Shield tunneling machine hob abrasion prediction method and device, computer equipment, readable storage medium and program product

Abstract

The invention relates to a shield tunneling machine hob wear prediction method and device, computer equipment, a computer readable storage medium and a computer program product, relates to a geological excavation technology, and can improve the prediction precision of a shield tunneling machine on hob wear in an operation process. The method comprises the following steps: determining a geological information map of a shield tunneling machine hobbing cutter operation environment, and obtaining hobbing cutter arrangement data and tunneling data under the geological information map; according to the hob arrangement data, horizontally segmenting the geological information map to obtain geological information map horizontal slices; according to the tunneling data, vertically segmenting the geological information map to obtain vertical slices of the geological information map; determining stratigraphic data corresponding to the horizontal slices of the geological information map according to the vertical slices of the geological information map, and calculating geological data according to the horizontal slices of the geological information map and the stratigraphic data; and inputting the hob arrangement data, the tunneling data and the geological data into the hob wear prediction model to obtain a hob wear prediction result.

Description

Technical Field

[0001] The present application relates to the field of geological excavation technology, and in particular to a method, device, computer equipment, computer-readable storage medium and computer program product for predicting wear of a shield machine cutter. Background Art

[0002] Shield construction is the mainstream method of tunnel construction. The cutterhead is an important component of the shield machine cutterhead in the process of rock breaking and excavation. Their overall condition directly affects the efficiency and quality of the shield. Too frequent shutdowns for maintenance will increase economic and time costs, but too long a cutter change distance will reduce the efficiency of the shield and even pose a risk of damaging the cutterhead.

[0003] In traditional technology, the wear prediction of shield machine cutter mainly relies on analytical algorithms, simulation models and other methods. These methods take cutter radius, cutter structure, uniaxial compressive strength, total thrust, penetration, torque and other shield parameters as wear influencing factors. Such methods have achieved valuable results in the wear law of cutter in a specific scenario. Artificial intelligence provides a new way to understand the more complex nonlinear relationship between cutter wear, geological conditions and shield operation parameters.

[0004] However, there are some problems with the current prediction methods or traditional methods. First, because traditional methods adjust coefficients for specific geological conditions, they lack robustness for the formation complexity problems that inevitably accompany the ultra-large diameter shield process; in addition, when using statistical or machine learning models, in order to adapt to the structural characteristics of the current model, the most common method for processing geological parameter information is summing and averaging. This method cannot accurately restore the actual working conditions of each disc cutter, so the prediction effect is difficult to guarantee in the promotion of different shield projects.

[0005] Therefore, the purpose of this invention is to develop a universal geological parameter processing method that can perform feature engineering and enhancement based on limited geological exploration data and convert it into a sequence that can match the parameters of continuous shield operations, thereby reducing unreasonable assumptions. Through this training data input method with a higher degree of restoration of the processing process, real-time prediction of hob cutter wear becomes possible. Summary of the invention

[0006] Based on this, it is necessary to provide a shield machine cutter wear prediction method, device, computer equipment, computer readable storage medium and computer program product to address the above technical problems.

[0007] In a first aspect, the present application provides a method for predicting cutter wear of a shield machine, comprising:

[0008] Determine a geological information map of the shield machine cutter operating environment, and obtain cutter placement data and excavation data corresponding to the shield machine cutter under the geological information map;

[0009] According to the cutter placement data of the shield machine cutter, the geological information map is horizontally segmented to obtain horizontal slices of the geological information map corresponding to the geological information map, and, according to the excavation data of the shield machine cutter, the geological information map is vertically segmented to obtain vertical slices of the geological information map corresponding to the geological information map;

[0010] According to the vertical slice of the geological information map, the stratigraphic data corresponding to the horizontal slice of the geological information map is determined, and according to the horizontal slice of the geological information map and the stratigraphic data corresponding to the horizontal slice of the geological information map, the geological data corresponding to the shield machine cutter operation is calculated;

[0011] The cutter placement data, the excavation data and the geological data are input into a cutter wear prediction model to obtain a cutter wear prediction result output by the cutter wear prediction model.

[0012] In one embodiment, the horizontal slicing of the geological information map according to the cutter placement data of the shield machine cutter to obtain a horizontal slice of the geological information map corresponding to the geological information map includes:

[0013] According to the cutter head function partition of the cutter arrangement data, the geological information map is horizontally segmented to obtain the geological information map horizontal slices corresponding to the geological information map; or

[0014] According to the hob cutting radius of the hob cutting data, the geological information map is horizontally segmented to obtain a geological information map horizontal slice corresponding to the geological information map.

[0015] In one embodiment, the vertical slicing of the geological information map according to the excavation data of the shield machine cutter to obtain the vertical slice of the geological information map corresponding to the geological information map includes:

[0016] According to the excavation ring interval of the excavation data, the geological information map is vertically segmented to obtain a vertical slice of the geological information map corresponding to the geological information map; or,

[0017] The geological information map is vertically segmented according to a single excavation ring of the excavation data to obtain a vertical slice of the geological information map corresponding to the geological information map.

[0018] In one embodiment, the hob wear prediction model is trained by the following steps, including:

[0019] Inputting historical disc cutter placement data, historical excavation data and historical geological data into a disc cutter wear prediction model to be trained, wherein the disc cutter wear prediction model determines the operation characteristics corresponding to the disc cutter according to the historical operation data and historical geological data, and determines the wear amount prediction result according to the operation characteristics;

[0020] The model error is determined according to the wear prediction result and the actual hob wear, and the hob wear prediction model parameters are adjusted by particle swarm algorithm. When the model error meets the training end condition, the hob wear prediction model is obtained.

[0021] In one embodiment, adjusting the parameters of the hob wear prediction model by using a particle swarm algorithm includes:

[0022] Generate a particle swarm; each particle in the particle swarm represents the parameters of the hob wear prediction model;

[0023] Determine the inertia weight and fitness threshold of the particle swarm under the current number of iterations; the inertia weight is negatively correlated with the number of iterations, and the fitness threshold is positively correlated with the number of iterations;

[0024] According to the inertia weight and the fitness threshold, the particle group and the speed and position of the particles after iteration are determined. When the iteration end condition is met, the parameters corresponding to the global optimal particle in the particle group after iteration are determined as the parameters of the hob wear prediction model.

[0025] In one embodiment, determining the geological information map when the shield machine is in operation includes:

[0026] Obtaining a geological profile during shield machine operation and geological attribute data corresponding to the geological profile;

[0027] Different strata in the geological profile are segmented, corresponding numerical labels are assigned to the segmented strata, and the numerical labels are bound to the geological attribute data of the corresponding strata to obtain the geological information map in pixel form.

