A method and system for monitoring and analyzing cutterhead wear of shield machine during tunneling

Abstract

The present invention discloses a method and system for monitoring and analyzing the wear of the cutter disc of a shield machine during tunneling, and mainly relates to the technical field of shield machine engineering. It includes: an array of monitoring sensors interactively arranged on the surface of the cutter disc of the shield machine and on the cutter, extracting the monitoring data of the preset monitoring window, and obtaining a monitoring data array sequence; obtaining a set of cutter disc operation characteristics; obtaining working soil layer state characteristics; obtaining a cutter surface acquisition image; performing a cutter edge comparison between the cutter standard image and the cutter surface acquisition image to obtain a cutter comparison result, and using the cutter comparison result as a cutter disc wear analysis result. The beneficial effect of the present invention is that it solves the technical problems in the prior art of low mastery of the cutter disc stability situation implied in the monitoring data and poor monitoring and analysis reliability when analyzing the wear state of the cutter disc of a shield machine during tunneling, and achieves the technical effect of improving the accuracy of the cutter disc wear monitoring and analysis of the shield machine.

Description

Technical Field

[0001] The invention relates to the technical field of shield machine engineering, and in particular to a method and system for monitoring and analyzing cutter head wear of a shield machine during tunneling. Background Art

[0002] The shield machine is a widely used equipment in modern underground tunnel construction. Its cutterhead is one of the most core parts, responsible for excavating the soil. During the long-term operation of the shield machine, the wear of the cutterhead and the cutter directly affects the excavation efficiency and the working stability of the shield machine. Therefore, real-time monitoring and analysis of the cutterhead wear is crucial to improving the operating efficiency of the shield machine and extending the service life of the equipment. Traditional wear monitoring methods usually rely on manual inspection or external sensor detection, but these methods cannot accurately and comprehensively capture the wear information of the cutterhead and the cutter.

[0003] The existing technology has the technical problems of low understanding of the cutterhead stability conditions implicit in the monitoring data and poor reliability of monitoring and analysis when analyzing the wear status of the cutterhead of a shield machine during excavation. Summary of the invention

[0004] The present application provides a method and system for monitoring and analyzing the cutter disc wear of a shield machine during excavation, which is used to solve the technical problems in the prior art of analyzing the wear status of the cutter disc of a shield machine during excavation, such as low understanding of the cutter disc stability conditions implied in the monitoring data and poor monitoring and analysis reliability.

[0005] In view of the above problems, the present application provides a method and system for monitoring and analyzing cutter head wear of a shield machine during tunneling.

[0006] In a first aspect of the present application, a method for monitoring and analyzing cutter head wear of a shield machine during tunneling is provided, the method comprising:

[0007] The monitoring sensor arrays interactively arranged on the surface of the shield machine cutter head and the cutter are used to extract the monitoring data of a preset monitoring window to obtain a monitoring data array sequence, wherein each monitoring data array includes the monitoring data of the monitoring sensor array at a monitoring time point of the preset monitoring window;

[0008] According to the order of monitoring time points in the preset monitoring window from front to back, the cutter head data interactive analysis is performed on the monitoring data array sequence, and the cutter head data interactive analysis result is analyzed for the cutter head operation feature to obtain a cutter head operation feature set;

[0009] Calling a high-definition camera to collect images of the working soil layer and soil layer broken particles, obtaining a soil layer image and a soil layer broken particle image, traversing the soil layer image and the soil layer broken particle image to perform working soil layer state analysis, and obtaining working soil layer state characteristics, wherein the high-definition camera is arranged in the peripheral protection area of ​​the shield machine cutter head;

[0010] Performing a cutterhead operation stability analysis according to the cutterhead operation feature set and the working soil layer state feature, activating a coolant spray device to clean the surface of the shield machine cutterhead when the cutterhead operation stability result is lower than a preset threshold, and using the high-definition camera to capture an image of the tool surface of the cleaned shield machine cutterhead to obtain a tool surface capture image;

[0011] The tool standard image of the shield machine cutter disc is called, and the tool edge is compared between the tool standard image and the tool surface acquisition image to obtain a tool comparison result, and the tool comparison result is used as a cutter disc wear analysis result, wherein the acquisition position of the tool standard image and the tool surface acquisition image is the same.

[0012] The second aspect of the present application provides a shield machine cutter head wear monitoring and analysis system during tunneling, the system comprising:

[0013] A monitoring data array sequence acquisition module is used to interactively arrange the monitoring sensor arrays on the surface of the shield machine cutter head and the cutter, extract the monitoring data of the preset monitoring window, and obtain the monitoring data array sequence, wherein each monitoring data array includes the monitoring data of the monitoring sensor array at a monitoring time point of the preset monitoring window;

[0014] a cutter disc operation feature set acquisition module, configured to perform a cutter disc data interactive analysis on the monitoring data array sequence according to the order of monitoring time points from front to back in the preset monitoring window, and perform a cutter disc operation feature analysis on the cutter disc data interactive analysis result to obtain a cutter disc operation feature set;

[0015] A working soil layer state feature acquisition module is used to call a high-definition camera to collect images of the working soil layer and soil layer broken particles, obtain a soil layer image and a soil layer broken particle image, traverse the soil layer image and the soil layer broken particle image to perform working soil layer state analysis, and obtain working soil layer state features, wherein the high-definition camera is arranged in the peripheral protection area of ​​the shield machine cutter head;

[0016] a tool surface acquisition image acquisition module, which is used to perform a tool disc operation stability analysis based on the tool disc operation feature set and the working soil layer state feature; when the tool disc operation stability result is lower than a preset threshold, activate the coolant spraying device to clean the surface of the shield machine tool disc, and use the high-definition camera to acquire an image of the tool surface of the cleaned shield machine tool disc to obtain a tool surface acquisition image;

[0017] The cutterhead wear analysis result acquisition module is used to call the tool standard image of the shield machine cutterhead, compare the tool edge of the tool standard image with the tool surface acquisition image, obtain the tool comparison result, and use the tool comparison result as the cutterhead wear analysis result, wherein the acquisition position of the tool standard image and the tool surface acquisition image is the same.

