TBM hob abrasion identification quantitative monitoring system and method based on vibration and magnetic induction combination

Abstract

The invention discloses a TBM hob abrasion identification quantitative monitoring system and method based on vibration and magnetic induction combination. The TBM hob abrasion identification quantitative monitoring system comprises a hob assembly, a sensing monitoring unit, a data processing module and a terminal early warning platform. The hob assembly comprises a hob box body and a hob, a permanent magnet is arranged on the hob, the sensing monitoring unit comprises a vibration sensor and a magnetic induction monitoring module, and the data processing module is electrically connected with the vibration sensor and the magnetic induction monitoring module and is used for analyzing and processing a vibration signal and a magnetic induction signal; and the terminal early warning platform is in communication connection with the data processing module, receives information transmitted by the data processing module, and is used for displaying the abrasion state of the hob and giving out early warning. According to the system and the method, vibration and magnetic induction multi-parameter information are fused, the defect that a traditional single-parameter monitoring mode is easily interfered by geological sudden change is overcome, abnormal wear modes such as uniform wear and eccentric wear can be effectively distinguished, and the monitoring accuracy and reliability are improved.

Description

Technical Field

[0001] This invention relates to the field of intelligent monitoring technology for cutter wear of full-face tunnel boring machines (TBMs), specifically to a TBM cutter wear identification and quantitative monitoring system and method based on vibration-magnetic sensing composite monitoring. Background Technology

[0002] In recent years, with the accelerated pace of infrastructure construction, ultra-long and deep-buried tunnel projects have placed higher demands on the construction efficiency and reliability of full-face tunnel boring machines (TBMs). As the core rock-breaking component of the TBM, the disc cutterhead endures high stress, strong impact, and intense friction under complex geological conditions for extended periods. Its wear condition directly affects the stability of tunneling parameters and the service life of the equipment. Studies have shown that cutterhead wear significantly increases the thrust demand of the TBM, and cutterhead failures caused by abnormal wear can lead to huge downtime and maintenance costs, severely restricting construction progress. Therefore, real-time monitoring of the cutterhead wear condition has become a key technical bottleneck for ensuring efficient TBM tunneling.

[0003] Existing hob wear monitoring technologies have significant limitations: traditional manual inspection methods require opening the cutter face under high pressure, high humidity, and high dust conditions, resulting in lengthy inspection times and significant safety risks such as cutterhead jamming, water inrush, and mud inrush. Indirect monitoring technologies, such as the odor additive method, are significantly affected by the permeability characteristics of the formation, exhibiting low accuracy and insufficient reliability in complex geological conditions such as water-rich sand layers. Pressure sensor-based monitoring schemes suffer from severe signal drift due to the extreme vibration environment inside the cutter box, making it difficult to meet the accuracy requirements of engineering applications. More importantly, most existing monitoring methods can only provide a rough estimate of wear, failing to effectively distinguish abnormal wear patterns such as uneven wear, chipping, and bearing jamming. Engineering practice shows that abnormal conditions such as uneven wear drastically increase the lateral force on the hob, seriously threatening the life and safety of core components such as the cutterhead main bearing, while traditional monitoring methods suffer from a significant lag in providing early warnings for such high-risk conditions. In addition, single-parameter monitoring mode is susceptible to geological abrupt changes. In specific strata such as hard rock, the relationship between vibration signal and wear is highly nonlinear, resulting in large errors in monitoring results and low reliability.

[0004] Given the aforementioned shortcomings of existing tunnel boring machine (TBM) cutter wear monitoring devices and methods, which make it difficult to achieve accurate and comprehensive identification and evaluation of TBM cutter wear information, this paper proposes to establish a TBM cutter wear identification and quantitative monitoring system and method based on vibration-magnetic sensing composite monitoring, providing an intelligent real-time monitoring solution for TBM cutter wear. Summary of the Invention

[0005] The purpose of this invention is to address the problems existing in the prior art by providing a quantitative monitoring system and method for TBM hob wear identification based on vibration and magnetic induction composite, aiming to achieve real-time and accurate identification of hob wear status through multi-physics field coupling monitoring technology.