[0028] In a second aspect, the present application also provides a shield machine cutter wear prediction device, comprising:

[0029] A geological information analysis module is used to determine a geological information map of the shield machine cutter operating environment, and obtain cutter placement data and excavation data corresponding to the shield machine cutter under the geological information map;

[0030] A geological information slice generation module is used to horizontally slice the geological information map according to the cutter placement data of the shield machine cutter to obtain horizontal slices of the geological information map corresponding to the geological information map, and to vertically slice the geological information map according to the excavation data of the shield machine cutter to obtain vertical slices of the geological information map corresponding to the geological information map;

[0031] A geological data calculation module is used to determine the stratigraphic data corresponding to the horizontal slice of the geological information map according to the vertical slice of the geological information map, and calculate the geological data corresponding to the shield machine cutter operation according to the horizontal slice of the geological information map and the stratigraphic data corresponding to the horizontal slice of the geological information map;

[0032] The cutter wear prediction module is used to input the cutter placement data, the excavation data and the geological data into the cutter wear prediction model to obtain the cutter wear prediction result output by the cutter wear prediction model.

[0033] In a third aspect, the present application further provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the following steps are implemented:

[0034] Determine a geological information map of the shield machine cutter operating environment, and obtain cutter placement data and excavation data corresponding to the shield machine cutter under the geological information map;

[0035] According to the cutter placement data of the shield machine cutter, the geological information map is horizontally segmented to obtain horizontal slices of the geological information map corresponding to the geological information map, and, according to the excavation data of the shield machine cutter, the geological information map is vertically segmented to obtain vertical slices of the geological information map corresponding to the geological information map;

[0036] According to the vertical slice of the geological information map, the stratigraphic data corresponding to the horizontal slice of the geological information map is determined, and according to the horizontal slice of the geological information map and the stratigraphic data corresponding to the horizontal slice of the geological information map, the geological data corresponding to the shield machine cutter operation is calculated;

[0037] The cutter placement data, the excavation data and the geological data are input into a cutter wear prediction model to obtain a cutter wear prediction result output by the cutter wear prediction model.

[0038] In a fourth aspect, the present application further provides a computer-readable storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, the following steps are implemented:

[0039] Determine a geological information map of the shield machine cutter operating environment, and obtain cutter placement data and excavation data corresponding to the shield machine cutter under the geological information map;

[0040] According to the cutter placement data of the shield machine cutter, the geological information map is horizontally segmented to obtain horizontal slices of the geological information map corresponding to the geological information map, and, according to the excavation data of the shield machine cutter, the geological information map is vertically segmented to obtain vertical slices of the geological information map corresponding to the geological information map;

[0041] According to the vertical slice of the geological information map, the stratigraphic data corresponding to the horizontal slice of the geological information map is determined, and according to the horizontal slice of the geological information map and the stratigraphic data corresponding to the horizontal slice of the geological information map, the geological data corresponding to the shield machine cutter operation is calculated;

[0042] The cutter placement data, the excavation data and the geological data are input into a cutter wear prediction model to obtain a cutter wear prediction result output by the cutter wear prediction model.

[0043] In a fifth aspect, the present application further provides a computer program product, including a computer program, which implements the following steps when executed by a processor:

[0044] Determine a geological information map of the shield machine cutter operating environment, and obtain cutter placement data and excavation data corresponding to the shield machine cutter under the geological information map;

[0045] According to the cutter placement data of the shield machine cutter, the geological information map is horizontally segmented to obtain horizontal slices of the geological information map corresponding to the geological information map, and, according to the excavation data of the shield machine cutter, the geological information map is vertically segmented to obtain vertical slices of the geological information map corresponding to the geological information map;

[0046] According to the vertical slice of the geological information map, the stratigraphic data corresponding to the horizontal slice of the geological information map is determined, and according to the horizontal slice of the geological information map and the stratigraphic data corresponding to the horizontal slice of the geological information map, the geological data corresponding to the shield machine cutter operation is calculated;

[0047] The cutter placement data, the excavation data and the geological data are input into a cutter wear prediction model to obtain a cutter wear prediction result output by the cutter wear prediction model.

[0048] The above-mentioned shield machine cutter wear prediction method, device, computer equipment, computer-readable storage medium and computer program product determine the geological information map of the shield machine cutter operating environment, and obtain the cutter layout data and excavation data corresponding to the shield machine cutter under the geological information map; according to the cutter layout data of the shield machine cutter, the geological information map is horizontally divided to obtain the horizontal slice of the geological information map corresponding to the geological information map, and, according to the excavation data of the shield machine cutter, the geological information map is vertically divided to obtain the vertical slice of the geological information map corresponding to the geological information map; according to the vertical slice of the geological information map, the stratigraphic data corresponding to the horizontal slice of the geological information map is determined, and according to the horizontal slice of the geological information map and the stratigraphic data corresponding to the horizontal slice of the geological information map, the geological data corresponding to the shield machine cutter operation is calculated; the cutter layout data, excavation data and geological data are input into the cutter wear prediction model to obtain the cutter wear prediction result output by the cutter wear prediction model. In the present application, a geological information map of the shield machine cutter operating environment is determined and cutter placement data and excavation data are obtained. The geological information map is horizontally sliced ​​in combination with the cutter placement data, and vertically sliced ​​in combination with the excavation data. Through the linkage processing of horizontal and vertical slices, the geological information map is converted into partition slices suitable for cutter operations, thereby enhancing the correspondence between geological data and actual working conditions. The stratigraphic data corresponding to the horizontal slices are extracted through vertical slices, and the geological data are calculated in combination with the horizontal slices, so that the stratigraphic characteristics of the cutter in different operating areas can be accurately quantified, forming a composite expression of the cutter operating environment, and providing high-quality input for subsequent model predictions. The wear prediction model captures the dynamic accumulation law of cutter wear through multi-dimensional data integration and time series modeling, and determines the cutter wear prediction results, so that the prediction of shield machine cutter wear can adapt complex geological conditions to variable excavation conditions, thereby improving the prediction accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] In order to more clearly illustrate the technical solutions in the embodiments of the present application or related technologies, the drawings required for use in the embodiments of the present application or related technical descriptions will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0050] Figure 1 An application environment diagram of a shield machine cutter wear prediction method in one embodiment;

[0051] Figure 2 A schematic diagram of a process of predicting the wear of a shield machine cutter in one embodiment;

[0052] Figure 3 A schematic diagram of a process for constructing geological data by slicing a geological information map in one embodiment;

[0053] Figure 4 A schematic diagram of a flow chart of combining a cutter wear prediction model with formation slice data in one embodiment;

[0054] Figure 5 is a schematic diagram of a geological cross section in one embodiment;

[0055] Figure 6 A schematic diagram of data preprocessing of a hob wear prediction model in one embodiment;

[0056] Figure 7 A schematic diagram of constructing geological data by slicing a geological information map in another embodiment;

[0057] Figure 8 The processing results of four data processing methods in four hob wear prediction models in one embodiment are shown;

[0058] Fig. 9 It is a structural block diagram of a shield machine cutter wear prediction device in one embodiment;

[0059] Fig.10 FIG. 4 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0060] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0061] The shield machine cutter wear prediction method provided in the embodiment of the present application can be applied to Figure 1 In the application environment shown. The terminal communicates with the server through the network. The data storage system can store the data that the server needs to process. The data storage system can be integrated on the server, or placed on the cloud or other network servers. The terminal can be, but is not limited to, various personal computers, laptops, smart phones, and tablets. The server can be an independent physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server that provides cloud computing services.