[0018] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0019] The present application extracts monitoring data of a preset monitoring window through a monitoring sensor array interactively arranged on the surface of the cutter head and the cutter of a shield machine, and obtains a monitoring data array sequence, wherein each monitoring data array includes monitoring data of the monitoring sensor array at a monitoring time point in the preset monitoring window, and then performs cutter head data interactive analysis on the monitoring data array sequence in the order of the monitoring time points in the preset monitoring window from front to back, and performs cutter head operation feature analysis on the cutter head data interactive analysis result to obtain a cutter head operation feature set, and then calls a high-definition camera to collect images of the working soil layer and soil layer broken particles to obtain soil layer images and soil layer broken particle images, and traverses the soil layer images and soil layer broken particle images to perform working soil layer state analysis. Analysis is performed to obtain the working soil state characteristics, wherein a high-definition camera is arranged in the outer protection area of ​​the shield machine cutter disc, and then the cutter disc operation stability analysis is performed based on the cutter disc operation characteristic set and the working soil state characteristics. When the cutter disc operation stability result is lower than the preset threshold, the coolant injection device is activated to clean the surface of the shield machine cutter disc, and the high-definition camera is used to collect images of the tool surface of the cleaned shield machine cutter disc to obtain the tool surface collection image, and the tool standard image of the shield machine cutter disc is called, and the tool standard image is compared with the tool surface collection image to obtain the tool comparison result, and the tool comparison result is used as the cutter disc wear analysis result, wherein the collection position of the tool standard image and the tool surface collection image is the same. The technical effect of improving the reliability and accuracy of the cutter disc wear analysis is achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Attached Figure 1 It is a schematic flow chart of a method for monitoring and analyzing cutter head wear of a shield machine during tunneling provided by an embodiment of the present invention.

[0021] Attached Figure 2 It is a schematic diagram of the structure of a cutterhead wear monitoring and analysis system for a shield machine during tunneling provided by an embodiment of the present invention.

[0022] Numbers shown in the accompanying drawings:

[0023] Monitoring data array sequence acquisition module 11, cutterhead operation feature set acquisition module 12, working soil layer state feature acquisition module 13, tool surface acquisition image acquisition module 14, cutterhead wear analysis result acquisition module 15. DETAILED DESCRIPTION

[0024] The present invention is further described below in conjunction with specific embodiments. It should be understood that these embodiments are only used to illustrate the present invention and are not used to limit the scope of the present invention. In addition, it should be understood that after reading the content taught by the present invention, those skilled in the art can make various changes or modifications to the present invention, and these equivalent forms also fall within the scope defined by the claims attached to the present application. It should be noted that the terms "including" and "having" are intended to cover non-exclusive inclusions, for example, a process, method, system, product or server comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or modules that are not clearly listed or inherent to these processes, methods, products or devices.

[0025] Embodiment 1, as attached Figure 1 As shown, the present application provides a method for monitoring and analyzing cutter head wear of a shield machine during tunneling, wherein the method comprises:

[0026] Step S100: interactively deploying a monitoring sensor array on the surface of the shield machine cutter head and the cutter, extracting monitoring data of a preset monitoring window, and obtaining a monitoring data array sequence, wherein each monitoring data array includes monitoring data of the monitoring sensor array at a monitoring time point of the preset monitoring window;

[0027] In one possible embodiment, the monitoring sensor array is an array composed of multiple sensors, which is used to monitor different types of physical quantities on the surface of the cutter disc and the tool, such as temperature, vibration, pressure, strain, etc., to capture the working status and wear of the cutter disc. The preset monitoring window is a time range preset by a technician in this field. For example, this can be the working cycle of the shield machine within a specific time, which is usually set according to the operating status and task requirements of the shield machine. The monitoring data array sequence is a monitoring sensor that monitors the surface of the shield machine cutter disc and the state of the tool. It is a data sequence composed of monitoring data extracted from multiple time nodes. Each monitoring data array represents the monitoring data of the monitoring sensor array at a monitoring time point in the preset monitoring window. This data sequence can help analyze the dynamic changes of the cutter disc.

[0028] Preferably, sensors are arranged on the surface of the shield machine cutter disc and the cutter, and these sensors work in coordination with each other to collect data in real time to form a monitoring sensor array. The monitoring sensor array is used to collect the working status data of the cutter disc and the cutter in real time. Each sensor array collects data in a preset monitoring window, and the monitoring data at all these time points together constitute a data sequence. In this way, real-time and continuous monitoring of the shield machine cutter disc can be achieved. The collected monitoring data array sequence reflects the operating status of the cutter disc at different time points, providing a rich and continuous data basis for the subsequent cutter disc stability analysis. The role of this step is to ensure the accuracy and continuity of data acquisition, provide reliable data support for subsequent analysis, and thus improve the reliability of the entire monitoring process.

[0029] Step S200: performing a cutter head data interaction analysis on the monitoring data array sequence according to the order of monitoring time points in the preset monitoring window from the front to the back, and performing a cutter head operation feature analysis on the cutter head data interaction analysis result to obtain a cutter head operation feature set;

[0030] Further, according to the order of monitoring time points in the preset monitoring window from front to back, the cutter head data interaction analysis is performed on the monitoring data array sequence, and the cutter head data interaction analysis result is analyzed for the cutter head operation feature set to obtain the cutter head operation feature set. Step S200 of the embodiment of the present application further includes:

[0031] Performing data interaction analysis on the first monitoring data array and the second monitoring data array of the monitoring data array sequence to obtain a first interactive cutterhead monitoring data array;

[0032] Performing data interaction analysis on the first interactive cutter head monitoring data array and the third monitoring data array of the monitoring data array sequence to obtain a second interactive cutter head monitoring data array;

[0033] Perform data interaction analysis on the N-2th interactive tool disk monitoring data array and the Nth monitoring data array of the monitoring data array sequence to obtain the N-1th interactive tool disk monitoring data array, and use the N-1th interactive tool disk monitoring data array as the tool disk data interaction analysis result, wherein N is the total number of monitoring data arrays in the monitoring data array sequence, and N is an integer greater than or equal to 1.