[0006] To achieve the above objectives, the technical solution adopted by the present invention is as follows: A quantitative monitoring system for TBM cutter wear identification based on vibration and magnetic induction composite includes a cutter assembly, a sensing and monitoring unit, a data processing module, and a terminal early warning platform; The hobbing cutter assembly includes a hobbing cutter housing, in which a rotatable hobbing cutter is disposed. At least one end face of the hobbing cutter is provided with a permanent magnet, which generates a regularly changing magnetic field signal when rotating synchronously with the hobbing cutter. The sensing and monitoring unit includes a vibration sensor and a magnetic sensing monitoring module. The vibration sensor is disposed on the inner wall of the cutter box or on the cutter, and is used to collect the vibration signal generated by the cutter breaking the rock. The magnetic sensing monitoring module is disposed on the cutter box and maintains a radial distance from the rotation trajectory of the permanent magnet, and is used to monitor the magnetic induction signal generated by the rotation of the permanent magnet. The data processing module is electrically connected to the vibration sensor and the magnetic sensing module respectively, and is used to analyze and process the vibration signal and the magnetic sensing signal. The terminal early warning platform is communicatively connected to the data processing module and receives information transmitted by the data processing module via wireless communication to display the wear status of the hobbing cutter and issue early warnings.

[0007] This TBM cutter wear identification and quantitative monitoring system integrates multi-parameter information from vibration and magnetic induction, overcoming the shortcomings of traditional single-parameter monitoring modes that are susceptible to interference from sudden geological changes. It can effectively distinguish abnormal wear patterns such as uniform wear and uneven wear, improving the accuracy and reliability of monitoring. Furthermore, the entire system has a compact structure, occupies little space, and is suitable for the harsh environment inside the TBM cutter box, characterized by high temperature, high humidity, strong vibration, and multiple obstructions. Installation and maintenance are also simple and convenient. Through the coordinated setup of the sensing monitoring unit, the data processing module, and the cutter assembly, it can achieve accurate identification of cutter wear types and quantitative monitoring of wear amounts. It has strong anti-interference capabilities and high monitoring accuracy, providing reliable technical support for efficient and safe TBM construction.

[0008] Furthermore, the hob is connected to the hob housing via a bearing, and the permanent magnets are evenly distributed in a ring on the end face of the hob, with the magnetic poles of adjacent permanent magnets alternately arranged.

[0009] Furthermore, the vibration sensor is embedded in the inner wall of the cutter housing and is a triaxial accelerometer to capture the vibration acceleration signals of the cutter in the radial, axial and circumferential directions.

[0010] Furthermore, the hobbing cutter housing is provided with a mounting slot, and the magnetic sensing monitoring module is encapsulated in the mounting slot.

[0011] Furthermore, the magnetic sensing module comprises at least three Hall sensors evenly arranged along the circumferential direction.

[0012] Furthermore, the data processing module includes a signal preprocessing unit, a feature extraction unit, and a wear identification unit. The signal preprocessing unit uses a wavelet transform noise reduction algorithm to filter and reduce noise in the vibration signal and magnetic induction signal. The feature extraction unit extracts characteristic parameters such as vibration frequency, vibration amplitude, magnetic field strength change rate, and amplitude from the preprocessed vibration signal and magnetic induction signal to provide data support for the wear identification unit. The wear identification unit analyzes and judges the wear type and wear amount of the hob. The wear type includes at least uniform wear, uneven wear, severe wear, and chipped abnormal wear.

[0013] In the uniform wear state, the cutter makes uniform contact with the rock, and the vibration amplitude shows a steady increasing trend. The vibration frequency parameters such as the main frequency and the secondary main frequency remain basically stable without significant fluctuations. In the uneven wear state, the cutter is subjected to unbalanced forces, and there is a significant difference between the radial and axial vibration amplitudes. This difference continues to increase with the degree of uneven wear, and the vibration frequency shows small irregular fluctuations. In the severe wear state, the structural integrity of the cutter is damaged, and the rock-breaking ability decreases. This is manifested by a sharp increase in vibration amplitude, a significant decrease in vibration frequency, and a significant decrease in the stability of the vibration signal, with frequent amplitude abrupt changes.