[0062] In an exemplary embodiment, Figure 2 As shown in the figure, a shield machine cutter wear prediction method is provided. Figure 1 The terminal in is taken as an example to illustrate, including the following steps S202 to S208. Among them:

[0063] Step 202, determining a geological information map of the shield machine cutter operating environment, and obtaining cutter placement data and excavation data corresponding to the shield machine cutter under the geological information map.

[0064] Among them, the geological information map is a key data graphic used to represent the characteristics of the stratum during the tunneling operation of the shield machine. It is generated based on the drilling exploration data and core test results along the tunnel, and is used to describe the stratum distribution, lithology, joint characteristics and other geological attributes. Through graphical geological information, the tunneling environment is expanded from discrete data points to a complete and continuous spatial expression, improving the visualization effect and data adaptation ability of the stratum characteristics. The cutter layout data refers to the geometric layout and operation information of the cutters on the cutterhead of the shield machine, including but not limited to the installation radius, spacing, angle and partition function of the cutters (such as center cutters, front cutters and edge cutters). It is an important basis for analyzing the working status of the cutters. The tunneling data is a set of dynamic operating parameters during the operation of the shield machine, including cutterhead torque, propulsion speed, penetration, thrust, etc. They reflect the interaction between the cutters and the stratum and are an important basic data source for model evaluation of cutter wear.

[0065] For example, the terminal reads the drilling exploration results and the tunnel geological profile, processes the sparse point data through the interpolation algorithm, and generates a continuous geological information map covering the excavation path. Secondly, in the process of acquiring the disc cutter layout data, the terminal extracts the cutter layout parameters from the shield machine design drawings, and combines the on-site laser scanning to verify the actual installation parameters to ensure data consistency. Finally, the acquisition of excavation data relies on real-time collection by the terminal and is stored in the shield machine system in the form of a time series.

[0066] Step S204, according to the cutter placement data of the shield machine cutter, the geological information map is horizontally divided to obtain horizontal slices of the geological information map corresponding to the geological information map, and, according to the excavation data of the shield machine cutter, the geological information map is vertically divided to obtain vertical slices of the geological information map corresponding to the geological information map.

[0067] Among them, horizontal segmentation refers to the horizontal segmentation of the geological information map based on the shield machine cutter placement data. The result of segmentation is to divide the entire geological information map into different horizontal slices to achieve accurate modeling of the geological conditions faced by different cutter areas.

[0068] Vertical segmentation is to vertically segment the geological information map along the excavation direction according to the excavation data of the shield machine. After segmentation, the geological information map is decomposed into a series of vertical slices, each of which represents the geological information of the shield machine at a specific excavation position. Vertical segmentation can accurately express the dynamic changes of geological characteristics during the shield advancement process.

[0069] The geological information map slice is the result of horizontal slicing and vertical slicing, and is the segmentation unit of the geological information map. Through the slicing method, the continuous geological information map is refined according to the cutter layout data and excavation data, which not only retains the spatial distribution information of geological characteristics, but also enables each slice to be directly associated with the working status and operation path of the cutter.

[0070] For example, the terminal reads the design parameters or verification data of the shield machine cutterhead, identifies the roller cutter placement data, and generates horizontal slices according to the stratum range corresponding to the roller cutter placement data in the geological information map. After completing the horizontal segmentation, the terminal performs vertical segmentation according to the excavation data. The terminal reads the excavation ring number or the advancement distance, and decomposes the geological information map into a series of vertical slices along the advancement direction.

[0071] Optionally, try to slice the stratum information from different dimensions, densities, and methods, perform horizontal slicing at equal intervals based on the cutter placement data, and perform vertical slicing in units of cutter change intervals based on the excavation data.

[0072] Step S206, according to the vertical slice of the geological information map, determine the stratigraphic data corresponding to the horizontal slice of the geological information map, and calculate the geological data corresponding to the shield machine cutter operation according to the horizontal slice of the geological information map and the stratigraphic data corresponding to the horizontal slice of the geological information map.

[0073] Among them, the stratigraphic data is a collection of lithological characteristics and related physical parameters corresponding to each slice in the geological information map, including but not limited to uniaxial compressive strength (UCS), joint density, water content, cutterhead abrasion index (CAI), etc. These data reflect the nature of the strata that the shield cutter faces during operation and are an important basis for predicting cutter wear.

[0074] The geological data is the result of calculation based on the formation attribute information after comprehensive horizontal slicing and vertical slicing, combined with the actual conditions of the disc cutter operation. It is used to express the resistance or wear-related parameters that the disc cutter needs to overcome when acting on the formation under operating conditions.

[0075] Exemplarily, first, the terminal extracts the attribute data of the strata contained in the corresponding horizontal slice according to the vertical slice of the geological information map. Specifically, the terminal determines the strata types and the proportion of each type in the area by analyzing the intersection area of ​​the horizontal slice and the vertical slice, and extracts the relevant physical attribute data. In one embodiment, in a horizontal slice, if the area covered by the vertical slice has two strata, medium-weathered rock and hard rock, the terminal will calculate the proportion of the two strata in the intersection area and record their physical parameter values. Secondly, the terminal calculates the geological data of the roller cutter operation based on the horizontal slice and its corresponding stratum data. Exemplarily, the terminal combines the stratum attribute data and uses a weighted formula to calculate the composite parameters related to the roller cutter wear.

[0076] Step S208, inputting the cutter arrangement data, excavation data and geological data into the cutter wear prediction model to obtain the cutter wear prediction result output by the cutter wear prediction model.

[0077] Among them, the hob wear prediction model is a model used to predict the degree of wear of the hob under operating conditions. It can be based on a time series model or other machine learning algorithms, combined with historical data for training, and predicted through real-time data input.

[0078] For example, first, the terminal integrates the cutter placement data, excavation data and geological data into a standardized input data set to ensure that the data dimension and format meet the input requirements of the model. The terminal preprocesses these data, and optionally aligns the time series of different data sources, such as unifying the time dimensions of the cutter placement data and excavation data based on the excavation ring number or timestamp, and smoothes outliers.

[0079] Next, the terminal sends the standardized input data to the cutter wear prediction model. The model analyzes the cutter layout data, combines the dynamic working condition parameters in the excavation data, and uses the formation characteristics in the geological data to comprehensively predict the degree of cutter wear under the current operating conditions. Finally, the terminal outputs the prediction results of the model and marks the results as the cutter wear amount corresponding to the time series or ring number sequence. Optionally, the terminal generates a visual report, superimposes the wear prediction value on the cutter layout of the shield machine cutterhead, and intuitively displays the wear degree and distribution of each cutter. In addition, the terminal can also make cutter replacement suggestions based on the model prediction results, providing the construction team with an optimized reference for the cutter replacement plan.

