A method for constructing a tool wear law model

Abstract

This invention discloses a method for constructing a tool wear law model. This invention relates to the field of shield tunneling tool wear law research, addressing the problems of existing methods having complex model construction, susceptibility to fluctuations in working conditions during practical applications, and insufficient predictive stability. This invention obtains actual engineering tool wear data at tool replacement points, and uses K-means clustering and quadratic fitting to establish an actual engineering wear law model; based on the CSM model of rock-breaking force of approximately constant cross-section roller cutters and the Rabinowicz abrasive wear equation, a theoretical model of the wear law is established; for the blade yield strength in the theoretical model, a blade yield strength variation model is established; the blade width variation model and the blade yield strength variation model are added to the theoretical model to obtain a corrected tool wear law model. This model can provide accurate data support for tool replacement cycle planning in shield tunneling construction.

Description

Technical Field

[0001] This invention relates to the field of research on the wear law of tunnel boring machine cutters, and in particular to a method for constructing a model of cutter wear law. Background Technology

[0002] Shield tunneling, as the mainstream construction method for underground engineering projects such as subway tunnels, is widely used due to its advantages such as high safety, good construction efficiency, minimal impact on the surrounding environment, and strong adaptability to complex geological conditions. The cutterhead is a key tool for the shield machine to directly break rocks during tunneling, and its performance directly affects the progress and cost of the entire project. In recent years, with increasingly complex engineering geological conditions, especially in construction in composite strata of soft and hard rock, cutter wear has intensified, the frequency of cutter replacement has increased significantly, and construction risks have also increased. Therefore, accurately understanding the wear patterns of the cutterhead is of great significance for making scientific cutter replacement decisions and ensuring construction safety and efficiency.

[0003] Currently, research on the wear patterns of tunnel boring machine (TBM) cutters mainly focuses on analyzing the impact of tunneling parameters (such as thrust, torque, and tunneling speed) on cutter wear, predicting cutter life by establishing a correlation model between tunneling parameters and wear amount. These methods typically require comprehensive consideration of the interaction of multiple dynamic parameters, resulting in complex model construction and susceptibility to fluctuations in operating conditions in practical applications, leading to insufficient predictive stability. This limitation restricts the practicality and adaptability of cutter wear prediction models, especially when facing complex and variable geological conditions. Existing methods often rely on numerous parameter calibrations and operating condition assumptions, resulting in cumbersome calculations and prone to prediction deviations, affecting the accuracy of cutter replacement timing. Therefore, it is necessary to propose a method for constructing a cutter wear pattern model to address these issues. Summary of the Invention

[0004] The purpose of this invention is to provide a method for constructing a tool wear law model, so as to solve the problems of existing methods having complex model construction, being easily affected by working condition fluctuations in practical applications, and having insufficient prediction stability.

[0005] This invention provides a method for constructing a tool wear law model, comprising the following steps:

[0006] Step 1: Obtain actual tool change point wear data. Based on the wear data, use K-means clustering and quadratic interpolation to establish an actual engineering wear pattern model.

[0007] Step 2: Based on the CSM model of rock-breaking force of a roller cutter with an approximate constant cross section and the Rabinowicz abrasive wear equation, a theoretical model of wear law is established. The theoretical model includes the following parameters: blade width and blade yield strength.

[0008] Step 3: Establish a cutting edge width variation model for the cutting edge width in the theoretical model; establish a cutting edge yield strength variation model for the cutting edge yield strength in the theoretical model; add the cutting edge width variation model and the cutting edge yield strength variation model to the theoretical model to obtain the corrected tool wear law model.

[0009] Furthermore, in step one, the actual tool change point wear data includes: the wear of the tool removed during tool change. Wear of the replaced cutting tools And the tool change time, among which Representing the Next, change the blade.

[0010] The steps for establishing a model of actual engineering wear patterns include: calculating the wear amount of the cutter between two cutter change points based on the wear amount of the cutter removed and the wear amount of the cutter installed during cutter change; calculating the tunneling distance between two cutter change points by matching the cutter change time with the tunneling mileage; determining the rock-breaking stroke of each cutter based on the tunneling distance and the cutter installation radius; calculating the cutter wear rate based on the cutter wear amount and the rock-breaking stroke; establishing a wear-wear rate scatter plot; dividing the data into n segments according to the wear amount, performing K-means clustering on each segment to obtain cluster points; and performing a second fitting on the n cluster points to obtain the fitting function, thus obtaining the model of actual engineering wear patterns.