[0014] Furthermore, the terminal early warning platform uses an industrial computer or monitoring software to display the wear amount, wear type, vibration curve, and magnetic field change curve of the hob in real time.

[0015] A quantitative monitoring method for TBM hob wear identification based on vibration-magnetic induction composite is disclosed. This method employs the vibration-magnetic induction composite TBM hob wear identification and quantitative monitoring system described above, and includes the following steps: System initialization: Install and debug the monitoring system, calibrate the vibration sensor and the magnetic sensing module, and set the wear judgment threshold and data acquisition frequency of the hob; Signal acquisition: During the TBM tunneling process, the vibration sensor acquires the vibration signal of the cutter head in real time, and the magnetic induction monitoring module acquires the magnetic induction signal generated by the rotation of the permanent magnet synchronously. Signal preprocessing: The data processing module filters and reduces noise from the collected vibration and magnetic induction signals to remove environmental interference and noise; Feature extraction: Extracting feature parameters such as vibration frequency, vibration amplitude, rate of change of magnetic field strength, and amplitude from the preprocessed vibration signal and magnetic induction signal; Wear identification and quantitative calculation: Based on the multi-parameter fusion algorithm, and the feature parameters and the wear discrimination threshold, the wear type of the hob is identified and the wear amount of the hob is calculated; wherein, the identification of the wear type is related to the significant change of the vibration amplitude, and the abnormal increase of the vibration amplitude is associated with severe wear, uneven wear, and abnormal wear such as chipping. Status display and early warning: The terminal early warning platform displays the wear status, wear type and wear amount of the hob in real time. When the wear amount exceeds the wear judgment threshold or an abnormal wear type occurs, an early warning signal is issued. When the system determines that the wear is severe or abnormal and requires tool replacement, the early warning platform clearly displays a tool replacement suggestion.

[0016] Furthermore, in different full-face tunnel boring machines and different geological formations, the characteristic parameters and wear discrimination thresholds need to be re-obtained through in-depth big data mining and continuously iterated and updated as new monitoring data is added during the tunneling process.

[0017] Furthermore, wavelet transform denoising algorithm is used in signal preprocessing to remove high-frequency noise and low-frequency interference signals; the multi-parameter fusion algorithm adopts a fusion model based on BP neural network, and the fusion model is trained by training samples to achieve accurate identification of wear type and quantitative calculation of wear amount.

[0018] Compared with existing technologies, the beneficial effects of this invention are: 1. This TBM cutter wear identification and quantitative monitoring system integrates multi-parameter information of vibration and magnetic induction, clearly identifying the differences in vibration characteristics under different wear modes. It can accurately distinguish wear types through the specific variation patterns of vibration parameters, overcoming the shortcomings of traditional single-parameter monitoring modes which are easily affected by geological changes. It can effectively distinguish abnormal wear modes such as uniform wear and uneven wear, improving the accuracy and reliability of monitoring. Moreover, the entire system has a compact structure, occupies little space, and is suitable for the harsh environment of high temperature, high humidity, strong vibration, and multiple obstructions inside the TBM cutter box. Installation and maintenance are also simple and convenient. 2. This TBM cutter wear identification and quantitative monitoring system, through the coordinated setup of the sensing monitoring unit, the data processing module, and the cutter assembly, can achieve accurate identification of cutter wear types and quantitative monitoring of wear amount. It has strong anti-interference capabilities and high monitoring accuracy, and can provide a reliable monitoring solution for TBM cutter wear. 3. By integrating magnetic induction counting and vibration signal analysis technologies, the cutter vibration signal is converted into characteristic parameters such as vibration frequency and amplitude per unit time of tunneling, and the cutter magnetic induction signal is converted into characteristic parameters such as magnetic field strength change frequency and amplitude per unit time of tunneling. Through in-depth mining and iterative learning of monitoring big data, the identification threshold and model are adaptively updated. Through multi-parameter fusion, the accurate identification of wear type and dynamic quantitative monitoring of wear amount are achieved; 4. This monitoring system and method are easy to operate, realize real-time online monitoring, do not require opening the chamber, reduce the safety risks of manual inspection and long downtime, reduce construction costs, and significantly improve construction safety and efficiency. Attached Figure Description