[0080] In the above-mentioned shield machine cutter wear prediction method, the geological information map of the shield machine cutter operating environment is determined, and the cutter layout data and excavation data corresponding to the shield machine cutter under the geological information map are obtained; according to the cutter layout data of the shield machine cutter, the geological information map is horizontally cut to obtain the horizontal slice of the geological information map corresponding to the geological information map, and, according to the excavation data of the shield machine cutter, the geological information map is vertically cut to obtain the vertical slice of the geological information map corresponding to the geological information map; according to the vertical slice of the geological information map, the stratigraphic data corresponding to the horizontal slice of the geological information map is determined, and according to the horizontal slice of the geological information map and the stratigraphic data corresponding to the horizontal slice of the geological information map, the geological data corresponding to the shield machine cutter operation is calculated; the cutter layout data, excavation data and geological data are input into the cutter wear prediction model to obtain the cutter wear prediction result output by the cutter wear prediction model. In this embodiment, the geological information map of the shield machine cutter operating environment is determined and the cutter placement data and excavation data are obtained. The geological information map is horizontally sliced ​​in combination with the cutter placement data, and vertically sliced ​​in combination with the excavation data. Through the linkage processing of horizontal and vertical slices, the geological information map is converted into partition slices suitable for the cutter operation, thereby enhancing the correspondence between the geological data and the actual working conditions. The stratigraphic data corresponding to the horizontal slices are extracted through vertical slices, and the geological data are calculated in combination with the horizontal slices, so that the stratigraphic characteristics of the cutter in different operating areas can be accurately quantified, forming a composite expression of the cutter operating environment, and providing high-quality input for subsequent model prediction. The wear prediction model captures the dynamic accumulation law of cutter wear through multi-dimensional data integration and time series modeling, and determines the cutter wear prediction result, so that the prediction of shield machine cutter wear can adapt complex geological conditions to variable excavation conditions, thereby improving the prediction accuracy.

[0081] In an exemplary embodiment, according to the cutter placement data of the shield machine cutter, the geological information map is horizontally divided to obtain horizontal slices of the geological information map corresponding to the geological information map, including: according to the cutter disc functional zoning of the cutter placement data, the geological information map is horizontally divided to obtain horizontal slices of the geological information map corresponding to the geological information map; or, according to the cutter placement radius of the cutter placement data, the geological information map is horizontally divided to obtain horizontal slices of the geological information map corresponding to the geological information map.

[0082] Among them, the cutter head functional zoning refers to the regional division of the shield machine cutter head according to the function and position of the cutter. The cutters in each zone have different positions and ranges of action, and the mechanical loads and interaction characteristics with the strata are also different, which can be used to refine the geological conditions input into the cutter wear prediction model.

[0083] The cutter radius refers to the radial distance from the center axis to the installation position of the cutter on the shield machine cutterhead. The different cutter radius determines the area and force of the stratum cut by the cutter, affecting the working state of each cutter during operation. Slicing based on the cutter radius can express the stratum contact characteristics of the cutter at different radius positions.

[0084] For example, the terminal reads the cutter arrangement data in the cutterhead design and divides the cutterhead into functional areas, such as the center cutter area, the front cutter area, and the edge cutter area. Then, the terminal performs corresponding horizontal segmentation on the geological information map according to these functional divisions to generate horizontal slices of the geological information map corresponding to each functional area.

[0085] For example, the terminal reads the installation radius data of the roller cutter, divides the geological information map according to the radial position of the roller cutter, and generates horizontal slices corresponding to the cutter radius. Each slice corresponds to the roller cutter operating range of a specific radius on the cutter disc, which facilitates the model to subdivide the working areas of roller cutters of different radii for analysis.

[0086] In this embodiment, the geological information map is horizontally divided according to the cutter function zoning and cutter placement radius of the cutter placement data, so that the model can distinguish the operating range of cutters of different radii and their interaction characteristics with the formation, thereby achieving high-precision cutter wear prediction in multiple regions and multiple radius levels.

[0087] In an exemplary embodiment, the geological information map is vertically segmented according to the excavation data of the shield machine cutter to obtain vertical slices of the geological information map corresponding to the geological information map, including: vertically segmenting the geological information map according to the excavation ring interval of the excavation data to obtain vertical slices of the geological information map corresponding to the geological information map; or vertically segmenting the geological information map according to a single excavation ring of the excavation data to obtain vertical slices of the geological information map corresponding to the geological information map.

[0088] Among them, the tunneling ring interval refers to the segmentation of a certain number of tunneling rings during the advancement of the shield machine. Using the tunneling ring interval for vertical segmentation can reflect the geological change characteristics of the shield machine in a larger range. A single tunneling ring refers to the unit formed by each ring completed by the shield machine during the advancement process. Vertical segmentation uses a single tunneling ring as a unit, which can capture the most detailed geological changes and fluctuations in shield operation parameters.

[0089] Exemplarily, the terminal vertically segments the geological information map according to the excavation ring interval of the excavation data. The excavation ring interval is used as a segmentation unit, so that the geological information map is segmented into vertical slices corresponding to a group of excavation rings. For example, the terminal can segment according to every 10 rings as an interval, and generate vertical slices of the geological information map covering these 10 rings. This interval segmentation method is helpful to observe and analyze the trend of geological characteristics and stratum changes within a certain advancement range, and is particularly suitable for analyzing geological continuity and changes at a larger scale.

[0090] For example, the terminal vertically slices the geological information map according to the excavation data of a single excavation ring. Each individual excavation ring, as the smallest segmentation unit, can carefully reflect the geological changes and details of the excavation process of each ring. For example, the terminal reads the excavation data such as the advancement speed and penetration of each ring, and divides the geological information corresponding to these parameters into single-ring vertical slices.

[0091] In this embodiment, by vertically dividing the excavation data into excavation ring intervals and single excavation rings, the terminal provides high-resolution geological data input for the wear prediction model, which can not only meet the needs of large-scale stratigraphic analysis, but also adapt to changes in small-scale excavation conditions.

[0092] In a specific embodiment, Figure 3 The binary address information map shown in (a) is sliced. First, the terminal slices the geological profile horizontally according to the cutter placement data, and divides the profile into the central cutter area, the front cutter area and the outer edge cutter area according to the functional area of ​​the cutter head. Figure 3 (b) Next, the terminal uses the excavation data to divide the geological profile into ring intervals in the vertical direction, forming vertical slices of 10 rings. For example, the 200-210 ring and the 320-330 ring are used as independent vertical slices. In the intersection area of ​​each horizontal and vertical slice, the terminal calculates the corresponding geological attribute data, such as the CAI value. For the 200-210 ring, the excavation object of the central cutter area and the front cutter area has only one stratum, so there is a CAI value. 200-210C = CAI 200-210F = σ1 (MPa). In the outer edge knife area, about 10% of the working area is layer 2, CAI 200-210E = σ2(MPa), and the CAI value is calculated according to the formation ratio: CAI 200-210E = 0.90×σ1 + 0.10×σ2 (MPa). Optionally, the formation-by-formation CAI calculation process is calculated by formula (1).