[0034] Furthermore, step S200 of the embodiment of the present application further includes:

[0035] Performing inner product mapping identification on the first monitoring data array and the second monitoring data array of the monitoring data array sequence to obtain an inner product mapping calculation result array;

[0036] Traversing the inner product mapping calculation result array to perform normalization processing to obtain an inner product mapping normalized calculation result array;

[0037] Interactive convolution calculation is performed on the inner product mapping normalized calculation result array and the first monitoring data array to obtain a first interactive cutter head monitoring data array.

[0038] In the embodiments of the present application, interactive analysis of cutter disc data refers to revealing key information such as the dynamic behavior pattern and wear trend of the cutter disc by comparing and interactively analyzing the monitoring data arrays at different time points. Interactive analysis helps to reveal the correlation between time-related change trends and the state of the cutter disc. The cutter disc operation feature set is a multi-dimensional set of the operating state of the cutter disc obtained by interactive analysis of multiple monitoring data arrays, including the stability, wear degree, work efficiency, etc. of the cutter disc. These feature sets can be used to evaluate the current operating state and potential faults of the cutter disc.

[0039] First, the monitoring data array is interactively analyzed in order from the front to the back of the preset monitoring window time points. This analysis can not only reveal the independent data of each monitoring time point, but also reveal the dynamic relationship between them, help discover the change trend of the cutterhead at different time points, and gradually obtain a higher level of interactive cutterhead monitoring data array. Through the mutual correlation and interactive analysis, a set containing the cutterhead operation characteristics is formed. This set provides an important basis for the subsequent cutterhead stability analysis.

[0040] The data similarity calculation of the first monitoring data array and the second monitoring data array of the monitoring data array sequence at the same monitoring position is performed using the cosine similarity calculation formula, and the inner product mapping calculation result array is obtained according to the calculation result. The inner product mapping calculation result array reflects the data similarity between the first monitoring data array and the second monitoring data array. By obtaining the inner product mapping calculation result, it is convenient to identify the wear pattern of the cutter disc and the operation characteristics of the cutter disc at different time points, so as to judge whether there is a potential problem with the cutter disc.

[0041] The inner product mapping calculation result array is normalized using the minimum-maximum normalization formula to maintain the consistency of data analysis and obtain the inner product mapping normalized calculation result array. Further, the inner product mapping normalized calculation result array and the first monitoring data array are interactively convolved using a graph convolutional network to analyze the interactive relationship between the data and obtain the first interactive cutter head monitoring data array. The first interactive cutter head monitoring data array reflects the implicit relationship between the data changes from the first monitoring data array to the second monitoring data array.

[0042] Based on the same calculation principle as that for obtaining the first interactive cutter disc monitoring data array, the first interactive cutter disc monitoring data array and the third monitoring data array of the monitoring data array sequence are subjected to data interaction analysis to obtain the second interactive cutter disc monitoring data array. Furthermore, based on the same principle, the N-2th interactive cutter disc monitoring data array and the Nth monitoring data array of the monitoring data array sequence are subjected to data interaction analysis to obtain the N-1th interactive cutter disc monitoring data array. Among them, the N-1th interactive cutter disc monitoring data array reflects the implicit relationship of the data changes of the entire monitoring data array sequence, and provides reliable data support for subsequent wear monitoring analysis. Furthermore, the N-1th interactive cutter disc monitoring data array is used as the cutter disc data interaction analysis result.

[0043] Preferably, multiple sample analysis results and multiple sample cutterhead operation feature sets are obtained as feature analysis training data, and the feature analysis training data is used to perform supervised training on a framework built based on a convolutional neural network to learn the mapping relationship between the analysis results and the cutterhead operation feature set until the training converges, thereby obtaining a trained feature extractor. The feature extractor is used to perform cutterhead operation feature analysis on the cutterhead data interaction analysis results to obtain the cutterhead operation feature set. Through the systematic data interaction analysis, the operation characteristics of the cutterhead are excavated, thereby achieving the technical effect of laying the foundation for subsequent stability analysis and providing a reliable data basis for the maintenance and operation of the shield machine.

[0044] Step S300: calling a high-definition camera to collect images of the working soil layer and soil layer broken particles, obtaining a soil layer image and a soil layer broken particle image, traversing the soil layer image and the soil layer broken particle image to perform working soil layer state analysis, and obtaining working soil layer state characteristics, wherein the high-definition camera is arranged in the peripheral protection area of ​​the shield machine cutter head;

[0045] Furthermore, a soil layer state analysis network layer is pre-constructed, and the soil layer image and the soil layer broken particle image are analyzed and identified using the soil layer state analysis network layer to obtain the working soil layer state characteristics.

[0046] In one embodiment, the high-definition camera is used to collect high-quality images of soil layers and soil layer broken particles to help analyze the state of the soil layer. The working soil layer is the soil or rock layer that the cutter disc contacts during the operation of the shield machine, and is usually divided according to the type, humidity, hardness and other properties of the soil layer. The state of the soil layer has a direct impact on the wear of the cutter disc, the soil layer cutting effect, etc. The soil layer broken particles are granular substances formed after the soil layer is broken during the operation of the shield machine cutter disc. The size, shape and distribution of the broken particles can reflect the hardness and soil condition of the soil layer. The soil layer state analysis network layer is a pre-built neural network or deep learning model that is specifically used to analyze and identify features in soil layer images and broken particle images, thereby extracting the working state characteristics of the soil layer.