[0019] Figure 1 A schematic diagram showing the arrangement of the hobbing cutter assembly and the sensing and monitoring unit of the present invention; Figure 2 This is a schematic diagram of the magnetic sensing monitoring module structure of the present invention; Figure 3 This is a flowchart of vibration-magnetic data processing in the TBM hobbing cutter wear identification and quantitative monitoring system based on vibration-magnetic induction composite of the present invention. Figure 4 This is a flowchart of the hobbing cutter wear identification and quantitative monitoring method of the present invention; Figure 5 This is a schematic diagram of the interface of the terminal early warning platform involved in the present invention; In the diagram: 1. Roller cutter; 2. Roller cutter housing; 3. Permanent magnet; 4. Magnetic sensing monitoring module; 5. Vibration sensor; 6. Fixed resistor; 7. Protective housing; 8. Circuit switch; 9. Flexible insulating spring; 10. Magnetic switch; 11. Positioning plate; 12. Strong adhesive layer; 13. Wireless transmitter; 14. Power supply. Detailed Implementation

[0020] The technical solution of the present invention will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are merely 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.

[0021] In the description of this invention, it should be noted that the terms "middle", "upper", "lower", "left", "right", "inner", "outer", etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing this invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this invention.

[0022] Example 1: Combination Figure 1 As shown, a quantitative monitoring system for TBM cutter wear identification based on vibration and magnetic induction composite includes a cutter assembly, a sensing and monitoring unit, a data processing module, and a terminal early warning platform. The hobbing cutter assembly includes a hobbing cutter housing 2, in which a rotatable hobbing cutter 1 is provided. At least one end face of the hobbing cutter 1 is provided with a permanent magnet 3, which generates a regularly changing magnetic field signal when it rotates synchronously with the hobbing cutter 1. The sensing and monitoring unit includes a vibration sensor 5 and a magnetic sensing monitoring module 4. The vibration sensor 5 is disposed on the inner wall of the cutter box or on the cutter, and is used to collect the vibration signal generated by the cutter breaking the rock. The magnetic sensing monitoring module 4 is disposed on the cutter box 2 and maintains a radial distance from the rotation trajectory of the permanent magnet 3, and is used to monitor the magnetic induction signal generated by the rotation of the permanent magnet. The data processing module is electrically connected to the vibration sensor and the magnetic sensing module respectively, and is used to analyze and process the vibration signal and the magnetic sensing signal. The terminal early warning platform is communicatively connected to the data processing module and receives information transmitted by the data processing module via wireless communication to display the wear status of the hobbing cutter and issue early warnings.

[0023] This TBM cutter wear identification and quantitative monitoring system integrates multi-parameter information from vibration and magnetic induction, overcoming the shortcomings of traditional single-parameter monitoring modes that are susceptible to interference from geological changes. It can effectively distinguish abnormal wear modes such as uniform wear and uneven wear, improving the accuracy and reliability of monitoring. Moreover, the entire system has a compact structure, occupies little space, and is suitable for the harsh environment of high temperature, high humidity, strong vibration, and multiple obstructions inside the TBM cutter box. Installation and maintenance are also simple and convenient.

[0024] This TBM cutter wear identification and quantitative monitoring system, through the cooperation of the sensing monitoring unit, the data processing module, and the cutter assembly, can accurately identify the type of cutter wear and quantitatively monitor the amount of wear. It has strong anti-interference ability and high monitoring accuracy, and can provide reliable technical support for efficient and safe TBM construction.

[0025] The permanent magnet 3 is fixedly installed on the end face of the roller cutter 1 and rotates synchronously with the roller cutter 1. The permanent magnet 3 is made of neodymium iron boron strong magnetic material, which has high magnetic energy product and stability, and can generate the magnetic signal required for magnetic induction monitoring to reflect the periodic changes generated when the roller cutter rotates.

[0026] The sensing and monitoring unit consists of a vibration sensor 5 and a magnetic sensing module 4, which work together to achieve comprehensive monitoring of the cutting head's operating status. The vibration sensor 5 is specifically used to collect the vibration signals generated by the cutting head breaking rock; the magnetic sensing module 4 can detect the frequency and amplitude of the magnetic field changes generated by the rotation of the permanent magnet 3 in real time with high precision.