[0011] Furthermore, the amount of wear between the two tool change points for:

[0012]

[0013] in, For the first The wear and tear on the knife that was replaced this time. For the first The amount of wear on the tool removed during each tool change. For the first The amount of wear on the tool replaced during each tool change.

[0014] tunneling distance between two tool change points for:

[0015]

[0016] in, For the first Compared to the distance the shield tunneled during the second cutterhead change, the distance traveled during the third cutterhead change was [missing information]. For the first The tunneling distance of the shield machine during the next cutterhead change For the first The tunneling distance of the shield machine during the next cutterhead replacement.

[0017] Rock Breaking Journey for:

[0018]

[0019] in, For the first During the second tool change The path of the knife breaking through the rocks, For the first The installation radius of the blade, For penetration degree.

[0020] Tool wear rate :

[0021]

[0022] in, For a knife The wear rate of the tool that was replaced during the next tool change.

[0023] Furthermore, by matching the cutterhead replacement time with the tunneling mileage, the tunneling distance between the two cutterhead replacement points is calculated, including: according to the cutterhead replacement time, the tunneling distance of the shield tunnel at the two cutterhead replacement time points is obtained by looking up a table, and then the difference is calculated to obtain the tunneling distance between the two cutterhead replacement points.

[0024] Furthermore, a wear-wear rate scatter plot is established, including: plotting the wear rate of a certain tool at the [missing information]. Wear of the tool replaced in the second tool change With the wear rate of the knife Composition data pairs ,by For independent variable, Create a scatter plot of tool change point data for all tools on the tool turret as the dependent variable.

[0025] Furthermore, the data is subdivided into n segments based on wear amount, and K-means clustering is performed on each segment to obtain cluster points. The n cluster points are then subjected to a second-order fitting to obtain a fitting function, resulting in a model of the actual engineering wear pattern, including: subdividing the data according to wear amount... For each segment of data, perform K-means clustering to obtain cluster points. The independent variable is divided into different intervals, and the representative value of the wear amount interval is found by K-means clustering in the interval.

[0026] The cluster points are connected using a quadratic fitting method, and the resulting curve is a model of the actual wear law in engineering.

[0027] Furthermore, in step two, the following theoretical model is established:

[0028]

[0029] in, The wear rate is the theoretical model. The abrasive wear coefficient, The distance between the blades. For rock compressive strength, For penetration, The initial blade width, Where is the hob radius, The yield strength of the cutting edge.

[0030] Furthermore, in step three, a blade width variation model is established for the blade width in the theoretical model, including:

[0031] For the parameter blade width in the theoretical model Establish the following change model:

[0032]

[0033] in, The initial blade width, It is half the cutting edge angle of the hobbing cutter. This represents the wear of the cutting tool.

[0034] Furthermore, in step three, a blade yield strength variation model is established for the blade yield strength in the theoretical model, including: for the blade yield strength parameter in the theoretical model... ,Establish Piecewise variation model:

[0035] j={1,2,3…,n}

[0036] Where n is the number of K-means cluster points. To obtain the minimum value, To obtain the maximum value.

[0037] Further, in step three, the cutting edge width variation model and the cutting edge yield strength variation model are added to the theoretical model to obtain a modified tool wear law model, including: adding the cutting edge width variation model and the cutting edge yield strength variation model to the theoretical model to obtain the modified model as follows:

[0038] j={1,2,3…,n}

[0039] in, The wear rate of the corrected model, This refers to the amount of wear. The abrasive wear coefficient, The distance between the blades. For rock compressive strength, For penetration, The initial blade width, It is half the cutting edge angle of the hobbing cutter. Where is the hob radius, The yield strength of the cutting edge.

[0040] The beneficial effects of this invention are as follows: The method for constructing the tool wear law model of this invention can obtain a relatively universal relationship between tool wear amount and wear rate. By obtaining the wear rate from the wear amount of the tool at a certain moment, the wear trend of the tool can be predicted. This provides accurate data support for the planning of tool replacement cycle in shield tunneling construction, effectively avoiding the waste of engineering costs caused by premature replacement, or the risks of tool failure, reduced tunneling efficiency, or even equipment damage caused by delayed replacement. Attached Figure Description

[0041] To more clearly illustrate the technical solution of the present invention, the drawings used in the embodiments will be briefly introduced below. Obviously, those skilled in the art can obtain other drawings based on these drawings without creative effort.