[0047] Preferably, images of the working soil layer and broken particles are collected by a high-definition camera to analyze the state of the soil layer. In this process, the camera is installed in the peripheral protection area of ​​the shield machine cutter disc to ensure that key image information of the contact with the soil layer during the operation of the cutter disc can be captured. By traversing and analyzing the soil layer image and the broken particle image, the surface state of the soil layer after the shield machine cutter disc is working, such as uniformity, compaction, etc., can be extracted, as well as the state of the broken particles of the soil layer. Furthermore, through the pre-constructed soil layer state analysis network layer (usually an image recognition network based on deep learning), the collected images are intelligently analyzed and identified, and accurate soil layer working state characteristics are finally obtained. These features not only help to judge the specific condition of the soil layer, but also provide valuable auxiliary information for the wear analysis of the cutter disc, thereby improving the prediction accuracy of the influence of the soil layer on the cutter disc during the operation of the shield machine.

[0048] Preferably, multiple sample soil layer images and multiple sample soil layer broken particle images, as well as corresponding working soil layer state characteristics are obtained as state analysis training data, and the state analysis training data are used to train the framework based on the construction of the convolutional neural network. During the training, the back propagation algorithm is used to optimize the network parameters so that the network can accurately predict the soil layer state characteristics according to the input image until the training converges, thereby obtaining a trained soil layer state analysis network layer.

[0049] The soil state analysis network layer after training and optimization can be applied to the actual soil state analysis, automatically analyzing the soil state and extracting relevant features, thereby providing real-time feedback data support for the working quality of the shield machine cutter head, achieving the technical effect of providing auxiliary data for the subsequent analysis of the cutter head wear state.

[0050] Step S400: performing a cutterhead operation stability analysis according to the cutterhead operation feature set and the working soil layer state feature, and when the cutterhead operation stability result is lower than a preset threshold, activating a coolant spraying device to clean the surface of the shield machine cutterhead, and using the high-definition camera to capture an image of the tool surface of the cleaned shield machine cutterhead to obtain a tool surface capture image;

[0051] In one embodiment, the operation stability of the cutter disc is evaluated by comprehensively analyzing the operation characteristics of the cutter disc and the state characteristics of the working soil layer. If the analysis results show that the stability of the cutter disc is lower than the preset threshold (for example, abnormal wear or unstable operation), the coolant injection device (including coolant source, injection device, control device, etc.) is started to clean the surface of the cutter disc to facilitate subsequent image acquisition. After cleaning, the surface image of the tool is collected again by a high-definition camera, and the state of the tool on the cutter disc is further evaluated by analyzing these images. This process can not only monitor the wear of the cutter disc in real time, but also detect potential problems in an automated manner in a timely manner to avoid affecting the construction efficiency due to excessive wear of the cutter disc. If the cutter disc operation stability result is not lower than the preset threshold, it indicates that the wear of the shield machine cutter disc is within an acceptable range, and no further tool evaluation is required at this time.

[0052] Preferably, the cutterhead operation feature set and the working soil layer state feature are weightedly calculated according to a preset weight value to obtain the cutterhead operation stability result, wherein the preset threshold is the minimum cutterhead operation stability result set by a person skilled in the art.

[0053] Step S500: calling the standard tool image of the shield machine cutter disc, performing tool edge comparison between the standard tool image and the tool surface acquisition image to obtain a tool comparison result, and using the tool comparison result as a cutter disc wear analysis result, wherein the acquisition positions of the standard tool image and the tool surface acquisition image are the same.

[0054] Further, calling the tool standard image of the shield machine cutter head, comparing the tool edge of the tool standard image with the tool surface acquisition image, obtaining the tool comparison result, and using the tool comparison result as the cutter head wear analysis result, step S500 of the embodiment of the present application also includes:

[0055] Mapping the tool surface acquisition image to the tool standard image, calculating edge distance difference according to a preset edge sampling point position set, and obtaining an edge distance difference set;

[0056] Based on the size of the edge distance difference and the position of the edge distance difference, the edge distance difference set is screened for the edge distance difference center value to determine the edge distance difference center value set;

[0057] Taking the edge distance difference center value set as the starting point, performing center iteration on the edge distance difference set, when the number of center iterations meets the preset number of iterations, obtaining the edge distance difference center neighborhood set according to the minimum value of the cost loss function in the iteration process, wherein each edge distance difference center neighborhood corresponds to an edge distance difference center;

[0058] Traversing the edge distance difference center neighborhood set to perform mean calculation to obtain the edge distance difference center neighborhood mean set;

[0059] The tool comparison result is obtained by performing weighted calculation based on the edge distance difference center neighborhood mean set.

[0060] Further, based on the size of the edge distance difference and the position of the edge distance difference, the edge distance difference set is screened for the edge distance difference center value to determine the edge distance difference center value set. In the embodiment of the present application, step S500 further includes:

[0061] Randomly extracting a preset number of edge distance difference values ​​from the edge distance difference value set as an initial edge distance difference value center value set;

[0062] Performing central value authentication on the initial edge distance difference central value set from two dimensions, namely, the size of the edge distance difference and the position of the edge distance difference. If the authentication is passed, taking the initial edge distance difference central value set as the edge distance difference central value set;

[0063] If the authentication fails, the edge distance difference center value in the edge distance difference set is screened again.

