Shield tunneling machine hob damage monitoring method and system, medium and product

Abstract

The invention discloses a shield tunneling machine hob damage monitoring method and system, a medium and a product, and relates to the technical field of shield construction monitoring. The hob damage monitoring system firstly obtains a vibration signal and an acoustic emission signal of a hob, and divides the signals into time sequence synchronization segments to guarantee data correspondence. Then extracting a vibration feature vector and calculating an Euclidean distance so as to reflect a vibration change trend; and acoustic emission characteristic parameters and change rates are determined, and acoustic emission signal characteristics are reflected. After the interference data fragments are removed, abnormal signals are marked, and whether the hob is damaged or not is judged according to the abnormal marking duration and the number density. All the steps cooperate with one another to form a complete process, vibration signals caused by instantaneous impact, acoustic emission signal abrupt change and signal characteristics generated by real damage of the hob can be effectively distinguished, the damage condition of the hob can be accurately recognized, early warning can be given out in time, the monitoring accuracy is improved, the false alarm rate is reduced, and construction safety and continuity are guaranteed.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of shield construction monitoring, and in particular to a shield machine cutter damage monitoring method, system, medium and product. BACKGROUND

[0002] The shield machine is important equipment in tunneling engineering. The cutters on the cutter head of the shield machine are often subjected to harsh working environments as the components directly breaking rocks. The cutters of the shield machine are subjected to great impact and wear during tunneling, and are prone to damage such as chipping and breaking. The cutters are key vulnerable components in the entire device. If the damage of the cutters of the shield machine is not discovered in time, not only will the tunneling efficiency be reduced, but also the cutter head of the shield machine may be damaged again, resulting in huge economic losses and delay of the construction period. Therefore, it is crucial to monitor the working state of the cutters of the shield machine in real time and realize timely damage warning to ensure the safety and efficiency of tunnel engineering.

[0003] In order to realize real-time monitoring of the cutters of the shield machine, a method based on vibration analysis is usually used in the related art. The method installs a vibration sensor near the cutter head of the shield machine to collect vibration signals generated by the cutters during operation. The monitoring system performs frequency spectrum analysis on the collected vibration signals and calculates the total vibration energy in a preset characteristic frequency band. When the total vibration energy in the preset characteristic frequency band exceeds a preset threshold, the system determines that the cutter of the shield machine may be damaged and issues an alarm.

[0004] In actual tunneling operation of the shield machine, the cutter head often encounters sudden changes in geology, such as from soft soil layer to hard rock layer or boulder. Such changes in working conditions will cause the cutters of the shield machine to produce instantaneous and intense impact vibration, resulting in instantaneous increase of the total vibration energy in a preset characteristic frequency band and exceeding of the preset threshold, thereby causing the monitoring system to possibly issue a false alarm and reducing the accuracy of cutter damage monitoring. SUMMARY

[0005] The present application provides a shield machine cutter damage monitoring method, system, medium and product for improving the accuracy of monitoring and reducing the false alarm rate under complex tunneling conditions.

[0006] In a first aspect, the application provides a method for monitoring damage of a disc cutter of a tunnel boring machine, applied to a disc cutter damage monitoring system, the method comprising: obtaining a vibration signal and an acoustic emission signal of the disc cutter of the tunnel boring machine, the acoustic emission signal being a transient elastic wave generated when the disc cutter of the tunnel boring machine contacts and grinds a tunnel; dividing the vibration signal into a plurality of vibration signal segments and dividing the acoustic emission signal into a plurality of acoustic emission signal segments, the vibration signal segments being in a time sequence synchronization relationship with the corresponding acoustic emission signal segments in a rotation period of a cutter head of the tunnel boring machine; extracting a vibration feature vector corresponding to each of the vibration signal segments based on a preset standard working frequency, the preset standard working frequency being a frequency at which a single disc cutter of the tunnel boring machine contacts and grinds a tunnel once completely under a current working condition of the tunnel boring machine, and calculating an Euclidean distance between the vibration feature vectors corresponding to adjacent vibration signal segments; determining an acoustic emission feature parameter corresponding to each of the acoustic emission signal segments according to spectral data of the acoustic emission signal segments, the acoustic emission feature parameter being an amplitude ratio of a main frequency component to a secondary frequency component determined according to the spectral data, and calculating a change rate between the acoustic emission feature parameters corresponding to adjacent acoustic emission signal segments; when the Euclidean distance exceeds a preset Euclidean distance threshold or / and the change rate exceeds a preset change rate threshold, marking and removing the corresponding vibration signal segment and acoustic emission signal as an interference data segment; when the vibration feature vector corresponding to the vibration signal segment exceeds a preset vibration feature vector range, recording the vibration signal segment as a first abnormality mark; when the acoustic emission feature parameter corresponding to the acoustic emission signal segment exceeds a preset acoustic emission feature parameter threshold, recording the acoustic emission signal segment as a second abnormality mark; and if a duration of the first abnormality mark exceeds a preset duration and a number density of the second abnormality mark exceeds a preset density threshold, determining that the disc cutter of the tunnel boring machine is damaged and issuing a warning signal.

[0007] By adopting the above technical solution, the disc cutter damage monitoring system first obtains the vibration signal and the acoustic emission signal of the disc cutter of the tunnel boring machine, and accurately collects information containing the working state of the disc cutter. The system divides the signals into time sequence synchronization segments to ensure data correspondence. The vibration feature vector is extracted based on the preset standard working frequency, and the Euclidean distance is calculated to reflect the vibration change trend; the acoustic emission feature parameter and the change rate are determined according to the spectral data to reflect the characteristics of the acoustic emission signal. After removing the interference data segment, the system marks the abnormal signal by the abnormality mark duration and the number density to determine the disc cutter damage. The system can effectively distinguish the vibration signal and acoustic emission signal mutation caused by instantaneous impact from the signal characteristics generated by the real disc cutter damage, accurately identify the disc cutter damage, and timely issue a warning to improve the accuracy of monitoring and reduce the false positive rate.

[0008] In some embodiments in combination with the first aspect, in some embodiments, when the Euclidean distance exceeds the preset Euclidean distance threshold or / and the rate of change exceeds the preset rate of change threshold, the corresponding vibration signal segment and acoustic emission signal are marked as interference data segments and removed, specifically comprising: inputting two adjacent vibration signal segments and two adjacent acoustic emission signal segments into a time sequence data processing queue; comparing the Euclidean distance with the preset Euclidean distance threshold and comparing the rate of change with the preset rate of change threshold; when the Euclidean distance exceeds the preset Euclidean distance threshold or / and the rate of change exceeds the preset rate of change threshold, determining the vibration signal segment and the acoustic emission signal segment in the time sequence as interference data segments and removing them.

[0009] By using the above technical solution, the adjacent vibration and acoustic emission signal segments are sequentially input into the queue for unified processing. The Euclidean distance is compared with the preset threshold, the rate of change is compared with the preset threshold, and the judgment standard is set. When the Euclidean distance or the rate of change exceeds the threshold, the signal segment in the time sequence is determined as interference data and removed, which can remove invalid data generated by external interference or abnormal working conditions. This process effectively excludes the influence of interference signals generated by instantaneous impact vibration signals and acoustic emission signals on the analysis results through quantitative comparison and threshold judgment, ensures the effectiveness and accuracy of the vibration and acoustic emission signal data used for judging the state of the cutter, avoids misjudgment of cutter damage caused by interference signals, and improves the reliability of the cutter damage monitoring system.

[0010] In some embodiments in combination with the first aspect, in some embodiments, before the vibration signal segment and the acoustic emission signal segment in the time sequence are determined as interference data segments and removed, the method further comprises: obtaining the vibration feature vector corresponding to the vibration signal segment in the time sequence and the acoustic emission feature parameter corresponding to the acoustic emission signal segment; determining whether the vibration feature vector exceeds a preset vibration feature vector range or whether the acoustic emission feature parameter exceeds a preset acoustic emission feature parameter threshold; and if any or all of the determination results are yes, the vibration signal segment and the acoustic emission signal segment in the time sequence are retained.

[0011] By using the above technical solution, the vibration feature vector and the acoustic emission feature parameter of the signal segment in the time sequence are obtained before the signal segment is removed, and the signal information is further mined. The vibration feature vector is determined whether to exceed the range and the acoustic emission feature parameter is determined whether to exceed the threshold, which provides a basis for retaining the signal segment. If any condition is met, the signal segment is retained, which avoids mistakenly deleting the signal reflecting the abnormal state of the cutter due to the Euclidean distance or the rate of change exceeding the threshold. The verification link is added in the process of removing the interference signal, which balances the relationship between removing interference and retaining effective abnormal signals, avoids removing abnormal signals as interference signals, and reduces the accuracy of judging the damage of the cutter.