[0027] The system also includes a power module, which adopts a waterproof and sealed design, has good waterproof and dustproof performance, adapts to the harsh environment inside the TBM toolbox, provides stable power supply for the sensing and monitoring unit and data processing module, and has a battery life of no less than 30 days, reducing the trouble of frequent power supply replacements.

[0028] Furthermore, the hob 1 is connected to the hob housing 2 via a bearing, and the permanent magnets 3 are evenly distributed in a ring on the end face of the hob end. The magnetic poles of adjacent permanent magnets 3 are alternately arranged, for example, the N poles and S poles of multiple permanent magnets 3 are arranged in sequence to ensure that a regularly changing magnetic field signal is generated during rotation.

[0029] Furthermore, the vibration sensor 5 is embedded in the inner wall of the hob housing 2. It is a triaxial accelerometer with a high sampling frequency and a suitable measurement range. It can capture the vibration acceleration signals of the hob in the radial, axial and circumferential directions, thereby comprehensively reflecting the vibration state of the hob.

[0030] Furthermore, the cutter housing 2 is provided with a mounting slot, and the magnetic sensing monitoring module 4 is encapsulated in the mounting slot, maintaining an appropriate radial distance from the rotation trajectory of the permanent magnet 3, so as to monitor the magnetic induction signal generated by the rotation of the permanent magnet. The embedded installation and waterproof sealing design not only saves space but also effectively adapts to the harsh environment of high temperature, high humidity, strong vibration, and multiple obstructions inside the TBM cutter housing.

[0031] Furthermore, the magnetic sensing monitoring module 4 consists of at least three Hall sensors evenly arranged along the circumferential direction. The pulse signals output by the Hall sensors can effectively reflect the number of rotations, frequency, and rotational stability of the cutter head within a unit time or unit distance of tunnel boring machine excavation.

[0032] In some implementations, such as Figure 2 As shown, the magnetic sensing monitoring module includes a protective housing 7 disposed on the cutter box 2. A positioning plate 11 is connected inside the protective housing 7 by a strong adhesive layer 12. A magnetic sensor switch 10 and a magnetic sensing circuit connected to the magnetic sensor switch 10 are disposed on the positioning plate 11. A flexible insulating spring 9 connected to the circuit is provided on the opening and closing side of the magnetic sensor switch 10. A wireless transmitter 13, a power supply 14, a circuit switch 8 and a stator resistor 6 are connected in series from one side of the magnetic sensor switch 10 to the other side.

[0033] Furthermore, the data processing module includes a signal preprocessing unit, a feature extraction unit, and a wear identification unit. The signal preprocessing unit uses a wavelet transform denoising algorithm to filter and denoise the vibration signal and magnetic induction signal, removing environmental interference and noise to improve signal quality. The feature extraction unit extracts feature parameters such as vibration frequency, vibration amplitude, magnetic field strength change rate, and amplitude from the preprocessed vibration signal and magnetic induction signal, providing data support for the wear identification unit. The wear identification unit is based on a multi-parameter fusion algorithm, using a fusion model based on a BP neural network. The model is trained using training samples to analyze and determine the wear type and wear amount of the hob. The wear type includes at least uniform wear, uneven wear, severe wear, and chipped abnormal wear.

[0034] For the uniform wear described above, its vibration amplitude shows a steady increasing trend with an increasing rate of approximately 0.05–0.2 g / h (g is the unit of gravitational acceleration), without any abrupt changes; its vibration frequency is stable at 50–200 Hz with a fluctuation range of ≤ ±5 Hz, and the secondary frequency shows no significant change; its core characteristics are that the radial, axial, and circumferential vibration parameters are balanced with no significant differences.

[0035] For the aforementioned uneven wear, its vibration amplitude is as follows: the radial and axial amplitudes differ significantly, with a difference ≥2g (which continues to increase as wear intensifies); its vibration frequency exhibits small, irregular fluctuations, with a fluctuation range of ±5 to 15Hz and a main frequency deviation of no more than 20%; its core characteristic is that the imbalance of force leads to significant differences in vibration direction and a decrease in frequency stability.