[0064] Further, taking the edge distance difference center value set as the starting point, performing center iteration on the edge distance difference set, when the number of center iterations meets the preset number of iterations, obtaining the edge distance difference center neighborhood set according to the minimum value of the cost loss function in the iteration process, step S500 of the embodiment of the present application further includes:

[0065] Calculate the difference between each edge distance difference in the edge distance difference set and the edge distance difference center value set respectively, assign each edge distance difference to the neighborhood corresponding to the edge distance difference center value with the smallest difference, and obtain the edge distance difference starting center neighborhood set;

[0066] Considering the size and position of the edge distance difference, a cost loss function is constructed, wherein the cost loss function is:

[0067] ;

[0068] in, is the cost loss function value, is the edge distance difference starting center neighborhood set The edge distance difference starts from the center of the neighborhood The edge distance difference, is the edge distance difference starting center neighborhood set The edge distance difference starts from the center of the neighborhood The edge distance difference position corresponding to the edge distance difference, is the edge distance difference center value set The center value of the edge distance difference, is the edge distance difference center value set The edge distance difference position corresponding to the edge distance difference center value, is the weight of the edge distance difference and the edge distance difference position, For the The total number of edge distance differences contained in the neighborhood of the starting center of the edge distance difference, is the number of central neighbors of the starting central neighborhood set of edge distance difference, is an integer greater than or equal to 1;

[0069] Calling a cost loss function to perform a cost analysis on the starting center neighborhood set of the edge distance difference to obtain a starting cost loss function value;

[0070] The center value of the edge distance difference center value set is re-iteratively screened in the edge distance difference set, and the cost loss function is used to calculate the cost loss function value of the edge distance difference iterative center neighborhood set obtained in each iteration, until the number of center iterations meets the preset number of iterations, the edge distance difference iterative center neighborhood set corresponding to the minimum value of the cost loss function in the iterative process is used as the edge distance difference center neighborhood set.

[0071] In one possible embodiment, the tool standard image refers to a standard reference image of the tool of the shield machine cutter disc, which is usually an image taken when the tool is new or undamaged. These standard images are used as a benchmark for subsequent comparison and analysis, and are used to compare with the tool surface images collected in actual work to evaluate the wear of the tool. The tool comparison result reflects the degree of tool wear of the current shield machine cutter disc. The degree of tool wear is quantified by calling the standard image and the actual image for edge comparison, and then calculating the edge distance difference. Then, based on these differences, through a series of screening and iterative processing, the degree of wear of the tool edge is accurately determined, thereby obtaining the wear analysis result of the cutter disc. The technical effect of improving the reliability of wear monitoring and analysis of the shield machine cutter disc is achieved.

[0072] In a possible embodiment, the preset edge sampling point position set is a position set of tool edge points preset by a person skilled in the art. The tool surface acquisition image is mapped to the tool standard image, and the edge distance difference between the position corresponding to the preset edge sampling point position set on the tool standard image and the corresponding position on the tool surface acquisition image is determined to obtain the edge distance difference set.

[0073] Preferably, in order to be able to screen the accidental values ​​in the edge distance difference set and obtain the same type of edge distance differences of different sizes from the two dimensions of size and position, it is necessary to screen the edge distance difference center value of the edge distance difference set and determine the edge distance difference center value set. Thus, the technical effect of providing data support for determining the general situation of tool wear of the shield machine cutter head is achieved.

[0074] Optionally, a preset number (a data amount pre-set by a person skilled in the art) of edge distance differences are randomly extracted from the edge distance difference set as an initial edge distance difference center value set. The initial edge distance difference center value set is authenticated based on the two dimensions of the size of the edge distance difference and the position of the edge distance difference. Preferably, the initial edge distance difference center value set is enumerated in pairs to obtain an enumerated combination set. Using the cosine similarity calculation formula, the enumerated combination set is traversed to perform similarity identification based on the two dimensions of the size of the edge distance difference and the position of the edge distance difference to obtain an enumerated combination similarity set.

[0075] The number of enumerated combination similarities in the enumerated combination similarity set that is greater than or equal to the preset maximum similarity is counted, and when the statistical result meets the requirements of those skilled in the art, the authentication is passed. If the authentication is passed, the initial edge distance difference center value set is used as the edge distance difference center value set.

[0076] When the statistical result does not meet the requirements of the technicians in this field, the authentication fails. If the authentication fails, the edge distance difference center value in the edge distance difference set is screened again.

[0077] After obtaining the edge distance difference center value set, the difference from each edge distance difference in the edge distance difference set to the edge distance difference center value set is calculated respectively, and each edge distance difference is classified into the neighborhood corresponding to the edge distance difference center value with the smallest difference, so as to obtain the edge distance difference starting center neighborhood set.

[0078] Furthermore, the cost loss function is used to perform a cost loss analysis of neighborhood construction on the starting center neighborhood set of the edge distance difference from two dimensions: the size of the edge distance difference and the position of the edge distance difference, to obtain the starting cost loss function value.

[0079] Optionally, the center value of the edge distance difference center value set is re-iteratively screened in the edge distance difference value set to obtain the edge distance difference iterative center value set that has passed the certification, and based on the same principle as the edge distance difference starting center neighborhood set, the center neighborhood of the edge distance difference set is constructed according to the edge distance difference iterative center value set to obtain the edge distance difference iterative center neighborhood set. After multiple iterations, the cost loss function is used to calculate the cost loss function value of the edge distance difference iterative center neighborhood set obtained in each iteration until the number of center iterations meets the preset number of iterations, and the edge distance difference iterative center neighborhood set corresponding to the minimum value of the cost loss function in the iterative process is used as the edge distance difference center neighborhood set.

[0080] By comparing the tool edge of the tool standard image with the tool surface acquisition image, it is possible to more accurately identify the local changes and overall trends of tool wear, determine the tool edge wear, obtain the tool comparison result, and use the tool comparison result as the tool head wear analysis result, thereby achieving the technical effect of improving the accuracy of wear monitoring analysis results.