[0012] In some embodiments of the first aspect, in some embodiments, if the duration of the first abnormal marker exceeds the preset duration and the number density of the second abnormal marker exceeds the preset density threshold, it is determined that the cutter of the shield tunneling machine is damaged and a warning signal is sent, specifically comprising: obtaining all first abnormal markers and second abnormal markers in a to-be-processed signal segment, the to-be-processed signal segment including the vibration signal segment and the acoustic emission signal segment that are not processed after the interference data segment is removed; calculating the duration of the first abnormal marker in the to-be-processed signal segment; counting the number of occurrences of the second abnormal marker in the to-be-processed signal segment, and calculating the number density of the second abnormal marker per unit time; if the duration of the first abnormal marker exceeds the preset duration and the number density of the second abnormal marker exceeds the preset density threshold, it is determined that the cutter of the shield tunneling machine is damaged and a warning signal is sent.

[0013] By adopting the above technical solution, the abnormal markers in the to-be-processed signal segment after the interference data is removed are obtained, and the abnormal information in the effective signal is focused. The duration of the first abnormal marker is calculated, which reflects the continuity of the vibration abnormality; the number of the second abnormal markers is counted and the density is calculated, which reflects the frequency of the acoustic emission abnormality. The cutter damage is judged by the dual conditions of duration and density, and the vibration and acoustic emission signal abnormal conditions are considered comprehensively. Only when the duration and density of the abnormal markers meet the threshold conditions at the same time, it is determined that the cutter is damaged, which avoids misjudgment caused by single abnormal condition or temporary fluctuation, and improves the accuracy and reliability of the cutter damage judgment.

[0014] In some embodiments of the first aspect, in some embodiments, the vibration signal is divided into a plurality of vibration signal segments, and the acoustic emission signal is divided into a plurality of acoustic emission signal segments, specifically comprising: obtaining real-time rotating speed and angle position information of the cutter head of the shield tunneling machine; according to the real-time rotating speed, the rotating period required for the cutter head to complete one complete rotation is calculated; according to the rotating period and the angle position information, the vibration signal is divided into a plurality of equal-length vibration signal segments, the vibration signal segments respectively containing corresponding time stamps; according to the rotating period and the angle position information, the acoustic emission signal is divided into a plurality of equal-length acoustic emission signal segments, the acoustic emission signal segments respectively containing corresponding time stamps, and corresponding to the time stamps of the vibration signal segments.

[0015] By adopting the above technical solution, the real-time rotating speed and angle position information of the cutter head of the shield tunneling machine are obtained, the rotating period is calculated combined with the rotating speed, and a time reference is provided for signal division. According to the rotating period and the angle position, the vibration and acoustic emission signals are divided into equal-length segments and assigned corresponding time stamps, so that the signals are aligned in the time dimension. This division method keeps the vibration signal segments and the acoustic emission signal segments in time sequence synchronization, ensures that the subsequent feature analysis of the two types of signals is based on the corresponding working conditions, and improves the accuracy and relevance of the vibration and acoustic emission signal analysis.

[0016] In some embodiments in combination with the first aspect, based on the preset standard working frequency, the vibration feature vectors corresponding to each vibration signal segment are extracted, and the Euclidean distances between the vibration feature vectors corresponding to adjacent vibration signal segments are calculated, specifically including: based on the preset standard working frequency, the effective vibration signals in each vibration signal segment are extracted; the effective vibration signals are wavelet transformed to obtain effective vibration sub-signals in multiple frequency bands; the energy values of the effective vibration sub-signals in each frequency band are arranged according to the frequency bands to obtain the vibration feature vectors; and the Euclidean distances between the vibration feature vectors corresponding to adjacent vibration signal segments are calculated.

[0017] By using the above technical solution, the effective vibration signals in the vibration signal segment are extracted based on the preset standard working frequency, which can exclude irrelevant frequency interference. The effective vibration signals are wavelet transformed to decompose into effective vibration sub-signals in multiple frequency bands, realizing fine analysis of the signals. The energy values of the sub-signals in each frequency band are arranged to obtain the vibration feature vectors, which quantitatively represent the vibration characteristics. The Euclidean distances between adjacent vibration feature vectors are calculated to quantify the degree of change of the vibration features. This provides an accurate basis for judging whether the vibration signal is abnormal, and further assists in judging whether the cutter working state is stable.

[0018] In some embodiments in combination with the first aspect, the acoustic emission feature parameters corresponding to each acoustic emission signal segment are determined according to the frequency spectrum data corresponding to each acoustic emission signal segment, and the change rates between the acoustic emission feature parameters corresponding to adjacent acoustic emission signal segments are calculated, specifically including: the frequency spectrum data corresponding to each acoustic emission signal segment is extracted; according to the frequency spectrum data, the frequency component with the highest amplitude in the acoustic emission signal is determined as the main frequency component, and the frequency component with the second highest amplitude is determined as the secondary frequency component; the amplitude ratio of the main frequency component to the secondary frequency component is calculated to obtain the acoustic emission feature parameter; and based on the acoustic emission feature parameters arranged in time sequence, the change rates between the acoustic emission feature parameters corresponding to adjacent acoustic emission signal segments are calculated.

[0019] By using the above technical solution, the frequency spectrum data of the acoustic emission signal segment is extracted to analyze the signal from the frequency perspective. The main frequency component and the secondary frequency component are determined to grasp the main frequency characteristics of the signal. The amplitude ratio of the two is calculated to obtain the acoustic emission feature parameter, which represents the frequency characteristics of the acoustic emission signal in a single parameter. Based on the time sequence, the change rates of adjacent acoustic emission feature parameters are calculated to reflect the change trend of the parameters over time. This analysis process from the frequency spectrum data to the feature parameter and then to the change rate converts the complex acoustic emission signal into quantifiable and comparable parameters and change trends, which can effectively identify abnormal changes in the acoustic emission signal and provide a reliable basis for judging the cutter working state based on the acoustic emission signal, which is complementary to the vibration signal analysis.

[0020] In a second aspect, the embodiments of the present application provide a cutter damage monitoring system, comprising: one or more processors and a memory; the memory is coupled with the one or more processors, and the memory is configured to store computer program codes, the computer program codes comprising computer instructions, and the one or more processors invoke the computer instructions to enable the cutter damage monitoring system to perform the method described in the first aspect and any possible implementation manner of the first aspect.

[0021] In a third aspect, the embodiments of the present application provide a computer program product comprising instructions, which, when executed on a cutter damage monitoring system, enable the cutter damage monitoring system to perform the method described in the first aspect and any possible implementation manner of the first aspect.

[0022] In a fourth aspect, the embodiments of the present application provide a computer-readable storage medium comprising instructions, which, when executed on a cutter damage monitoring system, enable the cutter damage monitoring system to perform the method described in the first aspect and any possible implementation manner of the first aspect.

[0023] It can be understood that the cutter damage monitoring system provided in the second aspect, the computer program product provided in the third aspect, and the computer storage medium provided in the fourth aspect are all used to execute the method provided in the embodiments of the present application. Therefore, the beneficial effects that can be achieved are referred to the beneficial effects in the corresponding method, which will not be described here.

[0024] The one or more technical solutions provided in the embodiments of the present application have at least the following technical effects or advantages: 1. Since the technical solutions of acquiring the vibration signal and the acoustic emission signal of the cutter of the shield tunneling machine, dividing the signals, extracting features, removing interference, and judging abnormal marking are adopted, the working state of the cutter can be comprehensively and accurately analyzed, the problem of inaccurate and untimely judgment of cutter damage in the related art is effectively solved, and accurate monitoring of cutter damage under complex tunneling conditions of the shield tunneling machine is realized, thereby reducing the risk of construction interruption or secondary damage of equipment caused by cutter failure.

[0025] 2. Since the technical solutions of inputting adjacent vibration and acoustic emission signal segments into a queue in order, comparing the Euclidean distance and the change rate with the preset threshold value, and removing the time sequence of the signal segment after the threshold value as interference data are adopted, invalid signal data generated by external interference or abnormal working conditions can be effectively removed, the problem of interference signal affecting the accuracy of cutter damage judgment in the related art is effectively solved, the signal data used for analyzing the state of the cutter is ensured to be real and reliable, and the effect of improving the accuracy and reliability of the monitoring system is improved.