[0036] For severe wear or abnormal wear with chipping, the vibration amplitude increases sharply, with an increase of ≥50% compared to the normal state (often exceeding 10g), accompanied by frequent amplitude abrupt changes; the vibration frequency decreases significantly, with the main frequency decreasing by ≥30% compared to the normal state (e.g., from 150Hz to ≤105Hz); its core characteristics are: a significant decrease in vibration signal stability and violent synchronous fluctuations in the magnetic field signal.

[0037] Furthermore, the terminal early warning platform uses an industrial computer or monitoring software to display information such as the wear amount, wear type, vibration curve, and magnetic field change curve of the hob in real time, so that operators can keep track of the hob's status in real time.

[0038] Example 2: This example provides a quantitative monitoring method for TBM hobbing cutter wear identification based on vibration and magnetic induction composite.

[0039] This method employs a TBM hob wear identification and quantitative monitoring system based on vibration-magnetic induction composite as described in Example 1, combined with... Figures 3-5 As shown, the method includes the following steps: Step 1: System initialization: Install and debug the monitoring system, calibrate the vibration sensor and the magnetic sensing module, and set the wear judgment threshold and data acquisition frequency of the hob; Specifically, the vibration sensor and the magnetic sensing monitoring module are installed on the cutter housing at preset positions and connected to the data processing module and the terminal early warning platform; the vibration sensor and the magnetic sensing monitoring module are calibrated by using a standard vibration source and a standard magnetic field to ensure measurement accuracy; the wear discrimination threshold under different wear levels is determined according to the cutter type and engineering requirements, and the data acquisition frequency is usually set to 1kHz, which can be adjusted according to actual needs.

[0040] Step 2, Signal Acquisition: During TBM tunneling, the vibration sensor collects the vibration signals of the cutter head in real time, including radial, axial, and circumferential vibration acceleration data; the magnetic induction monitoring module synchronously collects the magnetic induction signals generated by the rotation of the permanent magnet, including the frequency and amplitude data of magnetic field strength changes. The collected signals are transmitted to the data processing module via wired or wireless transmission.

[0041] Step 3, Signal Preprocessing: The data processing module filters and reduces noise from the collected vibration and magnetic induction signals to remove environmental interference and noise.

[0042] Step 4: Feature extraction: Extract the vibration frequency, vibration amplitude, magnetic field strength change rate and amplitude from the preprocessed vibration signal and magnetic induction signal.

[0043] Step 5, Wear Identification and Quantitative Calculation: Based on the multi-parameter fusion algorithm, and the feature parameters and the wear discrimination threshold, the wear type of the hob is identified, and the wear amount of the hob is calculated; wherein, the identification of the wear type is related to the significant change of the vibration amplitude, and the abnormal increase of the vibration amplitude is associated with severe wear, uneven wear, and abnormal wear such as chipping. When the vibration amplitude increases abnormally, it is used as a key basis for determining severe wear, uneven wear, or abnormal wear patterns such as chipping. The specific criteria are as follows: when the radial, axial, and circumferential vibration parameters collected by the vibration sensor show a steady increase in vibration amplitude and a stable vibration frequency, it is determined to be uniform wear; when the radial and axial vibration amplitudes differ significantly and the difference widens, and the vibration frequency fluctuates slightly, it is determined to be uneven wear; when the vibration amplitude increases sharply, the vibration frequency decreases significantly, and the stability of the vibration signal decreases, it is determined to be severe wear.

[0044] Step 6, Status Display and Early Warning: The terminal early warning platform displays the wear status, wear type and wear amount of the hob in real time. When the wear amount exceeds the wear judgment threshold or an abnormal wear type occurs, an early warning signal is issued. When the system determines that the wear is severe or abnormal and requires tool replacement, the early warning platform clearly displays a tool replacement suggestion.

[0045] The above monitoring method enables real-time online monitoring without the need for warehouse opening, reducing the safety risks of manual inspections and lengthy downtime, lowering construction costs, and significantly improving construction safety and efficiency.

