[0051] Solving formula ③ gives: ……⑤

[0052] T f is the frictional torque generated by the height of the broken body during the idle rotation of the cutter head, μ is the friction coefficient between the surrounding rock and the cutter head, R is the radius of the cutter head, and H is the height of the collapse body.

[0053] According to the above theory, the spatial range of the length, width and height of the front collapse body is predicted by the seismic wave method, and the theoretical force T of the cutter head during tunneling is calculated by using the detected height parameter H1 of the collapse body and formula ⑤. f1 , the breakout torque Tmax of the cutter head is a design parameter of the equipment performance and is a known quantity. When T f1 < Tmax, the equipment can pass through this stratum and there is no risk of jamming. At this time, the geological prediction result is that the risk of this stratum is no jamming; when T f1 > Tmax, there is a risk of jamming for the equipment; the geological prediction result is that the risk of this stratum is jamming.

[0054] The integrity detection of S2 surrounding rock is as follows: during the construction of the roadheader, the particle size of the slag chips is identified by using the CNN or YOLO image recognition method. Based on the identification results of the slag chip particle size, the data statistics of the particle size distribution are carried out, and the slag chips with larger particle sizes are identified. It is defined that the slag chip particle size greater than 1.5 times the cutter spacing is the abnormal large particle size, the slag chip particle size greater than 1 and less than or equal to 1.5 times the cutter spacing is the large particle size, and the slag chip particle size less than or equal to 1 times the cutter spacing is the medium particle size. The integrity of the surrounding rock corresponding to the abnormal large particle size and the large particle size is broken, and the integrity of the surrounding rock corresponding to the medium particle size is intact; finally, the integrity kv of the surrounding rock of the working face during tunneling is judged based on the slag chip image.

[0055] The surrounding rock strength detection is based on the tunneling parameter results. That is, during the construction of the roadheader, the main equipment status monitoring parameters are obtained, such as thrust, torque, rotation speed, propulsion speed, etc. According to the relationship between the monitoring parameters in the rock breaking relationship, through the analysis and excavation of the tunneling parameters, a comprehensive rock breaking index k = Fv / Tn is established by using the parameters of thrust F, torque T, rotation speed n, and propulsion speed v; then the comprehensive rock breaking index k = Fv / Tn, and the greater the k value, the greater the surrounding rock strength. A database of k and the surrounding rock strength UCS is established, and a prediction model of the surrounding rock strength UCS based on k is constructed through data statistics. The surrounding rock strength during tunneling is evaluated through the tunneling parameters, and the rapid perception of the surrounding rock strength UCS is realized.

[0056] S3 integrates the results of cuttings particle size identification and tunneling parameters. Using the integrity (kv) from cuttings particle identification and the surrounding rock strength (UCS) from tunneling parameters, a fusion analysis is performed to obtain a comprehensive evaluation result of the surrounding rock quality at the tunneling face, and to establish a risk assessment result for equipment jamming during the tunneling process. When performing the fusion analysis of surrounding rock integrity and strength, statistical analysis and mechanistic analysis using big data reveal that when the surrounding rock integrity is broken and the surrounding rock strength is low, the comprehensive evaluation result of the surrounding rock quality at the tunneling face is poor, and the risk assessment result is high. It should be noted that the torque characteristic index is used in the comprehensive evaluation of the surrounding rock quality at the tunneling face. This index is calculated using the torque and penetration depth during tunneling, which avoids the influence of shield jamming on the surrounding rock quality evaluation result.

[0057] When S4 corrects the geological prediction results, if the jamming risk assessment during excavation indicates a high jamming risk and the geological prediction indicates a jamming risk in that stratum, the geological interpretation result at that location is accurate. Conversely, if the jamming risk assessment during excavation indicates a high jamming risk but the geological prediction indicates no jamming risk in that stratum, it indicates a deviation in the geological prediction result. The geological interpretation result at that location is then corrected, and simultaneously, a jamming risk is confirmed. In other words, the geological interpretation results can be dynamically updated. Using the identification results of tunnel collapse bodies, the geological prediction results are updated. When the prediction result matches the collapse body identification result, the original interpretation result is maintained; when there is a discrepancy, the geological interpretation result ahead is dynamically corrected based on the construction identification results. Each "prediction-excavation-correction" cycle generates new samples, forming a data-model dual closed loop. The longer the tunnel, the higher the accuracy. It should be noted that, in the risk assessment of the jamming at the front, the expert experience method was used for risk assessment. Alternatively, a neural network method can be used, with the results of the surrounding rock quality assessment at the working face, the current jamming risk, and the surrounding rock quality assessment at the front as the input of the neural network, and the jamming risk level as the output, to assess the risk of jamming at the front.