[0026] 3. Due to the adoption of a technical solution that calculates the duration of the first abnormal mark and the number density of the second abnormal mark, and uses dual conditions to judge the damage of the roller cutter, it is possible to comprehensively consider the abnormal conditions of vibration and acoustic emission signals, avoid misjudgment caused by a single abnormality or short-term fluctuation, and effectively solve the problem of high false alarm rate of roller cutter damage monitoring in related technologies, thereby achieving accurate judgment of the roller cutter damage status, timely and accurate issuance of early warning signals, and ensuring the normal operation of the shield machine. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] Figure 1 This is a structural diagram of a shield machine cutter head in an embodiment of the present application; Figure 2 This is a flow chart of a shield machine cutter damage monitoring method according to an embodiment of the present application; Figure 3 This is another flow chart of the shield machine cutter damage monitoring method according to an embodiment of the present application; Figure 4 This is a schematic diagram of the structure of a physical device of the hob damage monitoring system in an embodiment of the present application. DETAILED DESCRIPTION

[0028] The terms used in the following examples of the present application are only for the purpose of describing specific embodiments and are not intended to limit the present application. As used in the specification of the present application, the singular expressions "a", "an", "above", "the", and "this" are intended to include plural expressions as well, unless the context clearly indicates otherwise. It should also be understood that the term "and / or" used in the present application refers to any or all possible combinations of one or more of the listed items.

[0029] In the following, the terms "first" and "second" are used for descriptive purposes only and should not be understood to imply or suggest relative importance or implicitly indicate the number of the technical features indicated. Therefore, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. In the description of the embodiments of this application, unless otherwise specified, "plurality" means two or more.

[0030] For ease of understanding, the following Figure 1 This article introduces an application scenario of shield machine cutter. Figure 1 , Figure 1 It is a structural schematic diagram of the shield machine cutter head in the embodiment of the present application.

[0031] The shield machine cutterhead includes: a cutterhead body 101 , a roller cutter 102 , a vibration sensor 103 and an acoustic emission sensor 104 .

[0032] The cutter body 101 is a key component of the cutter head of the shield tunneling machine, continuously rotates in the shield tunneling process, and drives the mounted roller cutter to crush and break the tunnel face. The roller cutter 102 is mounted on the cutter body 101 and is a key component that directly interacts with the rock-soil in the shield tunneling process. The vibration sensor 103 is installed on the cutter body 101 and is used to collect vibration signals. The acoustic emission sensor 104 is installed on the roller cutter 102 and is used to collect acoustic emission signals to obtain signals generated by the roller cutter 102 in real time. The signals collected by the vibration sensor 103 and the acoustic emission sensor 104 are sent to the roller cutter damage monitoring system for processing.

[0033] In the related art, a single vibration signal monitoring method can be used to monitor the damage of the roller cutter by setting a fixed vibration energy threshold. The following describes a scenario using the shield tunneling machine roller cutter damage monitoring method in the related art.

[0034] During the tunneling process, when the cutter body 101 suddenly encounters a hard boulder, the roller cutter 102 in a healthy state will produce a violent impact vibration in the moment of contact with the boulder. The monitoring method of the related art only analyzes the vibration signal, and the monitored vibration energy will instantaneously surge and exceed the preset threshold. Therefore, the method will issue an alarm, incorrectly judging the instantaneous impact experienced by the roller cutter 102 in a healthy state as roller cutter damage. This false alarm caused by non-damage factors reduces the reliability of the monitoring system.

[0035] However, by using the shield tunneling machine roller cutter damage monitoring method provided in the present application, the vibration and acoustic emission dual signals are comprehensively processed and judged to achieve high-precision and low-false-alarm early warning of roller cutter damage. Not only can it identify persistent and developing damage, but it can also effectively exclude the interference of events such as instantaneous impact. The following describes a scenario using the shield tunneling machine roller cutter damage monitoring method in the present application.

[0036] Similarly, in the scenario where the cutter body 101 encounters a hard boulder, the vibration signal corresponding to the roller cutter 102 in a healthy state will produce a short and violent spike, which may cause the corresponding vibration feature vector to exceed the range within 1 to 2 data segments, thus being recorded as a first abnormality marker. However, since the impact is instantaneous, this abnormal state will not last, so its duration will not exceed the preset duration. At the same time, although the acoustic emission signal will also have a transient wave generated by the impact, since the material itself has not been continuously damaged, the frequency of the second abnormality marker is low, and the number density will not exceed the preset density threshold. Finally, since both the duration and the number density do not meet the two warning conditions, the method of the present application will not issue an alarm, successfully avoiding false alarms.

[0037] For the damaged cutter 102: the tiny cracks inside it will continue to expand when normal excavation and impact are encountered. Its vibration eigenvector will continue to exceed the range due to abnormal rolling, causing the first abnormal mark to appear continuously, and its duration is likely to exceed the preset duration. More importantly, each tiny expansion of the crack will release an acoustic emission signal, causing the second abnormal mark to appear densely in this time period, and the number density will significantly exceed the preset density threshold. Finally, since the two warning conditions are met at the same time, the method of the present application accurately determines that the cutter 102 is damaged and issues a warning signal.

[0038] It can be seen that by adopting the cutter damage monitoring method of the shield tunneling machine provided in the present application, the reliable monitoring of the real damage of the cutter is realized, and the problem of false alarm caused by instantaneous impact and other non-damage factors is effectively solved, thereby significantly improving the accuracy and reliability of cutter damage monitoring.

[0039] For ease of understanding, the cutter damage monitoring method provided in the present embodiment can be applied to a cutter damage monitoring system (hereinafter referred to as a monitoring system) for process description in combination with the above scenario. Please refer to Figure 2 , a flowchart of the cutter damage monitoring method of the shield tunneling machine in the present embodiment.

[0040] S201, acquiring the vibration signal and the acoustic emission signal of the cutter of the shield tunneling machine, the acoustic emission signal being a transient elastic wave generated when the cutter contacts and rolls the tunnel; The vibration signal is a signal formed by mechanical vibration generated in the working process of the cutter of the shield tunneling machine due to various forces, which contains relevant information of the running state of the cutter, such as impact vibration signal when the cutter contacts the rock-soil; the acoustic emission signal is a signal formed by transient elastic wave generated inside the material due to local stress concentration, crack propagation, etc. when the cutter of the shield tunneling machine contacts and rolls the tunnel, which can reflect the wear and damage of the cutter, such as specific acoustic emission signal generated when the cutter has cracks.

[0041] During the whole process of tunnel excavation of the shield tunneling machine, the signals need to be continuously acquired as long as the shield tunneling machine is in working state. Specifically, when the shield tunneling machine is running, the vibration signal generated during the working of the cutter is collected by the vibration sensor installed near the cutter of the shield tunneling machine, the vibration sensor converts the mechanical vibration into an electrical signal output and sends it to the monitoring system; at the same time, the acoustic emission signal generated during the process of the cutter contacting and rolling the tunnel is collected by the acoustic emission sensor, the acoustic emission sensor can capture the transient elastic wave generated inside the material due to stress change, etc. and convert it into an electrical signal and send it to the monitoring system. The monitoring system receives the vibration signal and the acoustic emission signal collected by the sensor.

[0042] S202, dividing the vibration signal into a plurality of vibration signal segments, and dividing the acoustic emission signal into a plurality of acoustic emission signal segments, wherein the vibration signal segments are respectively in time-series synchronization with the corresponding acoustic emission signal segments in a rotation period of the shield machine cutterhead; Among them, the vibration signal segment refers to the small signal segments obtained by dividing the continuous vibration signal. Each small signal segment contains the vibration information of the roller cutter within a certain time period; the acoustic emission signal segment refers to the small signal segments obtained by dividing the continuous acoustic emission signal. Each acoustic emission signal segment records the acoustic emission status of the roller cutter within the corresponding time period; the rotation period of the shield machine cutter head is used to represent the time required for the shield machine cutter head to rotate one circle under actual working conditions. During the rotation of the cutter head, the roller cutter will continuously contact and crush the tunnel excavation surface, generating vibration signals and acoustic emission signals.

[0043] After receiving the vibration and acoustic emission signals, the monitoring system segments the continuous signals. Specifically, the collected continuous vibration signal is divided into multiple vibration signal segments in chronological order based on the shield machine cutterhead rotation cycle. Each vibration signal segment corresponds to the time period during which a single shield machine cutter makes complete contact and rolls the tunnel during the cutterhead rotation cycle. Similarly, the continuous acoustic emission signal is divided into multiple acoustic emission signal segments in the same chronological order and corresponding relationship. This ensures that each vibration signal segment and its corresponding acoustic emission signal segment maintain time synchronization over the shield machine cutterhead rotation cycle. In other words, they reflect the working status of the cutterhead at the same stage of the cutterhead rotation, allowing for subsequent comprehensive analysis of the cutterhead status from both vibration and acoustic emission perspectives.