[0046] This monitoring method integrates magnetic induction counting and vibration signal analysis techniques to convert the cutter vibration signal into characteristic parameters such as vibration frequency and amplitude per unit time during tunneling, and the cutter magnetic induction signal into characteristic parameters such as the frequency and amplitude of magnetic field strength changes per unit time during tunneling. Through in-depth mining and iterative learning of the monitoring big data, the method adaptively updates the identification threshold and model. By fusing multiple parameters, it achieves accurate identification of wear types (uniform wear, uneven wear, severe wear, etc.) and dynamic quantitative monitoring of wear amount. This method can simultaneously complete the identification of wear type and calculation of average wear value for a single cutter. It is simple, reliable, and possesses self-learning and adaptive capabilities, making it suitable for monitoring the wear of various cutters in tunnel boring machines. It has significant implications for effectively controlling construction costs through cutter wear monitoring and provides a key intelligent monitoring technology solution for the efficient and safe construction of tunnel boring machines.

[0047] Furthermore, in different full-face tunnel boring machines and different geological formations, the characteristic parameters and wear discrimination thresholds need to be re-obtained through in-depth big data mining and continuously iterated and updated as new monitoring data is added during the tunneling process.

[0048] Furthermore, in the signal preprocessing of step 3, a wavelet transform denoising algorithm is used to remove high-frequency noise and low-frequency interference signals. The multi-parameter fusion algorithm adopts a fusion model based on a BP neural network. The fusion model is trained using training samples to achieve accurate identification of wear type and quantitative calculation of wear amount. For example, uniform wear is characterized by a uniform increase in vibration amplitude, uneven wear is characterized by a significant difference between radial and axial vibration amplitudes, and severe wear is characterized by a decrease in vibration frequency and violent fluctuations in magnetic field signal. At the same time, based on the mapping relationship between characteristic parameters and wear amount, the wear amount of the hob is calculated, which can achieve a high-precision quantitative assessment of wear amount.

[0049] When the wear exceeds the set threshold or an abnormal wear type occurs, a corresponding warning signal is issued. These warning signals include audible and visual alarms (an alarm device is installed in the on-site control room) and remote data push notifications (notifying relevant management personnel via SMS, APP push, etc.). Warning levels are divided into three levels based on the wear amount and type: Level 1 (minor wear), Level 2 (moderate wear), and Level 3 (severe wear / abnormal wear, including uneven wear, chipping, etc.). Different levels correspond to different recommended countermeasures. The Level 3 warning message explicitly includes the operational instruction to "replace the hob."

[0050] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A quantitative monitoring system for TBM hobbing cutter wear identification based on vibration-magnetic induction composite, characterized in that, It includes a roller cutter assembly, a sensing and monitoring unit, a data processing module, and a terminal early warning platform; The hobbing cutter assembly includes a hobbing cutter housing, in which a rotatable hobbing cutter is disposed. At least one end face of the hobbing cutter is provided with a permanent magnet, which generates a regularly changing magnetic field signal when rotating synchronously with the hobbing cutter. The sensing and monitoring unit includes a vibration sensor and a magnetic sensing monitoring module. The vibration sensor is disposed on the inner wall of the cutter box or on the cutter, and is used to collect the vibration signal generated by the cutter breaking the rock. The magnetic sensing monitoring module is disposed on the cutter box and maintains a radial distance from the rotation trajectory of the permanent magnet, and is used to monitor the magnetic induction signal generated by the rotation of the permanent magnet. The data processing module is electrically connected to the vibration sensor and the magnetic sensing module respectively, and is used to analyze and process the vibration signal and the magnetic sensing signal. The terminal early warning platform is communicatively connected to the data processing module and receives information transmitted by the data processing module via wireless communication to display the wear status of the hobbing cutter and issue early warnings.

2. The TBM hobbing cutter wear identification and quantitative monitoring system based on vibration magnetic induction composite as described in claim 1, characterized in that, The hob is connected to the hob housing via a bearing. The permanent magnets are evenly distributed in a ring on the end face of the hob, and the magnetic poles of adjacent permanent magnets are alternately arranged.