[0042] Figure 1 This is a schematic diagram of the tool wear law model of the present invention.

[0043] Figure 2 This is a schematic diagram comparing the results of the three models of this invention. Detailed Implementation

[0044] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below in conjunction with specific embodiments and corresponding drawings. Obviously, the described embodiments are only a part of the embodiments of this invention, and not all of them. Based on the embodiments of this invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this invention. The technical solutions provided by various embodiments of this invention will be described in detail below with reference to the accompanying drawings.

[0045] Please see Figure 1 The present invention provides a method for constructing a tool wear law model, comprising the following steps:

[0046] Step 1: Obtain actual tool change point wear data. Based on the wear data, use K-means clustering and quadratic interpolation to establish an actual engineering wear pattern model.

[0047] Specifically, actual tool change point wear data in engineering includes: the wear of the tool removed during tool change. Wear of the replaced cutting tools And the tool change time, among which Representing the Next, change the blade.

[0048] The steps for establishing a wear law model for actual engineering include: calculating the tool wear between two tool change points based on the wear of the removed tool and the wear of the installed tool during tool change; and calculating the wear between the two tool change points. for:

[0049]

[0050] in, For the first The wear and tear on the knife that was replaced this time. For the first The amount of wear on the tool removed during each tool change. For the first The amount of wear on the tool replaced during each tool change.

[0051] By matching the cutter change time with the tunneling mileage, the tunneling distance between two cutter change points can be calculated; the tunneling distance between two cutter change points for:

[0052]

[0053] in, For the first Compared to the distance the shield tunneled during the second cutterhead change, the distance traveled during the third cutterhead change was [missing information]. For the first The tunneling distance of the shield machine during the next cutterhead change For the first The tunneling distance of the shield machine at the time of the cutterhead change is calculated. Based on the cutterhead change time, the tunneling distance of the shield machine at the two cutterhead change times is obtained by looking up a table, and then the difference is calculated to obtain the tunneling distance between the two cutterhead change points.

[0054] The rock-breaking stroke of each cutter is determined based on the tunneling distance and the cutter installation radius; rock-breaking stroke for:

[0055]

[0056] in, For the first During the second tool change The path of the knife breaking through the rocks, For the first The installation radius of the blade, For penetration degree.

[0057] The tool wear rate is calculated based on the tool wear and rock-breaking stroke; tool wear rate :

[0058]

[0059] in, For a knife The wear rate of the tool that was replaced during the next tool change.

[0060] Establish a wear-wear rate scatter plot; where a certain tool's wear rate is... Wear of the tool replaced in the second tool change With the wear rate of the knife Composition data pairs ,by For independent variable, Create a scatter plot of tool change point data for all tools on the tool turret as the dependent variable.

[0061] The data is divided into n segments based on wear amount, and K-means clustering is performed on each segment to obtain cluster points; where the data is divided into... For each segment of data, perform K-means clustering to obtain cluster points. The independent variable is divided into different intervals, and the representative value of the wear amount interval is found by K-means clustering method in the interval.

[0062] A quadratic fitting is performed on the n cluster points to obtain the fitting function, thus yielding the actual engineering wear law model. Specifically, the curve obtained by connecting the cluster points using the quadratic fitting method represents the actual engineering wear law model. The actual engineering wear law model is as follows:

[0063]

[0064] in, The wear rate represents the wear rate of the actual engineering wear law model. This represents the wear amount in a model representing the actual wear patterns in engineering projects. This represents the mapping relationship between the two, i.e., the fitting function.

[0065] Step 2: Based on the CSM model of rock-breaking force of a constant cross-section roller cutter and the Rabinowicz abrasive wear equation, a theoretical model of wear law is established. The theoretical model includes the following parameters: blade width and blade yield strength.

[0066] The following theoretical model is established:

[0067]

[0068] in, The wear rate is the theoretical model. The abrasive wear coefficient, The distance between the blades. For rock compressive strength, For penetration, The initial blade width, Let the hob radius be... The yield strength of the cutting edge.