[0044] S203. Extracting vibration eigenvectors corresponding to the vibration signal segments based on a preset standard operating frequency, and calculating the Euclidean distances between the vibration eigenvectors corresponding to adjacent vibration signal segments. The preset standard operating frequency refers to the frequency at which a single shield cutter makes complete contact with and rolls the tunnel under the current shield machine operating condition. Among them, the preset standard operating frequency refers to the frequency when a single shield machine cutter makes complete contact and rolls the tunnel under the current shield machine operating conditions. This frequency is pre-set based on factors such as the shield machine's design parameters and excavation geological conditions. The vibration feature vector represents a set of data vectors obtained by feature extraction of vibration signal segments. It contains key feature information of the vibration signal segments, such as frequency and amplitude, and is used to describe the characteristics of the vibration signal segments. The Euclidean distance is a measurement method for measuring the distance between two vectors. In this step, it is used to calculate the degree of difference between the vibration feature vectors corresponding to adjacent vibration signal segments. The greater the degree of difference, the greater the Euclidean distance. After completing the segmentation of the vibration signal, the monitoring system analyzes the characteristics of the vibration signal segments and the differences between adjacent segments. Specifically, according to the preset standard operating frequency, a feature extraction operation is performed on each vibration signal segment. Parameters that can reflect its essential characteristics are extracted from the vibration signal segment through an algorithm (such as wavelet transform, etc.) to form a corresponding vibration feature vector; then, the Euclidean distance between the vibration feature vectors corresponding to two adjacent vibration signal segments is calculated. For example, for the vibration feature vector V corresponding to the nth vibration signal segment, n The vibration feature vector V corresponding to the n+1th vibration signal segment n+1 ,The distance between them is calculated using the Euclidean distance formula to determine the feature changes between adjacent vibration signal segments.

[0045] S204. Determine, based on the spectral data corresponding to each of the acoustic emission signal segments, an acoustic emission characteristic parameter corresponding to each of the acoustic emission signal segments, and calculate a change rate between the acoustic emission characteristic parameters corresponding to adjacent acoustic emission signal segments, where the acoustic emission characteristic parameter refers to an amplitude ratio of a primary frequency component to a secondary frequency component determined based on the spectral data. Among them, the spectrum data represents the data obtained after converting the acoustic emission signal fragments into the frequency domain through signal processing methods such as Fourier transform, which shows the distribution of the signal on different frequency components; the acoustic emission characteristic parameter refers to the amplitude ratio of the main frequency component to the secondary frequency component determined according to the spectrum data, which is used to quantify the characteristics of the acoustic emission signal and reflect the working status of the hob; the main frequency component represents the frequency component with the highest energy in the spectrum data, which corresponds to the main frequency characteristics of the acoustic emission signal; the secondary frequency component refers to the frequency component with the second highest energy in the spectrum data, which is an important component of the acoustic emission signal; the rate of change is used to indicate the degree of difference between the acoustic emission characteristic parameters corresponding to adjacent acoustic emission signal fragments, which is obtained by calculating the ratio of the difference between the two parameters and the previous parameter.

[0046] After completing the segmentation of the acoustic emission signal, the monitoring system analyzes the characteristic changes of the acoustic emission signal. Specifically, first, a spectrum analysis is performed on each acoustic emission signal segment to obtain spectrum data. The main frequency component and the secondary frequency component are identified from the spectrum data, and their amplitude ratio is calculated to obtain the acoustic emission characteristic parameters. Then, the change rate between the acoustic emission characteristic parameters corresponding to two adjacent acoustic emission signal segments is calculated. For example, for the acoustic emission characteristic parameter P corresponding to the nth acoustic emission signal segment, n The acoustic emission characteristic parameter P corresponding to the n+1th acoustic emission signal segment n+1 , calculate the rate of change = (P n+1 -P n ) / P n The monitoring system can capture the dynamic changes in the characteristics of the acoustic emission signal.

[0047] S205, when the Euclidean distance exceeds the preset Euclidean distance threshold or / and the rate of change exceeds the preset rate of change threshold, the corresponding vibration signal segment and acoustic emission signal are marked as interference data segments and removed; wherein, the preset Euclidean distance threshold refers to a distance limit preset by the monitoring system for judging whether the vibration signal segment is abnormal; the preset rate of change threshold refers to a change limit preset by the monitoring system for judging whether the acoustic emission signal segment is abnormal; the interference data segment refers to the vibration signal segment and the acoustic emission signal segment that do not conform to the characteristics of the normal working state, which may be caused by external interference or equipment abnormality.

[0048] After completing the Euclidean distance calculation of the vibration feature vector and the rate of change calculation of the acoustic emission feature parameter, the monitoring system screens the effective data. Specifically, the monitoring system compares the Euclidean distance corresponding to each vibration signal segment with the preset Euclidean distance threshold, and at the same time compares the rate of change corresponding to each acoustic emission signal segment with the preset rate of change threshold. When the Euclidean distance exceeds the preset Euclidean distance threshold, it indicates that the characteristics of the vibration signal segment are too different from the adjacent segments, which may be abnormal; when the rate of change exceeds the preset rate of change threshold, it indicates that the characteristics of the acoustic emission signal segment change too violently, which may also be abnormal. As long as any of the above conditions is met, the monitoring system will mark the corresponding vibration signal segment and acoustic emission signal segment as interference data segments, and remove the interference data segments and the corresponding vibration signal segments or acoustic emission signal segments from the subsequent analysis data set, to ensure that only normal signal segments participate in the subsequent abnormal detection.

[0049] S206, when the vibration feature vector corresponding to the vibration signal segment exceeds the preset vibration feature vector range, the vibration signal segment is recorded as a first abnormal mark; wherein, the first abnormal mark refers to the record made by the monitoring system on the vibration signal segment that exceeds the preset vibration feature vector range, for identifying the vibration signal segment that may exist abnormality.

[0050] After completing the removal of the interference data segments, the monitoring system preliminarily detects the abnormal signals in the vibration signal segments. Specifically, the monitoring system compares the vibration feature vector corresponding to each remaining vibration signal segment with the preset vibration feature vector range. The preset vibration feature vector range can be determined by statistical analysis of historical normal data or machine learning method, and is usually represented as a hypercube or ellipsoid in a multidimensional space. If the vibration feature vector of a certain vibration signal segment is located outside the range, it indicates that the characteristics of the vibration signal are significantly different from the characteristics under the normal working state, and the monitoring system will record the vibration signal segment as a first abnormal mark.

[0051] S207、When the acoustic emission characteristic parameter corresponding to the acoustic emission signal segment exceeds the preset acoustic emission characteristic parameter threshold, the acoustic emission signal segment is recorded as a second abnormal marker; The second abnormal marker refers to the record made by the monitoring system on the acoustic emission signal segment that exceeds the preset acoustic emission characteristic parameter threshold, and is used to identify the acoustic emission signal segment that may have an abnormality.

[0052] After the interference data segment is removed, the monitoring system detects the abnormality in the acoustic emission signal. Specifically, the monitoring system compares the acoustic emission characteristic parameter corresponding to each remaining acoustic emission signal segment with the preset acoustic emission characteristic parameter threshold. The preset acoustic emission characteristic parameter threshold can be determined by statistical analysis of historical normal data or machine learning method, and represents the reasonable range of the acoustic emission characteristic parameter under normal working conditions. If the acoustic emission characteristic parameter of a certain acoustic emission signal segment exceeds the threshold, it indicates that the characteristic of the acoustic emission signal is significantly different from that under normal working conditions, and the monitoring system will record the acoustic emission signal segment as a second abnormal marker.

[0053] S208、If the duration of the first abnormal marker exceeds the preset duration and the number density of the second abnormal marker exceeds the preset density threshold, it is determined that the TBM cutter is damaged and a warning signal is issued.

[0054] The preset duration refers to a time limit set by the monitoring system, which is used to determine whether the duration of the vibration signal abnormality of the first abnormal marker is long enough. The number density refers to the ratio of the number of second abnormal markers to the length of the window within a certain time window, which is used to measure the frequency of acoustic emission signal abnormalities.