[0093]

[0094] in is the rock abrasiveness coefficient, is the uniaxial compressive strength (MPa), is the equivalent quartz content (%). Similarly, the 320-330 ring is also composed of two strata. Assuming that the proportions of stratum 1 are 65%, 62%, and 45% respectively according to the area calculation, the CAI of the corresponding functional zone can be calculated according to formulas (2)-(4).

[0095]

[0096]

[0097]

[0098] Similarly, horizontal slices of the cutter radius and vertical slices of the excavation ring can be combined, such as Figure 3 (c) The terminal uses the cutter installation radius as the basis for horizontal slicing and the excavation ring number as the unit for vertical slicing. The terminal calculates the excavation trajectory of each disc cutter in each ring and its abrasiveness index in each stratum. For example, taking the disc cutter with a rotation radius of R2 as an example, when it excavates to excavation ring 2, the cutter alternately cuts 4 strata, and its trajectory is the sum of the trajectories in each stratum, as shown in formula (5),

[0099]

[0100] Among them, R2 is the installation radius of S02 hob, L1 R02002 It represents the excavation trajectory of the S02 cutter in the 002th ring and the first stratum. At this time, the CAI corresponding to the cutter R2 in the 002th ring can be obtained by formula (6).

[0101]

[0102] Therefore, the abrasiveness index of each cutter in each ring of the expandable cutter head can be expressed as shown in formula (7):

[0103]

[0104] Among them, i is the cutter number, j is the excavation ring number, and k = 1, 2, ..., n is the kth stratum in the current excavation ring.

[0105] In this embodiment, through the combination of horizontal slicing of the cutter radius and vertical slicing of the tunneling ring, the terminal can capture the wear characteristics of the cutter when it interacts with various strata at different radial positions, effectively converting complex geological information and shield operating conditions into high-quality data that can be used for model training and prediction, significantly improving the prediction accuracy and adaptability of the model.

[0106] In an exemplary embodiment, the cutter wear prediction model is trained through the following steps, including: inputting historical cutter placement data, historical excavation data and historical geological data into the cutter wear prediction model to be trained, and the cutter wear prediction model determines the operating characteristics corresponding to the cutter based on the historical operating data and historical geological data, and determines the wear amount prediction result based on the operating characteristics; determines the model error based on the wear amount prediction result and the actual cutter wear amount, and adjusts the cutter wear prediction model parameters through the particle swarm algorithm. When the model error meets the training end condition, the cutter wear prediction model is obtained.

[0107] Specifically, Figure 4 As shown on the left, first, the terminal takes the historical cutter placement data, historical excavation data, and historical geological data as input, and imports the cutter wear prediction model Long Short-Term Memory (LSTM) to be trained. The model extracts the operation characteristics of the cutter, such as the load mode and friction characteristics of the cutter under specific layout, working conditions, and geological conditions, by analyzing the input multi-source data. Based on these characteristics, the model generates wear prediction results and forms a preliminary prediction output.

[0108] Subsequently, the terminal compares the predicted output wear with the actual historical measured hob wear and calculates the prediction error of the model. This step is used to measure the accuracy of the model under the current parameter settings. To optimize the model performance, the terminal applies the particle swarm optimization (PSO) algorithm to adjust the model parameters. The PSO algorithm generates a swarm of particles, each particle represents a combination of model parameters, and updates the position and velocity of the particles according to the individual best fitness and the global best fitness in multiple iterations. The terminal evaluates the model error in each round of iteration until the error meets the training end conditions, such as reaching the preset error threshold or the maximum number of iterations, and obtains the optimized hob wear prediction model.

[0109] In this embodiment, by inputting historical cutter placement data, historical excavation data and historical geological data into the cutter wear prediction model to be trained, the complex relationship between the cutter and the formation and operating conditions can be fully captured during the training phase, and the wear characteristics and laws of the cutter under different geological conditions can be gradually learned, thereby improving the prediction ability and robustness of the model.

[0110] In an exemplary embodiment, the parameters of the roller cutter wear prediction model are adjusted by a particle swarm algorithm, including: generating a particle swarm; each particle in the particle swarm represents the parameters of the roller cutter wear prediction model; determining the inertia weight and fitness threshold of the particle swarm under the current number of iterations; the inertia weight is negatively correlated with the number of iterations, and the fitness threshold is positively correlated with the number of iterations; according to the inertia weight and the fitness threshold, determining the particle swarm and the speed and position of the particles after the iteration, and when the iteration end condition is met, determining the parameters corresponding to the global optimal particle in the particle swarm after the iteration as the roller cutter wear prediction model parameters.

[0111] Specifically, Figure 4 As shown on the right, first, the terminal generates an initial particle swarm, where each particle represents a parameter combination of a hob wear prediction model, such as learning rate, number of hidden layer nodes, and number of iterations. Each particle in the particle swarm is initialized to a random value to cover multiple potential solutions in the search space.

[0112] On the basis of the original PSO, the inertia weight and fitness threshold Improvements are made to obtain the improved particle swarm optimization algorithm (IPSO). Specifically, the terminal calculates the inertia weight of the particles in each iteration. , whose value decreases according to formula (8).

[0113]

[0114] in is the current iteration number, is the maximum number of iterations, and are the initial and final inertia weights, respectively, and optionally Set to 0.9, Set to 0.4. The dynamic adjustment of the inertia weight ensures that the search process gradually shifts from global exploration to local search, increasing the ability of the algorithm to jump out of the local optimum. The terminal adjusts the current number of iterations The fitness threshold is calculated as shown in formula (9) to screen particles with high fitness.

[0115]

[0116] The fitness threshold gradually increases with the number of iterations, ensuring that more excellent particles are retained in the later stages of the iteration to improve search efficiency. The fitness screening strategy introduces a "survival of the fittest" mechanism, selecting particles with a fitness above K to continue updating and iterating, and the remaining particles are eliminated and randomly redistributed. As the iterations proceed, in order to retain more preferred particles, the percentage of K gradually increases. Particles with low fitness are eliminated and randomly regenerated into new particles to maintain population diversity.

[0117] Then, the terminal calculates the inertia weight and fitness threshold , update the velocity and position of each particle, calculated by formula (10) (11).

[0118]

[0119]

[0120] in, and is the learning factor, and is a random variable. When the iteration end condition (such as the maximum number of iterations or the error meets the preset threshold) is met, the terminal determines the global optimal particle in the current particle swarm and uses its corresponding parameters as the final optimization parameters of the hob wear prediction model.

[0121] In this embodiment, the particle swarm algorithm performs a global search with a larger inertia weight in the initial stage. As the iteration proceeds, the inertia weight gradually decreases, and the search gradually turns to local optimization, ensuring the precise adjustment of the model parameters. The fitness threshold gradually increases with the increase in the number of iterations, so that particles with low fitness can be eliminated and new particles can be introduced in each round of iteration, thereby maintaining the diversity of the group and the comprehensiveness of the search, realizing the automatic optimization of the parameters of the hob wear prediction model, and improving the training efficiency and prediction accuracy of the model.