[0081] In summary, the embodiments of the present application have at least the following technical effects:

[0082] This application collects the status data of the cutter disc and the cutter in real time through the sensor array arranged on the surface of the cutter disc and the cutter, ensuring the comprehensiveness and real-time nature of the data. Then, the cutter disc data is interactively analyzed, and combined with the working soil layer status characteristics, a more accurate cutter disc operation characteristic is obtained, which can identify the stability and wear trend of the cutter disc. When the cutter disc operation stability exceeds the threshold, the coolant injection device is activated in time for cleaning to prevent the cutter disc from being damaged due to dust accumulation or excessive wear, and to improve the accuracy of subsequent image acquisition. Then, the cutter disc is imaged through a high-definition camera, and the working status analysis of the cutter disc is further optimized, achieving a more accurate cutter disc wear analysis, achieving the technical effect of improving the accuracy of wear analysis.

[0083] Embodiment 2, based on the same inventive concept as the method for monitoring and analyzing the cutter head wear of a shield machine in tunneling in the aforementioned embodiment, as shown in the attached Figure 2 As shown, the present application provides a shield machine cutter head wear monitoring and analysis system during tunneling, and the system and method embodiments in the present application embodiments are based on the same inventive concept. The system includes:

[0084] The monitoring data array sequence acquisition module 11 is used to interactively arrange the monitoring sensor array on the surface of the shield machine cutter head and the cutter, extract the monitoring data of the preset monitoring window, and obtain the monitoring data array sequence, wherein each monitoring data array includes the monitoring data of the monitoring sensor array at a monitoring time point of the preset monitoring window;

[0085] The cutter head operation feature set acquisition module 12 is used to perform a cutter head data interactive analysis on the monitoring data array sequence according to the order of monitoring time points from the front to the back in the preset monitoring window, and perform a cutter head operation feature analysis on the cutter head data interactive analysis result to obtain a cutter head operation feature set;

[0086] The working soil layer state feature acquisition module 13 is used to call a high-definition camera to collect images of the working soil layer and soil layer broken particles, obtain a soil layer image and a soil layer broken particle image, traverse the soil layer image and the soil layer broken particle image to perform working soil layer state analysis, and obtain working soil layer state features, wherein the high-definition camera is arranged in the peripheral protection area of ​​the shield machine cutter head;

[0087] The tool surface acquisition image acquisition module 14 is used to perform a tool disc operation stability analysis based on the tool disc operation feature set and the working soil layer state feature. When the tool disc operation stability result is lower than a preset threshold, the coolant spraying device is activated to clean the surface of the shield machine tool disc, and the high-definition camera is used to acquire an image of the tool surface of the cleaned shield machine tool disc to obtain a tool surface acquisition image.

[0088] The cutterhead wear analysis result acquisition module 15 is used to call the tool standard image of the shield machine cutterhead, compare the tool edge of the tool standard image with the tool surface acquisition image, obtain the tool comparison result, and use the tool comparison result as the cutterhead wear analysis result, wherein the acquisition position of the tool standard image and the tool surface acquisition image is the same.

[0089] Furthermore, the cutter head wear analysis result obtaining module 15 is used to perform the following steps:

[0090] Mapping the tool surface acquisition image to the tool standard image, calculating edge distance difference according to a preset edge sampling point position set, and obtaining an edge distance difference set;

[0091] Based on the size of the edge distance difference and the position of the edge distance difference, the edge distance difference set is screened for the edge distance difference center value to determine the edge distance difference center value set;

[0092] Taking the edge distance difference center value set as the starting point, performing center iteration on the edge distance difference set, when the number of center iterations meets the preset number of iterations, obtaining the edge distance difference center neighborhood set according to the minimum value of the cost loss function in the iteration process, wherein each edge distance difference center neighborhood corresponds to an edge distance difference center;

[0093] Traversing the edge distance difference center neighborhood set to perform mean calculation to obtain the edge distance difference center neighborhood mean set;

[0094] The tool comparison result is obtained by performing weighted calculation based on the edge distance difference center neighborhood mean set.

[0095] Furthermore, the cutter head wear analysis result obtaining module 15 is used to perform the following steps:

[0096] Randomly extracting a preset number of edge distance difference values ​​from the edge distance difference value set as an initial edge distance difference value center value set;

[0097] Performing central value authentication on the initial edge distance difference central value set from two dimensions, namely, the size of the edge distance difference and the position of the edge distance difference. If the authentication is passed, taking the initial edge distance difference central value set as the edge distance difference central value set;

[0098] If the authentication fails, the edge distance difference center value in the edge distance difference set is screened again.

[0099] Furthermore, the cutter head wear analysis result obtaining module 15 is used to perform the following steps:

[0100] Calculate the difference between each edge distance difference in the edge distance difference set and the edge distance difference center value set respectively, assign each edge distance difference to the neighborhood corresponding to the edge distance difference center value with the smallest difference, and obtain the edge distance difference starting center neighborhood set;

[0101] Considering the size and position of the edge distance difference, a cost loss function is constructed, wherein the cost loss function is:

[0102] ;

[0103] in, is the cost loss function value, is the edge distance difference starting center neighborhood set The edge distance difference starts from the center of the neighborhood The edge distance difference, is the edge distance difference starting center neighborhood set The edge distance difference starts from the center of the neighborhood The edge distance difference position corresponding to the edge distance difference, is the edge distance difference center value set The center value of the edge distance difference, is the edge distance difference center value set The edge distance difference position corresponding to the edge distance difference center value, is the weight of the edge distance difference and the edge distance difference position, For the The total number of edge distance differences contained in the neighborhood of the starting center of the edge distance difference, is the number of central neighbors of the starting central neighborhood set of edge distance difference, is an integer greater than or equal to 1;

[0104] Calling a cost loss function to perform a cost analysis on the starting center neighborhood set of the edge distance difference to obtain a starting cost loss function value;

[0105] The center value of the edge distance difference center value set is re-iteratively screened in the edge distance difference set, and the cost loss function is used to calculate the cost loss function value of the edge distance difference iterative center neighborhood set obtained in each iteration, until the number of center iterations meets the preset number of iterations, the edge distance difference iterative center neighborhood set corresponding to the minimum value of the cost loss function in the iterative process is used as the edge distance difference center neighborhood set.