[0055] After the first abnormal marker and the second abnormal marker are recorded, the monitoring system comprehensively judges whether the TBM cutter is damaged and whether a warning is issued. Specifically, the monitoring system first counts the duration of the first abnormal marker, i.e. the time span from the appearance of the first first abnormal marker to the end of the last first abnormal marker. At the same time, the monitoring system calculates the number density of the second abnormal marker within a sliding time window, for example, within a 10-second time window, if there are 5 second abnormal markers, the number density is 0.5 per second. Then, the monitoring system compares the duration of the first abnormal marker with the preset duration, and compares the number density of the second abnormal marker with the preset density threshold. Only when the duration of the first abnormal marker exceeds the preset duration, and the number density of the second abnormal marker exceeds the preset density threshold, the monitoring system will determine that the cutter is damaged, and issue a warning signal. The warning signal can be conveyed to the operator through sound and light alarm, short message notification or monitoring system interface prompt, so as to take measures such as replacing the cutter in time, and avoid further damage to the equipment and delay of the project.

[0056] By adopting the technical scheme, the embodiment captures abnormal changes of the running state of the shield machine cutter from multiple angles through joint analysis of the vibration signal and the acoustic emission signal, effectively improves the accuracy and reliability of the abnormal detection, divides the signal segments based on the rotation period of the cutter head, extracts features, dynamically calculates the difference between adjacent segments, and removes interference data. Combined with threshold analysis of the vibration feature vector and the acoustic emission feature parameter, the cutter damage state is comprehensively judged, and precise monitoring is realized. The safety and continuity of tunnel construction are ensured, and the engineering efficiency and economic benefits are improved.

[0057] In the above embodiment, the application collects vibration signals and acoustic emission signals of the shield machine cutter in the working state, removes interference signals, judges whether the shield machine cutter is damaged, and triggers a warning. In actual application, the process of removing interference signals may cause abnormal signals generated by cutter damage to be misjudged as interference and removed. In some embodiments, a verification mechanism can be introduced in the process of removing interference signals to avoid misdeletion of abnormal signals.

[0058] The scenario of the present embodiment is supplemented as follows. After combining the above scenario, the method provided by the present embodiment is further described in a more specific flow. Please refer to Figure 2 , which is another flowchart of the shield machine cutter damage monitoring method in the present embodiment.

[0059] S301, acquire vibration signals and acoustic emission signals of the shield machine cutter, the acoustic emission signals being transient elastic waves generated when the shield machine cutter contacts and rolls the tunnel; Step S301 is similar to step S201 in the above embodiment, and can refer to the above description, which will not be repeated here.

[0060] S302, acquire real-time rotation speed and angle position information of the cutter head of the shield machine; The real-time rotation speed of the cutter head of the shield machine indicates the rotation speed of the cutter head of the shield machine at the current time, which is usually measured in revolutions per minute (RPM) and used to measure the speed of the cutter head rotation. The angle position information refers to the angle value of the cutter head relative to a reference position during rotation, which is usually measured in degrees (°) and used to determine the specific position of the cutter head on the circumference.

[0061] After obtaining the vibration signal and acoustic emission signal of the tunneling machine cutter, time and position references are provided for subsequent signal segmentation. Specifically, the monitoring system monitors the rotation speed of the cutter head in real time through a rotation speed sensor (such as an optical encoder, a Hall sensor, etc.) installed on the cutter head drive shaft or related transmission components, converts mechanical rotation into electrical signals and processes them to obtain accurate real-time rotation speed data. At the same time, angle encoders and other devices are used to obtain the angular position information of the cutter head, which can accurately indicate the rotation angle of the cutter head at any time relative to the initial reference position.

[0062] S303、According to the real-time rotation speed, calculate the rotation period required for the cutter head to complete one complete rotation; Wherein, the rotation period represents the time required for the cutter head of the tunneling machine to complete one complete 360° rotation, and the unit is second (s), which is inversely proportional to the real-time rotation speed.

[0063] After obtaining the real-time rotation speed of the cutter head, the time window of signal segmentation is determined. Specifically, the monitoring system calculates the time required for the cutter head to complete one complete rotation according to the real-time rotation speed data through the formula: rotation period = 60 / real-time rotation speed. For example, if the real-time rotation speed is 3 RPM, the rotation period is 60 / 3 = 20 seconds. The system will continuously update the rotation period data to adapt to the change of the cutter head rotation speed of the tunneling machine under different working conditions, and ensure the accuracy and real-time performance of subsequent signal segmentation.

[0064] S304、According to the rotation period and the angular position information, the vibration signal is divided into a plurality of equal-length vibration signal segments, and each vibration signal segment includes a corresponding timestamp; According to the rotation period and the angular position information, the acoustic emission signal is divided into a plurality of equal-length acoustic emission signal segments, and each acoustic emission signal segment includes a corresponding timestamp, and the timestamp of each vibration signal segment corresponds one-to-one.

[0065] After calculating the rotation period, signal preprocessing is performed. Specifically, the monitoring system first determines the time length of signal segmentation according to the rotation period, for example, if the rotation period is 20 seconds, the vibration signal and the acoustic emission signal are divided into equal lengths according to the time window of 20 seconds. At the same time, combined with the angular position information, it is ensured that each signal segment corresponds to a specific angular interval of the cutter head on the circumference.

[0066] During the division process, the system adds accurate time stamps to each vibration signal segment and acoustic emission signal segment, which can be calculated based on the start time of signal collection and the position of the segment in the continuous signal. For example, the time stamp of the first signal segment is the start time of collection, and the time stamps of subsequent segments are sequentially increased by the duration of a rotation cycle. In this way, the system achieves strict synchronization of vibration signal segments and acoustic emission signal segments in time, providing an accurate data basis for subsequent feature extraction and anomaly detection.

[0067] S305, based on a preset standard working frequency, extracting an effective vibration signal in each vibration signal segment; The effective vibration signal is subjected to wavelet transform to obtain effective vibration sub-signals of multiple frequency bands; The preset standard working frequency represents the frequency at which a single cutter completes a complete contact and rolls the tunnel under the current working condition of the shield machine, with a unit of hertz (Hz), which is pre-set by factors such as shield machine design parameters and geological conditions. The effective vibration signal refers to the signal component selected from the vibration signal segment that is directly related to the normal working or abnormal state of the cutter, excluding irrelevant interference such as environmental noise. Wavelet transform is a mathematical method of decomposing time-domain signals into sub-signals of different frequency bands, extracting features of signals at different frequencies through multi-resolution analysis, such as decomposing vibration signals into high-frequency (reflecting impact characteristics) and low-frequency (reflecting steady-state vibration characteristics) sub-signals.

[0068] After completing the division of the vibration signal segments, the effective features are extracted from the vibration signal segments. Specifically, the monitoring system first performs band-pass filtering on each vibration signal segment according to the preset standard working frequency, filters out noise signals unrelated to the standard working frequency, and retains the effective vibration signal containing the working characteristics of the cutter.

[0069] Subsequently, the system performs wavelet transform on the effective vibration signal, selects an appropriate wavelet basis function (such as Daubechies wavelet) and decomposition level (such as 3 layers), and decomposes the signal into multiple effective vibration sub-signals of different frequency bands, each sub-signal corresponding to a specific frequency range, such as high-frequency and low-frequency bands obtained by first-level decomposition, and further decomposing the low-frequency band by second-level decomposition, and so on.

[0070] S306, arranging the energy values of the effective vibration sub-signals of each frequency band according to the frequency band to obtain a vibration feature vector; wherein the frequency band represents the frequency interval obtained by decomposing the signal after wavelet transform, such as high-frequency band (100-200 Hz), medium-frequency band (20-100 Hz), etc.; the effective vibration sub-signal refers to the signals of each frequency band decomposed from the effective vibration signal by wavelet transform, each sub-signal containing the vibration energy characteristics of the corresponding frequency band; the energy value is used to represent the energy size of each frequency band signal, which is obtained by calculating the root mean square value or power spectral density of the signal in the frequency band.

[0071] After the effective feature extraction of the vibration signal segment is completed, the corresponding frequency domain feature is converted into a calculable vector form. Specifically, the monitoring system calculates the energy of each effective vibration sub-signal of each frequency band.

[0072] For example, the root mean square energy of the high-frequency sub-signal in the time window is calculated: Where x i is the sampling point value of the sub-signal, and N is the number of sampling points.

[0073] After the calculation is completed, the system arranges the energy values of each frequency band in a vector according to the order from low to high (or from high to low) of the frequency band, forming a vibration feature vector. For example, if the signal is decomposed into three frequency bands: low frequency (0-20Hz), medium frequency (20-50Hz), and high frequency (50-100Hz), then the vibration feature vector is [low frequency energy, medium frequency energy, high frequency energy], which fully reflects the energy distribution characteristics of the vibration signal.

[0074] S307, calculate the Euclidean distance between the vibration feature vectors corresponding to each adjacent vibration signal segment.