[0122] In an exemplary embodiment, a geological profile of a shield machine during operation and geological attribute data corresponding to the geological profile are obtained; different strata in the geological profile are segmented, corresponding numerical labels are assigned to each segmented stratum, and the numerical labels and geological attribute data of the corresponding strata are bound to obtain a geological information map in pixel form.

[0123] Specifically, the terminal reads and obtains Figure 5The geological profile shown in the figure is pre-drawn on the shield machine construction path. The profile is derived from exploration data, core tests and laboratory analysis results (such as uniaxial compression tests, point load tests and corrosion tests). At the same time, the terminal obtains geological attribute data corresponding to the geological profile, such as the uniaxial compressive strength, abrasion index and water content of each stratum, providing specific physical property descriptions for each layer of the geological profile.

[0124] After acquiring the data, the terminal processes the geological profile and multi-values ​​the tunnel geological map (N value = the number of geological components) through the identification plug-in, so that it has the function of counting the composition ratio of each value pixel point in the corresponding interval in a custom slice method from the horizontal and vertical directions, and segmenting the different strata shown in the figure. Optionally, the segmentation algorithm can use edge detection and region growth-based methods to identify the boundaries of each stratum to ensure that different strata are clearly distinguished in the figure. Subsequently, the terminal assigns a unique numerical label to each segmented stratum, such as label "1" for hard rock strata, "2" for medium-weathered rock strata, etc. These labels serve as identifiers to bind each stratum to its specific geological attribute data.

[0125] Finally, the terminal combines the segmented geological profile and numerical labels to generate a pixel-based geological information map. Each pixel represents the spatial position and numerical label of the stratum, and through data binding, the geological attribute data is embedded in the pixel.

[0126] In this embodiment, by obtaining geological profiles and their geological attribute data, a geological information map in pixel form is generated, which realizes the structural and digital expression of geological information. It effectively combines the physical properties of the strata with their spatial positions, provides detailed stratum distribution and geological characteristics, and provides accurate and comprehensive data input for the cutter wear prediction model. It enables the function of counting the composition ratio of pixel points of each value in the corresponding interval in the horizontal and vertical directions according to the custom slice method.

[0127] In order to enable those skilled in the art to better understand the above steps, the embodiment of the present application is illustrated below by using an example, but it should be understood that the embodiment of the present application is not limited to this.

[0128] In an exemplary embodiment, the terminal device first passes Figure 6The process shown prepares a data set. Obtain the geological profile and its corresponding geological attribute data on the construction path of the shield machine. The geological profile is provided by geological exploration results, borehole sampling and laboratory analysis, including the physical properties of different strata, such as uniaxial compressive strength, abrasion index, etc. Then, the terminal uses image processing technology to segment the different strata in the geological profile and assigns numerical labels to each segmented stratum, which are bound to the corresponding geological attribute data. Through this process, the terminal generates a geological information map in pixel form, in which each pixel represents the numerical label of a certain stratum and its geological attributes. At the same time, the terminal obtains the cutter placement data and excavation data of the shield machine disc. The cutter placement data includes information such as the installation position of the disc on the cutterhead, functional partitions (such as the center cutter area, the front cutter area and the outer edge cutter area) and the cutter placement radius. The excavation data includes dynamic operating parameters such as propulsion speed, thrust, penetration and cutterhead torque, which are recorded according to the excavation ring number or ring interval. According to the disc cutter placement data, the terminal horizontally segments the geological information map to generate horizontal slices of the cutterhead functional partition or cutter placement radius. For example, the terminal divides the cutterhead into the center cutter area, the front cutter area, and the outer cutter area, and generates corresponding horizontal slices on the geological information map to distinguish the geological environment of each cutter area. On this basis, the terminal vertically slices the geological information map according to the excavation ring interval or a single excavation ring based on the excavation data. The terminal combines horizontal and vertical slices to calculate the geological data of each cutterhead partition or roller cutter radius within a specific ring interval.

[0129] The terminal inputs the calculated cutter placement data, excavation data and geological data into the cutter wear prediction model. The model is a deep learning model based on LSTM, and the parameters are adjusted by the IPSO algorithm. The historical cutter placement data, excavation data and geological data are used to train the model so that it can learn the cutter wear characteristics under different geological conditions and construction parameters. During the model training process, after the terminal inputs the data, the model outputs the wear prediction result. The terminal compares the prediction result with the actual measured historical wear and calculates the model error. To optimize the model performance, the terminal adjusts the model parameters, such as the learning rate and the number of hidden layer nodes, through the IPSO algorithm, so that the model gradually converges to the state with the minimum error.

[0130] The trained model is used to predict the wear of the disc cutter in real time. The terminal inputs the current disc cutter placement data, real-time excavation data and real-time calculated geological data into the model to obtain the disc cutter wear prediction results output by the model.

[0131] In this embodiment, by combining the horizontal slices of functional zones and the vertical slices of ring intervals, the geological data is processed in a refined manner and high-resolution modeling is achieved. This method not only improves the details and accuracy of the model input data, but also enables the cutter wear prediction model to provide more accurate prediction results under complex and variable geological conditions. The model optimization step automatically adjusts the parameters through the IPSO algorithm, so that the model converges faster and has higher prediction ability and stability, thereby significantly improving the overall efficiency and safety of shield construction.

[0132] In an exemplary embodiment, in order to verify the effectiveness of the slicing method, four models, namely Random Forest (RF), Feedforward Neural Network (BP), LSTM, and IPSO+LSTM (the present invention), are used to compare the tool wear prediction effects of four different geological information processing schemes.

[0133] Specifically, a dataset containing only cutter placement data and excavation data but not geological data is input into the cutter wear prediction model. In the same excavation stage, the UCS of cutters with different radii is the same, that is, the geological parameters are consistent. In one embodiment, only 12 shield operation parameters (such as torque, thrust, propulsion speed, etc.) and cutter structure information are used as input, but geological parameters are not included. For example, a cutter is used continuously from the 1st ring to the nth ring, and the jth parameter of the ith ring is set to , the hob corresponding model parameter input matrix, as expressed in formula (12),

[0134]

[0135] Each row of the matrix can be regarded as a continuous change sequence of a parameter during the excavation process. The number of matrix rows corresponds to the input dimension of the cutter wear prediction model. The matrix is ​​input into the LSTM or IPSO+LSTM model. For the RF and BP models, due to the difference in the model input structure, each row of X needs to be summed and averaged. The matrix obtained after summing and averaging is shown in formula (13). The summed and averaged matrix is ​​used as the input of RF and BP.

[0136]

[0137] Specifically, Figure 7 As shown in (a), if the B roller cutter is used for excavation from the 3rd stage to the 5th stage (i.e., the 90th ring to the 203rd ring), and there are three geological survey results T3, T4, and T5 in this interval, then the geological input of the roller cutter is the sum and average of the above three values, and the statistics of the geological parameters are not accurate enough.