[0106] Furthermore, a soil layer state analysis network layer is pre-constructed, and the soil layer image and the soil layer broken particle image are analyzed and identified using the soil layer state analysis network layer to obtain the working soil layer state characteristics.

[0107] Furthermore, the cutterhead operation feature set acquisition module 12 is used to perform the following steps:

[0108] Performing data interaction analysis on the first monitoring data array and the second monitoring data array of the monitoring data array sequence to obtain a first interactive cutterhead monitoring data array;

[0109] Performing data interaction analysis on the first interactive cutter head monitoring data array and the third monitoring data array of the monitoring data array sequence to obtain a second interactive cutter head monitoring data array;

[0110] Perform data interaction analysis on the N-2th interactive tool disk monitoring data array and the Nth monitoring data array of the monitoring data array sequence to obtain the N-1th interactive tool disk monitoring data array, and use the N-1th interactive tool disk monitoring data array as the tool disk data interaction analysis result, wherein N is the total number of monitoring data arrays in the monitoring data array sequence, and N is an integer greater than or equal to 1.

[0111] Furthermore, the cutterhead operation feature set acquisition module 12 is used to perform the following steps:

[0112] Performing inner product mapping identification on the first monitoring data array and the second monitoring data array of the monitoring data array sequence to obtain an inner product mapping calculation result array;

[0113] Traversing the inner product mapping calculation result array to perform normalization processing to obtain an inner product mapping normalized calculation result array;

[0114] Interactive convolution calculation is performed on the inner product mapping normalized calculation result array and the first monitoring data array to obtain a first interactive cutter head monitoring data array.

[0115] It should be noted that the above-mentioned sequence of the embodiments of the present application is only for description and does not represent the advantages and disadvantages of the embodiments. And the above-mentioned specific embodiments of this specification are described. In addition, the processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0116] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application should be included in the protection scope of the present application.

[0117] This specification and the drawings are merely exemplary illustrations of the present application and are deemed to cover any modification, variation, combination or equivalent within the scope of the present application. Obviously, a person skilled in the art may make various modifications and variations to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalents, the present application intends to include these modifications and variations.

Claims

1. A method for monitoring and analyzing cutterhead wear of a shield machine during tunneling, characterized in that: The method comprises: The monitoring sensor arrays interactively arranged on the surface of the shield machine cutter head and the cutter are used to extract the monitoring data of a preset monitoring window to obtain a monitoring data array sequence, wherein each monitoring data array includes the monitoring data of the monitoring sensor array at a monitoring time point of the preset monitoring window; According to the order of monitoring time points in the preset monitoring window from front to back, the cutter head data interactive analysis is performed on the monitoring data array sequence, and the cutter head data interactive analysis result is analyzed for the cutter head operation feature to obtain a cutter head operation feature set; Calling a high-definition camera to collect images of the working soil layer and soil layer broken particles, obtaining a soil layer image and a soil layer broken particle image, traversing the soil layer image and the soil layer broken particle image to perform working soil layer state analysis, and obtaining working soil layer state characteristics, wherein the high-definition camera is arranged in the peripheral protection area of ​​the shield machine cutter head; Performing a cutterhead operation stability analysis according to the cutterhead operation feature set and the working soil layer state feature, activating a coolant spray device to clean the surface of the shield machine cutterhead when the cutterhead operation stability result is lower than a preset threshold, and using the high-definition camera to capture an image of the tool surface of the cleaned shield machine cutterhead to obtain a tool surface capture image; Calling the tool standard image of the shield machine cutter head, comparing the tool standard image with the tool surface acquisition image to obtain a tool comparison result, and using the tool comparison result as the cutter head wear analysis result, wherein the collection position of the tool standard image and the tool surface acquisition image is the same; The method includes calling a standard tool image of the cutter head of the shield machine, comparing the tool edge of the standard tool image with the tool surface acquisition image, obtaining a tool comparison result, and using the tool comparison result as a cutter head wear analysis result, including: Mapping the tool surface acquisition image to the tool standard image, calculating edge distance difference according to a preset edge sampling point position set, and obtaining an edge distance difference set; Based on the size of the edge distance difference and the position of the edge distance difference, the edge distance difference set is screened for the edge distance difference center value to determine the edge distance difference center value set; Taking the edge distance difference center value set as the starting point, performing center iteration on the edge distance difference set, when the number of center iterations meets the preset number of iterations, obtaining the edge distance difference center neighborhood set according to the minimum value of the cost loss function in the iteration process, wherein each edge distance difference center neighborhood corresponds to an edge distance difference center; Traversing the edge distance difference center neighborhood set to perform mean calculation to obtain the edge distance difference center neighborhood mean set; The tool comparison result is obtained by performing weighted calculation based on the edge distance difference center neighborhood mean set.

2. A shield machine cutter head wear monitoring and analysis method during tunneling as claimed in claim 1, characterized in that: Based on the size of the edge distance difference and the position of the edge distance difference, the edge distance difference set is screened for the edge distance difference center value to determine the edge distance difference center value set, including: Randomly extracting a preset number of edge distance difference values ​​from the edge distance difference value set as an initial edge distance difference value center value set; Performing central value authentication on the initial edge distance difference central value set from two dimensions, namely, the size of the edge distance difference and the position of the edge distance difference. If the authentication is passed, taking the initial edge distance difference central value set as the edge distance difference central value set; If the authentication fails, the edge distance difference center value in the edge distance difference set is screened again.