[0075] Where adjacent vibration signal segments represent two vibration signal segments that are continuous in time series, such as the nth and (n+1)th segments; the Euclidean distance is used to measure the geometric distance of two vibration feature vectors in multi-dimensional space, reflecting the difference degree of adjacent segment features, and the greater the distance, the more significant the feature change. The calculation formula is: Where v n,i and v n+1,i are the ith components of the nth and (n+1)th feature vectors, respectively, and m is the vector dimension.

[0076] After generating the vibration feature vector, the time sequence change of the vibration signal feature vector is analyzed. Specifically, the monitoring system traverses all vibration signal segments in time sequence, and calculates the Euclidean distance of each pair of adjacent vibration feature vectors.

[0077] S308, extract the frequency spectrum data corresponding to each acoustic emission signal segment; According to the frequency spectrum data, the frequency component with the highest amplitude in the acoustic emission signal is determined as the main frequency component, and the frequency component with the second highest amplitude is determined as the secondary frequency component; The main frequency component refers to the frequency component with the highest amplitude in the frequency spectrum data, representing the main energy concentration frequency band of the acoustic emission signal. For example, if the amplitude of 200Hz frequency is the highest in a certain segment, then 200Hz is the main frequency component of the segment. The secondary frequency component refers to the frequency component with the second highest amplitude in the frequency spectrum data, which is an important auxiliary indicator of signal characteristics. For example, if the amplitude of 150Hz frequency is only second to that of 200Hz, then 150Hz is the secondary frequency component.

[0078] After completing the division of the acoustic emission signal segments, the frequency domain features corresponding to the acoustic emission signal segments are extracted. Specifically, the monitoring system performs fast Fourier transform (FFT) on each acoustic emission signal segment to convert the time domain signal into frequency domain data and generate a frequency spectrum graph. The system traverses all frequency points in the frequency spectrum graph, identifies the frequency point with the highest amplitude as the main frequency component, and then finds the frequency point with the second highest amplitude as the secondary frequency component. For example, in the frequency spectrum graph of a certain acoustic emission signal segment, the amplitude corresponding to 180Hz frequency is 60mV, the amplitude corresponding to 120Hz is 45mV, and the amplitudes of other frequencies are all less than 45mV. Therefore, the main frequency component is 180Hz and the secondary frequency component is 120Hz.

[0079] S309, calculating the amplitude ratio of the main frequency component and the secondary frequency component to obtain an acoustic emission feature parameter; The amplitude ratio represents the ratio of the amplitude of the main frequency component to the amplitude of the secondary frequency component, which is used to quantify the frequency feature distribution of the acoustic emission signal. The acoustic emission feature parameter refers to the signal characteristic quantity represented by the amplitude ratio, which is used to reflect the degree of abnormality of the working state of the cutter.

[0080] After determining the main frequency component and the secondary frequency component, the acoustic emission feature parameter is generated. Specifically, the monitoring system extracts the amplitudes of the main frequency component and the secondary frequency component from the frequency spectrum data, calculates the amplitude ratio, and obtains the acoustic emission feature parameter. For example, if the amplitude of the main frequency component is 50mV and the amplitude of the secondary frequency component is 25mV, then the amplitude ratio is 2.0. The monitoring system generates the corresponding acoustic emission feature parameter for each acoustic emission signal segment and stores them in chronological order.

[0081] S310, based on the acoustic emission feature parameters arranged in time sequence, calculating the change rate between the acoustic emission feature parameters corresponding to adjacent acoustic emission signal segments.

[0082] The time sequence arrangement refers to arranging the acoustic emission feature parameters in the order of the acquisition time of the signal segments to form time series data. After generating the acoustic emission feature parameters, the dynamic changes of the acoustic emission feature parameters are quantified. Specifically, the monitoring system traverses all acoustic emission feature parameters in chronological order, and calculates the change rate for each pair of adjacent parameters. The calculation formula is: P iThe acoustic emission characteristic parameter corresponding to the i-th acoustic emission signal segment.

[0083] S311, input two adjacent vibration signal segments and two adjacent acoustic emission signal segments into a data processing queue in time sequence; After the calculation of the Euclidean distance of the vibration characteristic vector and the calculation of the change rate of the acoustic emission characteristic parameter are completed, the monitoring system takes each group of adjacent vibration signal segments (the n-th and the n+1-th) and their corresponding acoustic emission signal segments (time stamp one-to-one) as a data group, and inputs them into the data processing queue in time sequence.

[0084] S312, compare the Euclidean distance with the preset Euclidean distance threshold value, and compare the change rate with the preset change rate threshold value; After the signal segments are stored in the data processing queue, the abnormal signal segments are preliminarily screened.

[0085] The monitoring system takes out the data of each group of adjacent signal segments from the data processing queue, respectively obtains the Euclidean distance and the change rate of the group. The Euclidean distance is compared with the preset Euclidean distance threshold value, and at the same time, the change rate is compared with the preset change rate threshold value.

[0086] S313, when the Euclidean distance exceeds the preset Euclidean distance threshold value, or / and the change rate exceeds the preset change rate threshold value, the vibration characteristic vector corresponding to the vibration signal segment after the time sequence and the acoustic emission characteristic parameter corresponding to the acoustic emission signal segment are obtained; When the Euclidean distance or the change rate exceeds the corresponding threshold value, the positioning of the abnormal signal segment is performed.

[0087] When the Euclidean distance of a certain group of adjacent signal segments exceeds the preset threshold value, or the change rate exceeds the preset threshold value, the monitoring system determines that the vibration signal segment (the n+1-th) after the time sequence in the group of signals and the corresponding acoustic emission signal segment are likely to be interference data. The vibration characteristic vector of the n+1-th vibration signal segment and the characteristic parameter of the corresponding acoustic emission signal segment are extracted for the next step of judgment.

[0088] S314, judge whether the vibration characteristic vector exceeds the preset vibration characteristic vector range, or whether the acoustic emission characteristic parameter exceeds the preset acoustic emission characteristic parameter threshold value; After the possible abnormal signal segment is located, the subsequent abnormality checking process is entered, so as to avoid directly eliminating the possible real abnormal signal.

[0089] The monitoring system first compares the feature vector of the time-lagged vibration signal segment with the preset vibration feature vector range. For example, if the feature vector is [15, 65, 25] and the lower limit of the low-frequency energy is 20, it is determined that the vibration feature is out of range. At the same time, the system compares the corresponding acoustic emission feature parameter with the preset threshold value, such as a parameter of 2.8 (exceeding the upper limit of 2.5), and determines that the acoustic emission feature is abnormal.

[0090] S315, if any or all of the judgment results are yes, the time-lagged vibration signal segment and acoustic emission signal segment are retained.

[0091] The monitoring system uses "or" logic for judgment, that is, as long as the vibration feature vector is out of range or the acoustic emission feature parameter is out of threshold, it is considered that the signal segment may have abnormality (including interference or real damage).

[0092] If any of the judgment results in step S314 is "yes" (such as the vibration feature is out of range or the acoustic emission parameter is out of threshold), the monitoring system considers that the time-lagged signal segment may reflect the real abnormality (such as wear, crack) of the cutter, rather than simply external interference. At this time, the system retains the vibration signal segment and the acoustic emission signal segment in the temporary storage area and does not perform the rejection operation.

[0093] S316, the time-lagged vibration signal segment and acoustic emission signal segment are determined as interference data segments and are rejected.

[0094] If the judgment results in step S314 are all "no", that is, the vibration feature vector is not out of the preset range, and the acoustic emission feature parameter is also not out of the preset threshold value, the monitoring system determines that the time-lagged signal segment is an interference data segment.

[0095] At this time, the monitoring system rejects the vibration signal segment and the acoustic emission signal segment from the data processing queue and marks it as an interference data segment. For example, if the Euclidean distance of a segment is 6.0 (exceeding the threshold value of 5.0), but the vibration feature vector and the acoustic emission parameter are checked to be within the normal range, the system will determine that the segment is a feature mutation caused by temporary interference of the sensor, and will be rejected. The rejection operation will update the time sequence index of the subsequent signal segment synchronously to ensure data continuity.

[0096] S317, judge whether the vibration feature vector corresponding to the vibration signal segment exceeds the preset vibration feature vector range; after the interference data is removed, the feature anomaly of the vibration signal segment is continuously monitored. Specifically, the monitoring system traverses all vibration signal segments that are not removed, compares the vibration feature vector of each segment with the preset range. For example, the feature vector of a certain vibration signal segment is [15, 65, 25], wherein the low-frequency energy 15 is lower than the preset lower limit 20, and the medium-frequency energy 65 is higher than the preset upper limit 60, so it is determined that the vector exceeds the normal range.