[0138] Specifically, the cutter head is first divided into three functional areas: center cutter, front cutter, and edge cutter. Figure 7 (b). Each cutterhead section has its corresponding working area during excavation. Figure 7 (c) Then, taking the excavation stage as the statistical unit, the volume ratio of medium-weathered granite and slightly-weathered granite crushed during excavation in different cutterhead areas in each stage was counted. Figure 7 (d), the proportion of moderately weathered granite in the center, front and edge of stage 5 is 65%, 62% and 47% respectively, and the rest is the volume proportion of slightly weathered granite. At this time, based on the original exploration core experimental data and combined with the results of stratigraphic slices, the geological data corresponding to different stages and different cutterhead areas are obtained. The obtained geological data are shown in Table 1. Taking UCS as an example, UCS C , UCS F , UCS E They represent the UCS corresponding to the center, front, and edge areas of the cutter disc at each stage.

[0139]

[0140] Table 1

[0141] For some disc cutters with continuous excavation distances of 2 or more stages, further summation statistics are performed in the horizontal direction as follows:

[0142]

[0143] Among them, zone represents the working area; and They represent the beginning and end stages of tool life respectively; count(i) represents the total number of stages. At this point, by combining with the geological profile, the piecewise discrete UCS test data has been converted into a continuous data set.

[0144] Specifically, based on the data in Table 1, the geological data corresponding to each roller cutter is obtained by weighted average calculation. The weight calculation is the number of rings in this stage / the number of rings in the entire excavation section. Based on this, the geological parameter calculation method of this scheme can be obtained as shown in formula (15), where ring stage(i) Represents the number of rings in the current i-th stage; Sum_ring is the total number of rings experienced by the hob during its entire life cycle.

[0145]

[0146] Specifically, the tool wear prediction effects of four different geological information processing schemes are compared as follows: Figure 8As shown in the figure, Scheme 1, which ignores geological parameters to predict tool wear during shield tunneling, performs the worst in the entire group of comparisons; Scheme 2 simply sums and averages the geological parameters, which can improve the prediction accuracy to a certain extent. However, based on the data of the stratigraphic slices of the present invention, both Scheme 3 and Scheme 4 are better than the above two schemes in the prediction results of various models. Among them, the combination of IPSO+LSTM and Scheme 3 performs best. It reached 0.70, which is 12% higher than the 0.58 of the combination of BP and Scheme 1. It can be seen that the combination of the LSTM model optimized by IPSO and the formation slicing method has achieved ideal results in the application of predicting tool wear. If converted into the mean absolute error (MAE) evaluation method, the error between the predicted value and the measured value is about ±0.49mm, which is close to the manual measurement error of ±0.5mm.

[0147] In this embodiment, the tool wear prediction effects of four different geological information processing solutions are compared. The processing of slice data can make the prediction of the cutter wear prediction model more accurate. At the same time, due to the characteristics of IPSO+LSTM, the cutter wear prediction can be output in real time according to the real-time advancing shield parameters and geological information. The reliable advantages of the combination of stratum slicing and IPSO+LSTM will be further highlighted in excavation projects with more complex and changeable stratum conditions and larger shield machine diameters.

[0148] It should be understood that, although the various steps in the flowcharts involved in the above-mentioned embodiments are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-mentioned embodiments can include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.

[0149] Based on the same inventive concept, the embodiment of the present application also provides a shield machine cutter wear prediction device for implementing the shield machine cutter wear prediction method involved above. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the above method, so the specific limitations in one or more shield machine cutter wear prediction device embodiments provided below can refer to the limitations of the shield machine cutter wear prediction method above, and will not be repeated here.

[0150] In an exemplary embodiment, Fig. 9 As shown, a shield machine cutter wear prediction device is provided, comprising: a geological information analysis module 910, a geological information slice generation module 920, a geological data calculation module 930 and a cutter wear prediction module 940, wherein:

[0151] The geological information analysis module 910 is used to determine the geological information map of the shield machine cutter operation environment, and obtain the cutter placement data and excavation data corresponding to the shield machine cutter under the geological information map;

[0152] The geological information slice generation module 920 is used to horizontally slice the geological information map according to the cutter placement data of the shield machine cutter to obtain the horizontal slice of the geological information map corresponding to the geological information map, and to vertically slice the geological information map according to the excavation data of the shield machine cutter to obtain the vertical slice of the geological information map corresponding to the geological information map;

[0153] The geological data calculation module 930 is used to determine the stratigraphic data corresponding to the horizontal slice of the geological information map according to the vertical slice of the geological information map, and calculate the geological data corresponding to the shield machine cutter operation according to the horizontal slice of the geological information map and the stratigraphic data corresponding to the horizontal slice of the geological information map;

[0154] The cutter wear prediction module 940 is used to input the cutter placement data, the excavation data and the geological data into the cutter wear prediction model to obtain the cutter wear prediction result output by the cutter wear prediction model.

[0155] In one embodiment, the geological information slice generation module 920 is also used to horizontally slice the geological information map according to the cutter placement data of the shield machine cutter to obtain horizontal slices of the geological information map corresponding to the geological information map, including: horizontally slice the geological information map according to the cutter head functional partition according to the cutter placement data to obtain horizontal slices of the geological information map corresponding to the geological information map; or horizontally slice the geological information map according to the cutter placement radius of the cutter placement data to obtain horizontal slices of the geological information map corresponding to the geological information map.

[0156] In one embodiment, the geological information slice generation module 920 is also used to vertically slice the geological information map according to the excavation data of the shield machine cutter to obtain vertical slices of the geological information map corresponding to the geological information map, including: vertically slice the geological information map according to the excavation ring interval of the excavation data to obtain vertical slices of the geological information map corresponding to the geological information map; or vertically slice the geological information map according to a single excavation ring of the excavation data to obtain vertical slices of the geological information map corresponding to the geological information map.

[0157] In one embodiment, the cutter wear prediction module 940 is also used for the cutter wear prediction model to be trained through the following steps, including: inputting historical cutter placement data, historical excavation data and historical geological data into the cutter wear prediction model to be trained, and the cutter wear prediction model determines the operating characteristics corresponding to the cutter according to the historical operating data and historical geological data, and determines the wear amount prediction result according to the operating characteristics; determines the model error according to the wear amount prediction result and the actual cutter wear amount, adjusts the cutter wear prediction model parameters through the particle swarm algorithm, and obtains the cutter wear prediction model when the model error meets the training end condition.

[0158] In one embodiment, the roller cutter wear prediction module 940 is also used to adjust the roller cutter wear prediction model parameters through the particle swarm algorithm, including: generating a particle swarm; each particle in the particle swarm represents the roller cutter wear prediction model parameters; determining the inertia weight and fitness threshold of the particle swarm under the current number of iterations; the inertia weight is negatively correlated with the number of iterations, and the fitness threshold is positively correlated with the number of iterations; according to the inertia weight and the fitness threshold, determining the particle swarm and the speed and position of the particles after the iteration, and when the iteration end condition is met, determining the parameters corresponding to the global optimal particle in the particle swarm after the iteration as the roller cutter wear prediction model parameters.