[0058] Based on the revised geological forecast results from step S4, step S5 classifies the risk of the preceding strata into two levels: no jamming and risk of jamming. A risk level of "risk of jamming" indicates a significant risk in the preceding strata, prompting a geological risk warning and the implementation of necessary safety measures to improve the safety of TBM construction.

[0059] This invention assesses the risks of TBM construction, primarily focusing on TBM jamming caused by fault fracture zone collapse. Based on multi-source data from tunnel construction, and through analysis and mining of excavation parameters, muck parameters, and advanced geological prediction data, it achieves a comprehensive evaluation of the surrounding rock quality at the tunnel face, the current equipment jamming situation, and the quality of the surrounding rock ahead of the tunnel face, thereby comprehensively assessing the TBM jamming risk in the strata ahead. Furthermore, the jamming risk is dynamically updated as TBM excavation progresses, effectively achieving a comprehensive assessment of jamming risk.

[0060] It should be noted that this invention improves the equipment components and does not involve improvements to the circuits or control programs. This invention only controls the operation and shutdown of various electronic devices through a PLC control system. Since the PLC control system is a mature automatic control system in industry, this invention will not elaborate on the circuit and control program content.

[0061] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A real-time early warning method for TBM construction risks based on multi-information fusion, characterized in that: The steps are as follows: S1 Geological Prediction and Detection: Conduct advanced geological prediction at the location of the tunneling machine to predict the spatial range of the collapse ahead, and use the formula of collapse height and cutterhead idling friction torque to calculate the theoretical force on the cutterhead during tunneling. Compare the theoretical force on the cutterhead with the cutterhead escape torque to give the geological prediction result of whether the risk of the stratum is jamming. S2 rock integrity and strength detection during excavation: Rock integrity detection is performed as follows: During tunneling machine construction, the video monitoring system monitors the muck discharge video stream of the conveyor belt, and the muck particle size is identified by image recognition method. The particle size distribution is statistically analyzed, and a relationship model between muck particle size and rock integrity is established through statistical analysis. Finally, the rock integrity of the tunnel face is identified based on muck images. The surrounding rock strength detection process involves: acquiring the tunneling parameters of the tunneling machine during construction, establishing a comprehensive rock breaking index k using these parameters, creating a database of the comprehensive rock breaking index k and the surrounding rock strength, constructing a surrounding rock strength prediction model based on the comprehensive rock breaking index k through data statistics, and evaluating the strength of the surrounding rock under tunneling using the tunneling parameters to achieve rapid perception of the surrounding rock strength. S3 Evaluation of the risk of equipment jamming during excavation: The integrity and strength of the surrounding rock are analyzed together to obtain a comprehensive evaluation result of the surrounding rock quality at the excavation face, and to establish the evaluation result of the risk of equipment jamming during the excavation process; S4 Correction of geological interpretation results: Using the risk assessment results of the machine in step S3, the geological prediction results in step S1 are corrected; S5. Risk warning of strata ahead of equipment construction: Based on the revised geological forecast results in step S4, the risk of the strata ahead is classified, and corresponding measures are taken according to the risk classification.