[0097] S318, when the vibration feature vector corresponding to the vibration signal segment exceeds the preset vibration feature vector range, the vibration signal segment is recorded as a first abnormal marker; Steps S318 and S320 are similar to steps S206 and S207 in the above embodiment, and can be referred to the above description, which will not be repeated here.

[0098] S319, judge whether the acoustic emission feature parameter corresponding to the acoustic emission signal segment exceeds the preset acoustic emission feature parameter threshold; After the interference data is removed, the feature anomaly of the acoustic emission signal segment is continuously monitored. Specifically, the monitoring system traverses all acoustic emission signal segments that are not removed, compares the acoustic emission feature parameter of each segment with the preset threshold. For example, the parameter value of a certain segment is 2.8 (exceeding the upper limit 2.5) or 0.5 (lower than the lower limit 0.8), so it is determined to be abnormal. When the parameter is abnormal, the system generates a second abnormal marker, which includes: segment timestamp, parameter specific value (such as 3.0), abnormal direction (upper limit exceeds), etc. S320, when the acoustic emission feature parameter corresponding to the acoustic emission signal segment exceeds the preset acoustic emission feature parameter threshold, the acoustic emission signal segment is recorded as a second abnormal marker; S321, acquire all first abnormal markers and second abnormal markers in the to-be-processed signal segment, wherein the to-be-processed signal segment includes the vibration signal segment and the acoustic emission signal segment that are not processed after the interference data segment is removed; Wherein, the to-be-processed signal segment refers to the vibration signal segment and the acoustic emission signal segment remaining after the interference data is removed, which contains normal signal segments that are not marked as interference and possible abnormal signal segments.

[0099] After the completion of the interference data rejection and the abnormality flag generation, the abnormality flags to be analyzed are summarized. Specifically, the monitoring system first determines the current signal segment time window to be processed, and then extracts all the first abnormality flags and the second abnormality flags in the window from the abnormality database. For example, the system sets the window to be processed as 10 cycles of the cutterhead rotation, and if the current cutterhead rotation cycle is 20 seconds, the window length is 200 seconds, and the system will extract all the abnormality flags in the 200 seconds. The extracted flags include the signal segment timestamp, characteristic parameter value and other information corresponding to each abnormality, so as to calculate the duration and the number density of the abnormality subsequently.

[0100] S322, calculating the duration of the first abnormality flag in the signal segment to be processed; S321, counting the number of the second abnormality flags in the signal segment to be processed, and calculating the number density of the second abnormality flags per unit time; Wherein, the duration of the first abnormality flag refers to the time interval from the appearance of the first first abnormality flag to the end of the last continuous first abnormality flag, for example, the time span of 3 continuous flags is 60 seconds; the preset duration is the time threshold set by the system to judge the duration validity of the abnormality, for example, 30 seconds, and if the duration exceeds the threshold, the abnormality is considered to have duration; the number density of the second abnormality flag refers to the number of the second abnormality flags per unit time, and the calculation formula is number density = number of flags / time window length, for example, 5 flags appear in 10 seconds, and the density is 0.5 times per second; the preset density threshold is the upper limit of the normal abnormality frequency set by the system, for example, 0.3 times per second, and if the frequency exceeds the threshold, the abnormality is considered to occur frequently.

[0101] After obtaining the abnormality flags in the signal segment to be processed, the time characteristics and the frequency characteristics of the abnormal signals are quantified respectively. Specifically, the monitoring system first sorts the first abnormality flags by time, finds the continuous abnormality segments, and calculates the duration. For example, the first abnormality flag starts at 10:30:00 and ends at 10:30:45, with a duration of 45 seconds. If the abnormality flags are not continuous, the duration is calculated in segments. At the same time, the system counts the total number of the second abnormality flags in the window to be processed, and divides the number density by the window duration. For example, the window duration is 50 seconds, and there are 20 second abnormality flags, so the density is 20 / 50=0.4 times per second.

[0102] S323, if the duration of the first abnormality flag exceeds the preset duration and the number density of the second abnormality flag exceeds the preset density threshold, it is determined that the TBM cutter is damaged and a warning signal is sent out; After the duration of the first abnormal mark and the number density of the second abnormal mark are calculated, it is determined whether the cutter is damaged and a warning is triggered. Specifically, the monitoring system simultaneously checks whether the duration of the first abnormal mark exceeds a preset duration and whether the number density of the second abnormal mark exceeds a preset density threshold. For example, if the duration is 45 seconds (exceeding the preset 40 seconds) and the number density is 0.4 times per second (exceeding the preset 0.3 times per second), it is determined that the cutter is damaged, and a warning signal is immediately generated. The warning signal contains information such as cutter number, damage type (such as cutter ring wear, crack), and abnormal start time. If only a single condition is met (such as meeting the duration but not the density), the system continues to accumulate data and re-determines to avoid false alarms. The system improves the accuracy and reliability of cutter damage monitoring through double condition cross-validation.

[0103] S324, if all the determination results are no, the current shield machine cutter working state is normal.

[0104] By adopting the above technical solutions, the embodiment realizes accurate signal segmentation by introducing the cutterhead speed and angle position information, extracts the multi-frequency band features of the vibration signal and the frequency features of the acoustic emission signal by wavelet transform and spectrum analysis respectively. By designing a double verification mechanism, the abnormal signal is first filtered based on the Euclidean distance and the change rate threshold, and then it is determined whether the features exceed the preset range to avoid false deletion of real abnormal signals. At the same time, an intersection verification mechanism is established by combining the duration of the vibration abnormality and the number density of the acoustic emission abnormality to realize more accurate cutter damage monitoring and effectively reduce the false alarm rate.

[0105] The cutter damage monitoring system in the embodiment of the application will be described from the perspective of hardware processing. Please refer to Figure 4 , which is a schematic diagram of an entity device structure of the cutter damage monitoring system in the embodiment of the application.

[0106] It should be noted that Figure 4 The structure of the cutter damage monitoring system shown is only an example and should not impose any limitation on the functions and use range of the embodiment of the application.

[0107] As Figure 4As shown, the electronic device includes a Central Processing Unit (CPU) 401 which can perform various appropriate actions and processes in accordance with a program stored in a Read-Only Memory (ROM) 402 or a program loaded from a storage section 408 into a Random Access Memory (RAM) 403, for example, to execute the methods described in the above-described embodiments. In the Random Access Memory (RAM) 403, various programs and data required for system operation are also stored. The Central Processing Unit (CPU) 401, the Read-Only Memory (ROM) 402, and the Random Access Memory (RAM) 403 are connected to each other through a bus 404. An Input / Output (I / O) interface 405 is also connected to the bus 404.

[0108] Connected to the Input / Output (I / O) interface 405 are an input section 406 including an audio input device, a button switch, and the like; an output section 407 including a display and an audio output device, an indicator, and the like; a storage section 408 including a hard disk and the like; and a communication section 409 including a network interface card such as a LAN (Local Area Network) card, a modem, and the like. The communication section 409 performs communication processing via a network such as the Internet. A drive 410 is also connected to the Input / Output (I / O) interface 405 as necessary. A removable medium 411 such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, and the like is mounted on the drive 410 as necessary, so that a computer program read therefrom is installed in the storage section 408 as necessary.

[0109] In particular, the processes described above with reference to the flow charts can be implemented as computer software programs in accordance with embodiments of the present application. For example, embodiments of the present application include a computer program product comprising a computer program carried on a computer readable medium, the computer program comprising computer programs for executing the methods illustrated by the flow charts. In such embodiments, the computer program can be downloaded and installed from a network via the communication section 409, and / or installed from the detachable medium 411. When the computer program is executed by the central processing unit (CPU) 401, various functions defined in the present application are executed. Note that specific examples of the computer readable storage medium can include, but are not limited to, an electrical connection having one or more wires, a portable computer diskette, a hard disk, a random access memory (RAM), a read only memory (ROM), an erasable programmable read only memory (EPROM), a flash memory, a fiber optical link, a portable compact disc read only memory (CD-ROM), an optical storage device, a magnetic storage device, or any appropriate combination thereof. In this application, the computer readable storage medium can be any tangible medium that contains or stores a program for use by or in connection with an instruction execution system, apparatus, or device.

[0110] The flow charts and block diagrams in the drawings are illustrations of possible architectures, functions, and operations of systems, methods, and computer program products according to various embodiments of the present application. It will be understood that each block of the flow charts or block diagrams, and combinations of blocks in the flow charts or block diagrams, can be implemented by a

[0111] Specifically, the cutter damage monitoring system of the embodiment includes a processor and a memory, and the memory stores a computer program. When the computer program is executed by the processor, the shield tunneling machine cutter damage monitoring method provided by the above-mentioned embodiment is implemented.