[0159] In one embodiment, the geological information analysis module 910 is also used to determine the geological information map when the shield machine is in operation, including: obtaining the geological profile map when the shield machine is in operation and the geological attribute data corresponding to the geological profile map; segmenting different strata in the geological profile map, assigning corresponding numerical labels to each segmented stratum, and binding the numerical labels and the geological attribute data of the corresponding strata to obtain the geological information map in pixel form.

[0160] Each module in the above-mentioned shield machine cutter wear prediction device can be implemented in whole or in part by software, hardware and a combination thereof. Each of the above-mentioned modules can be embedded in or independent of a processor in a computer device in the form of hardware, or can be stored in a memory in a computer device in the form of software, so that the processor can call and execute the operations corresponding to each of the above modules.

[0161] In an exemplary embodiment, a computer device is provided. The computer device may be a terminal, and its internal structure diagram may be as shown in FIG. Fig.10 As shown. The computer device includes a processor, a memory, an input / output interface, a communication interface, a display unit and an input device. The processor, the memory and the input / output interface are connected through a system bus, and the communication interface, the display unit and the input device are connected to the system bus through the input / output interface. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The input / output interface of the computer device is used to exchange information between the processor and the external device. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be realized through WIFI, a mobile cellular network, near field communication (NFC) or other technologies. When the computer program is executed by the processor, a shield machine cutter wear prediction method is realized. The display unit of the computer device is used to form a visually visible picture, which can be a display screen, a projection device or a virtual reality imaging device. The display screen can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device can be a touch layer covering the display screen, or a button, trackball or touchpad set on the computer device shell, or an external keyboard, touchpad or mouse.

[0162] Those skilled in the art will understand that Fig.10 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0163] In one embodiment, a computer device is provided, including a memory and a processor, wherein a computer program is stored in the memory, and the processor implements the steps in the above-mentioned method embodiments when executing the computer program.

[0164] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps in the above-mentioned method embodiments are implemented.

[0165] In one embodiment, a computer program product is provided, including a computer program, which implements the steps in the above method embodiments when executed by a processor.

[0166] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant regulations.

[0167] A person of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment method can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. Non-relational databases may include distributed databases based on blockchains, etc., but are not limited to this. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, an artificial intelligence (AI) processor, etc., but are not limited to this.

[0168] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0169] The above-described embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. A method for predicting cutter wear of a shield machine, characterized in that: The method comprises: Determine a geological information map of the shield machine cutter operating environment, and obtain cutter placement data and excavation data corresponding to the shield machine cutter under the geological information map; According to the cutter placement data of the shield machine cutter, the geological information map is horizontally segmented to obtain horizontal slices of the geological information map corresponding to the geological information map, and, according to the excavation data of the shield machine cutter, the geological information map is vertically segmented to obtain vertical slices of the geological information map corresponding to the geological information map; According to the vertical slice of the geological information map, the stratigraphic data corresponding to the horizontal slice of the geological information map is determined, and according to the horizontal slice of the geological information map and the stratigraphic data corresponding to the horizontal slice of the geological information map, the geological data corresponding to the shield machine cutter operation is calculated; The cutter placement data, the excavation data and the geological data are input into a cutter wear prediction model to obtain a cutter wear prediction result output by the cutter wear prediction model.

2. The method according to claim 1, characterized in that The step of horizontally slicing the geological information map according to the cutter placement data of the shield machine cutter to obtain a horizontal slice of the geological information map corresponding to the geological information map includes: According to the cutter head function partition of the cutter arrangement data, the geological information map is horizontally segmented to obtain the geological information map horizontal slices corresponding to the geological information map; or According to the hob cutting radius of the hob cutting data, the geological information map is horizontally segmented to obtain a geological information map horizontal slice corresponding to the geological information map.

3. The method according to claim 1, characterized in that: The step of vertically slicing the geological information map according to the excavation data of the shield machine cutter to obtain a vertical slice of the geological information map corresponding to the geological information map includes: According to the excavation ring interval of the excavation data, the geological information map is vertically segmented to obtain a vertical slice of the geological information map corresponding to the geological information map; or, The geological information map is vertically segmented according to a single excavation ring of the excavation data to obtain a vertical slice of the geological information map corresponding to the geological information map.

4. The method according to claim 1, characterized in that: The hob wear prediction model is trained by the following steps, including: Inputting historical disc cutter placement data, historical excavation data and historical geological data into a disc cutter wear prediction model to be trained, wherein the disc cutter wear prediction model determines the operation characteristics corresponding to the disc cutter according to the historical operation data and historical geological data, and determines the wear amount prediction result according to the operation characteristics; The model error is determined according to the wear prediction result and the actual hob wear, and the hob wear prediction model parameters are adjusted by particle swarm algorithm. When the model error meets the training end condition, the hob wear prediction model is obtained.

5. The method according to claim 4, characterized in that The method of adjusting the parameters of the hob wear prediction model by using a particle swarm algorithm includes: Generate a particle swarm; each particle in the particle swarm represents the parameters of the hob wear prediction model; Determine the inertia weight and fitness threshold of the particle swarm under the current number of iterations; the inertia weight is negatively correlated with the number of iterations, and the fitness threshold is positively correlated with the number of iterations; According to the inertia weight and the fitness threshold, the particle group and the speed and position of the particles after iteration are determined. When the iteration end condition is met, the parameters corresponding to the global optimal particle in the particle group after iteration are determined as the parameters of the hob wear prediction model.

6. The method according to claim 1, characterized in that The geological information map for determining the shield machine operation includes: Acquire a geological profile during shield machine operation and geological attribute data corresponding to the geological profile; Different strata in the geological profile are segmented, corresponding numerical labels are assigned to the segmented strata, and the numerical labels are bound to the geological attribute data of the corresponding strata to obtain the geological information map in pixel form.

7. A shield machine cutter wear prediction device, characterized in that: The device comprises: A geological information analysis module is used to determine a geological information map of the shield machine cutter operating environment, and obtain cutter placement data and excavation data corresponding to the shield machine cutter under the geological information map; A geological information slice generation module is used to horizontally slice the geological information map according to the cutter placement data of the shield machine cutter to obtain horizontal slices of the geological information map corresponding to the geological information map, and to vertically slice the geological information map according to the excavation data of the shield machine cutter to obtain vertical slices of the geological information map corresponding to the geological information map; A geological data calculation module is used to determine the stratigraphic data corresponding to the horizontal slice of the geological information map according to the vertical slice of the geological information map, and calculate the geological data corresponding to the shield machine cutter operation according to the horizontal slice of the geological information map and the stratigraphic data corresponding to the horizontal slice of the geological information map; The cutter wear prediction module is used to input the cutter placement data, the excavation data and the geological data into the cutter wear prediction model to obtain the cutter wear prediction result output by the cutter wear prediction model.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 6 are implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images