3. A shield machine cutter head wear monitoring and analysis method during tunneling as claimed in claim 2, characterized in that: Taking the edge distance difference center value set as the starting point, performing center iteration on the edge distance difference set, when the number of center iterations meets the preset number of iterations, obtaining the edge distance difference center neighborhood set according to the minimum value of the cost loss function in the iteration process, including: Calculate the difference between each edge distance difference in the edge distance difference set and the edge distance difference center value set respectively, assign each edge distance difference to the neighborhood corresponding to the edge distance difference center value with the smallest difference, and obtain the edge distance difference starting center neighborhood set; Considering the size and position of the edge distance difference, a cost loss function is constructed, wherein the cost loss function is: ； in, is the cost loss function value, is the edge distance difference starting center neighborhood set The edge distance difference starts from the center of the neighborhood The edge distance difference, is the edge distance difference starting center neighborhood set The edge distance difference starts from the center of the neighborhood The edge distance difference position corresponding to the edge distance difference, is the edge distance difference center value set The center value of the edge distance difference, is the edge distance difference center value set The edge distance difference position corresponding to the edge distance difference center value, is the weight of the edge distance difference and the edge distance difference position, For the The total number of edge distance differences contained in the neighborhood of the starting center of the edge distance difference, is the number of central neighbors of the starting central neighborhood set of edge distance difference, is an integer greater than or equal to 1; Calling a cost loss function to perform a cost analysis on the starting center neighborhood set of the edge distance difference to obtain a starting cost loss function value; The center value of the edge distance difference center value set is re-iteratively screened in the edge distance difference set, and the cost loss function is used to calculate the cost loss function value of the edge distance difference iterative center neighborhood set obtained in each iteration, until the number of center iterations meets the preset number of iterations, the edge distance difference iterative center neighborhood set corresponding to the minimum value of the cost loss function in the iterative process is used as the edge distance difference center neighborhood set.

4. The method for monitoring and analyzing cutter head wear of a shield machine during tunneling as claimed in claim 1, characterized in that: A soil layer state analysis network layer is pre-constructed, and the soil layer image and the soil layer broken particle image are analyzed and identified using the soil layer state analysis network layer to obtain the working soil layer state characteristics.

5. The method for monitoring and analyzing cutter head wear of a shield machine during tunneling as claimed in claim 1, characterized in that: According to the order of monitoring time points in the preset monitoring window from front to back, the cutter head data interaction analysis is performed on the monitoring data array sequence, and the cutter head data interaction analysis result is analyzed for the cutter head operation feature set to obtain the cutter head operation feature set, including: Performing data interaction analysis on the first monitoring data array and the second monitoring data array of the monitoring data array sequence to obtain a first interactive cutterhead monitoring data array; Performing data interaction analysis on the first interactive cutter head monitoring data array and the third monitoring data array of the monitoring data array sequence to obtain a second interactive cutter head monitoring data array; Perform data interaction analysis on the N-2th interactive tool disk monitoring data array and the Nth monitoring data array of the monitoring data array sequence to obtain the N-1th interactive tool disk monitoring data array, and use the N-1th interactive tool disk monitoring data array as the tool disk data interaction analysis result, wherein N is the total number of monitoring data arrays in the monitoring data array sequence, and N is an integer greater than or equal to 1.

6. A method for monitoring and analyzing cutterhead wear of a shield machine during tunneling as claimed in claim 5, characterized in that: include: Performing inner product mapping identification on the first monitoring data array and the second monitoring data array of the monitoring data array sequence to obtain an inner product mapping calculation result array; Traversing the inner product mapping calculation result array to perform normalization processing to obtain an inner product mapping normalized calculation result array; Interactive convolution calculation is performed on the inner product mapping normalized calculation result array and the first monitoring data array to obtain a first interactive cutter head monitoring data array.

7. A shield machine cutterhead wear monitoring and analysis system during tunneling, characterized in that: The system is used to implement a shield machine cutter head wear monitoring and analysis method during tunneling as described in any one of claims 1 to 6, and the system comprises A monitoring data array sequence acquisition module is used to interactively arrange the monitoring sensor arrays on the surface of the shield machine cutter head and the cutter, extract the monitoring data of the preset monitoring window, and obtain the monitoring data array sequence, wherein each monitoring data array includes the monitoring data of the monitoring sensor array at a monitoring time point of the preset monitoring window; a cutter disc operation feature set acquisition module, configured to perform a cutter disc data interactive analysis on the monitoring data array sequence according to the order of monitoring time points from front to back in the preset monitoring window, and perform a cutter disc operation feature analysis on the cutter disc data interactive analysis result to obtain a cutter disc operation feature set; A working soil layer state feature acquisition module is used to call a high-definition camera to collect images of the working soil layer and soil layer broken particles, obtain a soil layer image and a soil layer broken particle image, traverse the soil layer image and the soil layer broken particle image to perform working soil layer state analysis, and obtain working soil layer state features, wherein the high-definition camera is arranged in the peripheral protection area of ​​the shield machine cutter head; a tool surface acquisition image acquisition module, which is used to perform a tool disc operation stability analysis based on the tool disc operation feature set and the working soil layer state feature; when the tool disc operation stability result is lower than a preset threshold, activate the coolant spraying device to clean the surface of the shield machine tool disc, and use the high-definition camera to acquire an image of the tool surface of the cleaned shield machine tool disc to obtain a tool surface acquisition image; The cutterhead wear analysis result acquisition module is used to call the tool standard image of the shield machine cutterhead, compare the tool edge of the tool standard image with the tool surface acquisition image, obtain the tool comparison result, and use the tool comparison result as the cutterhead wear analysis result, wherein the acquisition position of the tool standard image and the tool surface acquisition image is the same.

Images