[0112] As another aspect, the application also provides a computer readable storage medium, which can be included in the cutter damage monitoring system described in the above embodiments, or can exist independently without being assembled into the cutter damage monitoring system. The storage medium carries one or more computer programs, which, when executed by a processor of the cutter damage monitoring system, enable the cutter damage monitoring system to implement the cutter damage monitoring method of the TBM provided in the above embodiments.

[0113] The above embodiments are only used to illustrate the technical solutions of the present application, but not limit the present application; although the present application has been described in detail with reference to the foregoing embodiments, those ordinarily skilled in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacements to some technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present application.

[0114] In the above embodiments, according to the context, the term "when" can be interpreted as meaning "if" or "after" or "in response to determining" or "in response to detecting". Similarly, according to the context, the phrase "upon determining" or "if detecting (a stated condition or event)" can be interpreted as meaning "if determining" or "in response to determining" or "upon detecting (a stated condition or event)" or "in response to detecting (a stated condition or event)".

[0115] Those ordinarily skilled in the art can understand that all or part of the processes in the above method embodiments can be instructed by a computer program to relevant hardware, and the program can be stored in a computer readable storage medium, and when executed, can include the processes of the above method embodiments. The foregoing storage medium includes ROM or random storage memory RAM, magnetic disc or optical disc and various program code storage media.

Claims

1. A shield machine cutter damage monitoring method, characterized in that: Applied to a hob damage monitoring system, the method comprises: Acquiring a vibration signal and an acoustic emission signal of a shield machine cutter, wherein the acoustic emission signal refers to a transient elastic wave generated when the shield machine cutter contacts and rolls the tunnel; Dividing the vibration signal into a plurality of vibration signal segments, and dividing the acoustic emission signal into a plurality of acoustic emission signal segments, wherein the vibration signal segments are respectively in a time-series synchronization relationship with the corresponding acoustic emission signal segments in a rotation period of the shield machine cutter head; extracting vibration eigenvectors corresponding to the vibration signal segments based on a preset standard operating frequency, and calculating the Euclidean distances between the vibration eigenvectors corresponding to adjacent vibration signal segments, wherein the preset standard operating frequency refers to the frequency at which a single shield cutter makes complete contact with and rolls the tunnel under the current shield machine operating condition; determining, based on the spectrum data corresponding to each of the acoustic emission signal segments, acoustic emission characteristic parameters corresponding to each of the acoustic emission signal segments, and calculating a rate of change between the acoustic emission characteristic parameters corresponding to adjacent acoustic emission signal segments, wherein the acoustic emission characteristic parameter refers to an amplitude ratio of a primary frequency component to a secondary frequency component determined based on the spectrum data; When the Euclidean distance exceeds a preset Euclidean distance threshold or / and the change rate exceeds a preset change rate threshold, the corresponding vibration signal segment and acoustic emission signal are marked as interference data segments and removed; When the vibration feature vector corresponding to the vibration signal segment exceeds a preset vibration feature vector range, recording the vibration signal segment as a first abnormality mark; When the acoustic emission characteristic parameter corresponding to the acoustic emission signal segment exceeds a preset acoustic emission characteristic parameter threshold, recording the acoustic emission signal segment as a second abnormality mark; If the duration of the first abnormal mark exceeds a preset duration and the number density of the second abnormal mark exceeds a preset density threshold, it is determined that the shield machine cutter is damaged and an early warning signal is issued.

2. The method according to claim 1, characterized in that When the Euclidean distance exceeds a preset Euclidean distance threshold or / and the change rate exceeds a preset change rate threshold, marking the corresponding vibration signal segment and acoustic emission signal as interference data segments and removing them specifically includes: Input two adjacent vibration signal segments and two adjacent acoustic emission signal segments into a data processing queue in time sequence; Comparing the Euclidean distance with the preset Euclidean distance threshold, and comparing the change rate with the preset change rate threshold; When the Euclidean distance exceeds the preset Euclidean distance threshold, or / and the change rate exceeds the preset change rate threshold, the vibration signal segment and acoustic emission signal segment that are later in time sequence are determined to be interference data segments and are removed.

3. The method according to claim 2, characterized in that Before determining the vibration signal segment and acoustic emission signal segment that are later in time sequence as interference data segments and removing them, the method further includes: Acquire the vibration feature vector corresponding to the vibration signal segment that is later in the time sequence, and the acoustic emission feature parameter corresponding to the acoustic emission signal segment; Determining whether the vibration characteristic vector exceeds a preset vibration characteristic vector range, or whether the acoustic emission characteristic parameter exceeds a preset acoustic emission characteristic parameter threshold; If any or all of the judgment results are yes, the vibration signal segment and acoustic emission signal segment with the later time sequence are retained.

4. The method according to claim 1, wherein If the duration of the first abnormal mark exceeds a preset duration and the number density of the second abnormal mark exceeds a preset density threshold, determining that the shield machine cutter is damaged and issuing a warning signal specifically includes: Acquire all first abnormality marks and second abnormality marks in a signal segment to be processed, wherein the signal segment to be processed includes the vibration signal segment and the acoustic emission signal segment that have not been processed after the interference data segment is removed; Calculating the duration of the first abnormal mark in the signal segment to be processed; Counting the number of occurrences of the second abnormal marker in the signal segment to be processed, and calculating the number density of the second abnormal marker per unit time; If the duration of the first abnormal mark exceeds a preset duration and the number density of the second abnormal mark exceeds a preset density threshold, it is determined that the shield machine cutter is damaged and an early warning signal is issued.

5. The method according to claim 1, wherein The step of dividing the vibration signal into a plurality of vibration signal segments and dividing the acoustic emission signal into a plurality of acoustic emission signal segments specifically includes: Obtaining real-time rotation speed and angular position information of the shield machine cutterhead; Calculating the rotation period required for the cutter head to complete one full rotation based on the real-time rotation speed; Dividing the vibration signal into a plurality of vibration signal segments of equal length according to the rotation period and angular position information, wherein each vibration signal segment includes a corresponding time stamp; According to the rotation period and angular position information, the acoustic emission signal is divided into a plurality of acoustic emission signal segments of equal length, each of which includes a corresponding time stamp and has a one-to-one correspondence with the time stamp of the vibration signal segment.

6. The method according to claim 1, characterized in that The extracting, based on the preset standard operating frequency, the vibration feature vectors corresponding to the respective vibration signal segments, and calculating the Euclidean distances between the vibration feature vectors corresponding to adjacent vibration signal segments specifically include: extracting effective vibration signals from each of the vibration signal segments based on a preset standard operating frequency; Performing wavelet transform on the effective vibration signal to obtain effective vibration sub-signals in multiple frequency bands; Arranging the energy values ​​of the effective vibration sub-signals in each frequency band according to the frequency band to obtain a vibration feature vector; Calculate the Euclidean distance between the vibration feature vectors corresponding to adjacent vibration signal segments.

7. The method according to claim 1, characterized in that The determining, based on the spectrum data corresponding to each of the acoustic emission signal segments, the acoustic emission characteristic parameters corresponding to each of the acoustic emission signal segments, and calculating the change rate between the acoustic emission characteristic parameters corresponding to adjacent acoustic emission signal segments specifically includes: Extracting spectrum data corresponding to each of the acoustic emission signal segments; Determining, based on the spectrum data, a frequency component with the highest amplitude in the acoustic emission signal as a primary frequency component and a frequency component with the second highest amplitude as a secondary frequency component; Calculating the amplitude ratio of the primary frequency component to the secondary frequency component to obtain acoustic emission characteristic parameters; Based on the acoustic emission characteristic parameters arranged in time sequence, the change rates between the acoustic emission characteristic parameters corresponding to adjacent acoustic emission signal segments are calculated.

8. A cutter damage monitoring system, characterized in that: The roller cutter damage monitoring system includes: one or more processors and a memory; the memory is coupled to the one or more processors, the memory is used to store computer program code, the computer program code includes computer instructions, and the one or more processors call the computer instructions to enable the roller cutter damage monitoring system to execute the method according to any one of claims 1 to 7.

9. A computer-readable storage medium comprising instructions, characterized in that: When the instructions are executed on a reel cutter damage monitoring system, the reel cutter damage monitoring system is caused to execute the method according to any one of claims 1 to 7.

10. A computer program product, characterized in that When the computer program product is run on a roller cutter damage monitoring system, the roller cutter damage monitoring system is enabled to perform the method according to any one of claims 1 to 7.

Images