2. The real-time early warning method for TBM construction risks based on multi-information fusion according to claim 1, characterized in that: The calculation process of the formula for the height of the collapsed body and the friction torque of the cutterhead during idle rotation in step S1 is as follows: Under the stress state of the cutterhead when the machine is stopped, based on the Coulomb earth pressure theory, when the cutterhead is subjected to a certain degree of collapsed body, each point on the cutterhead is subjected to the passive earth pressure of the collapsed body, and the torque dT at the discrete points is... f for: ……① ……② In the formula, K p ρ is the passive earth pressure coefficient; φ is the friction angle of the broken soil within the collapse body, with the internal friction angle of the crushed stone being 30°; ρ is the density of the broken soil (kg / m³). 3 g is the acceleration due to gravity in m / s². 2 y represents the vertical height from the discrete point to the center of the cutter head. Let dxdy be the distance from a discrete point to the center of the circle, and dxdy be the integral over the area. Integrating formula ②, we get: ……③ The integration domain is: ……④ Where r is the distance from the discrete point to the center of the cutter head, and θ is the rotation angle of the cutter head; Solving formula ③ yields: ……⑤ T f The frictional torque generated by the height of the crushed material on the idling of the cutterhead is given by denoted as μ, where μ is the friction coefficient between the surrounding rock and the cutterhead, R is the radius of the cutterhead, and H is the height of the collapsed material.

3. The real-time early warning method for TBM construction risks based on multi-information fusion according to claim 2, characterized in that: In step S1, the spatial range of the length, width, and height of the ahead collapse body is predicted by the seismic wave method, and the theoretical force T on the cutter head during tunneling is calculated using the height parameter H1 and formula ⑤. f1 , the breakout torque Tmax of the cutter head is a design parameter of the equipment performance. When T f1 < Tmax, the equipment can pass through this formation without the risk of getting stuck. At this time, the geological prediction result is that the risk of this formation is no getting stuck; when T f1 > Tmax, the equipment has the risk of getting stuck; the geological prediction result is that the risk of this formation is getting stuck.

4. The real-time early warning method for TBM construction risks based on multi-information fusion according to any one of claims 1 to 3, characterized in that: In step S2, CNN or YOLO image recognition methods are used to identify the particle size of the slag fragments. Abnormally large particle sizes and large particle sizes correspond to broken surrounding rock, while medium particle sizes correspond to intact surrounding rock.

5. The real-time early warning method for TBM construction risks based on multi-information fusion according to claim 4, characterized in that: Slag flakes with a diameter greater than 1.5 times the cutter spacing are defined as abnormally large slag flakes, slag flakes with a diameter greater than 1 and less than or equal to 1.5 times the cutter spacing are defined as large slag flakes, and slag flakes with a diameter less than or equal to 1 times the cutter spacing are defined as medium slag flakes.

6. The real-time early warning method for TBM construction risks based on multi-information fusion according to claim 1 or 5, characterized in that: In step S2, when testing the surrounding rock strength, the tunneling parameters used are thrust F, torque T, rotation speed n, and propulsion speed v. Then, the comprehensive rock breaking index k = Fv / Tn. The larger the value of k, the greater the surrounding rock strength.

7. The real-time early warning method for TBM construction risks based on multi-information fusion according to claim 6, characterized in that: In step S3, when the integrity and strength of the surrounding rock are analyzed together, statistical analysis and mechanism analysis of big data show that when the integrity of the surrounding rock is broken and the strength of the surrounding rock is low, the comprehensive evaluation result of the surrounding rock quality at the tunnel face is that the quality of the surrounding rock formation is poor and the risk of jamming is high.

8. The real-time early warning method for TBM construction risks based on multi-information fusion according to claim 7, characterized in that: When correcting the geological prediction results in step S4, if the risk assessment result of machine jamming during the excavation process is high and the geological prediction result is that the risk of machine jamming in this stratum is high, then the geological interpretation result at that location is accurate. If the risk assessment result for machine jamming during excavation is high, but the geological prediction result is no risk of machine jamming in that stratum, it indicates that there is a deviation in the geological prediction result. The geological interpretation result for that location should be corrected, and the geological interpretation result should be corrected to indicate that there is a risk of machine jamming.

9. The real-time early warning method for TBM construction risks based on multi-information fusion according to claim 1 or 8, characterized in that: Based on the revised geological forecast results in step S4, the risk level of the strata ahead is classified as either no jamming or jamming risk.

10. The real-time early warning method for TBM construction risks based on multi-information fusion according to claim 9, characterized in that: When the risk level is considered to be at risk of machine jamming, a geological risk warning will be issued and necessary safety measures will be taken.

Images