3. The TBM hobbing cutter wear identification and quantitative monitoring system based on vibration magnetic induction composite as described in claim 1, characterized in that, The vibration sensor is embedded in the inner wall of the cutter housing and uses a triaxial accelerometer to capture the vibration acceleration signals of the cutter in the radial, axial and circumferential directions.

4. The TBM hobbing cutter wear identification and quantitative monitoring system based on vibration magnetic induction composite as described in claim 1, characterized in that, The hobbing cutter housing is provided with a mounting slot, and the magnetic sensing monitoring module is encapsulated in the mounting slot.

5. The TBM hobbing cutter wear identification and quantitative monitoring system based on vibration magnetic induction composite as described in claim 1, characterized in that, The magnetic sensing module consists of at least three Hall sensors evenly arranged along the circumferential direction.

6. The TBM hob wear identification and quantitative monitoring system based on vibration magnetic induction composite as described in claim 1, characterized in that, The data processing module includes a signal preprocessing unit, a feature extraction unit, and a wear identification unit. The signal preprocessing unit uses a wavelet transform noise reduction algorithm to filter and reduce noise in the vibration signal and magnetic induction signal. The feature extraction unit extracts characteristic parameters such as vibration frequency, vibration amplitude, magnetic field strength change rate, and amplitude from the preprocessed vibration signal and magnetic induction signal to provide data support for the wear identification unit. The wear identification unit analyzes and judges the wear type and wear amount of the hob. The wear type includes at least uniform wear, uneven wear, severe wear, and chipped abnormal wear.

7. The TBM hobbing cutter wear identification and quantitative monitoring system based on vibration magnetic induction composite as described in claim 1, characterized in that, The terminal early warning platform uses an industrial computer or monitoring software to display the wear amount, wear type, vibration curve, and magnetic field change curve of the hob in real time.

8. A quantitative monitoring method for TBM hobbing cutter wear identification based on vibration-magnetic induction composite, characterized in that, This method employs the TBM hob wear identification and quantitative monitoring system based on vibration magnetic induction composite as described in any one of claims 1 to 7, and includes the following steps: System initialization: Install and debug the monitoring system, calibrate the vibration sensor and the magnetic sensing module, and set the wear judgment threshold and data acquisition frequency of the hob; Signal acquisition: During the TBM tunneling process, the vibration sensor acquires the vibration signal of the cutter head in real time, and the magnetic induction monitoring module acquires the magnetic induction signal generated by the rotation of the permanent magnet synchronously. Signal preprocessing: The data processing module filters and reduces noise from the collected vibration and magnetic induction signals to remove environmental interference and noise; Feature extraction: Extracting feature parameters such as vibration frequency, vibration amplitude, rate of change of magnetic field strength, and amplitude from the preprocessed vibration signal and magnetic induction signal; Wear identification and quantitative calculation: Based on the multi-parameter fusion algorithm, and the feature parameters and the wear discrimination threshold, the wear type of the hob is identified and the wear amount of the hob is calculated; wherein, the identification of the wear type is related to the significant change of the vibration amplitude, and the abnormal increase of the vibration amplitude is associated with severe wear, uneven wear, and abnormal wear such as chipping. Status display and early warning: The terminal early warning platform displays the wear status, wear type and wear amount of the hob in real time. When the wear amount exceeds the wear judgment threshold or an abnormal wear type occurs, an early warning signal is issued. When the system determines that the wear is severe or abnormal and requires tool replacement, the early warning platform clearly displays a tool replacement suggestion.

9. The quantitative monitoring method for TBM hobbing cutter wear identification based on vibration-magnetic induction composite as described in claim 8, characterized in that, In different full-face tunnel boring machines and different geological formations, the characteristic parameters and wear discrimination thresholds need to be re-obtained through in-depth big data mining and continuously iterated and updated as new monitoring data are added during the tunneling process.

10. The TBM hobbing cutter wear identification and quantitative monitoring method based on vibration-magnetic induction composite as described in claim 8, characterized in that, The signal preprocessing uses a wavelet transform denoising algorithm to remove high-frequency noise and low-frequency interference signals; the multi-parameter fusion algorithm uses a fusion model based on a BP neural network, and trains the fusion model with training samples to achieve accurate identification of wear type and quantitative calculation of wear amount.

Images