[0069] Step 3: Establish a cutting edge width variation model for the cutting edge width in the theoretical model; establish a cutting edge yield strength variation model for the cutting edge yield strength in the theoretical model; add the cutting edge width variation model and the cutting edge yield strength variation model to the theoretical model to obtain the corrected tool wear law model.

[0070] Specifically, for the parameter blade width in the theoretical model Establish the following change model:

[0071]

[0072] in, The initial blade width, It is half the cutting edge angle of the hobbing cutter. This represents the wear of the cutting tool.

[0073] Specifically, regarding the blade yield strength parameter in the theoretical model... ,Establish Piecewise variation model:

[0074] j={1,2,3…,n}

[0075] Where n is the number of K-means cluster points. To obtain the minimum value, To obtain the maximum value.

[0076] Specifically, by incorporating the blade width variation model and the blade yield strength variation model into the theoretical model, the modified model is obtained as follows:

[0077] j={1,2,3…,n}

[0078] in, The wear rate of the corrected model, This refers to the amount of wear. The abrasive wear coefficient, The distance between the blades. For rock compressive strength, For penetration, The initial blade width, It is half the cutting edge angle of the hobbing cutter. Where is the hob radius, The yield strength of the cutting edge.

[0079] When processing actual engineering data, a scatter plot is drawn, and the data is divided into 7 clusters to obtain cluster centers before a fitted curve is obtained. In the theoretical model, based on actual underground engineering parameters, the abrasive wear coefficient... Usually taken Dimensionless; tool spacing Pick Rock compressive strength Pick Penetration Pick Blade width Pick hob radius Pick ; blade yield strength Pick In the revised model, Take the x-coordinate value corresponding to the cluster center. The value range is determined based on engineering experience. Dimensions are Substituting the above engineering data into the model, we obtain the following: Figure 2 The results are shown.

[0080] In summary, the two variation models introduced in this invention significantly improve the original theoretical model and can be used as applicable models for predicting wear data.

[0081] The embodiments of the present invention described above do not constitute a limitation on the scope of protection of the present invention.

Claims

1. A method for constructing a tool wear law model, characterized in that, Includes the following steps: Step 1: Obtain actual engineering tool change point wear data. Based on the wear data, use K-means clustering and quadratic fitting to establish an actual engineering wear pattern model. Step 2: Based on the CSM model of rock-breaking force of an approximate constant cross-section roller cutter and the Rabinowicz abrasive wear equation, a theoretical model of wear law is established: in, The wear rate is the theoretical model. The abrasive wear coefficient, The distance between the blades. For rock compressive strength, For penetration, The initial blade width, Where is the hob radius, The yield strength of the blade; Step 3: For the parameter blade width in the theoretical model Establish a model for the variation of blade width: in, The initial blade width, It is half the cutting edge angle of the hobbing cutter. This represents the amount of wear on the cutting tool; For the parameter blade yield strength in the theoretical model Establish a model for the variation of the blade's yield strength: j={1,2,3…,n} Where n is the number of K-means cluster points. To obtain the minimum value, To obtain the maximum value; By incorporating the cutting edge width variation model and the cutting edge yield strength variation model into the theoretical model, a modified tool wear law model is obtained.

2. The method for constructing a tool wear law model according to claim 1, characterized in that, In step one, the actual tool change point wear data includes: the wear of the tool removed during tool change. Wear of the replaced cutting tools And the tool change time, among which Representing the Next tool change; The steps for establishing a model of actual engineering wear patterns include: Calculate the tool wear between two tool change points based on the wear of the removed tool and the wear of the installed tool during tool change. By matching the cutter change time with the tunneling mileage, the tunneling distance between two cutter change points can be calculated; The rock-breaking stroke of each cutter is determined based on the tunneling distance and the cutter installation radius. The tool wear rate is calculated based on the wear of the tool and the rock-breaking stroke. Create a wear-wear rate scatter plot; The data is divided into n segments based on wear amount, and K-means clustering is performed on each segment to obtain cluster points. By performing a second-order fitting on the n cluster points, a fitting function is obtained, which leads to a model of the actual engineering wear law.

3. The method for constructing a tool wear law model according to claim 2, characterized in that, Wear between two tool change points for: in, For the first The wear and tear on the knife that was replaced this time. For the first The amount of wear on the tool removed during each tool change. For the first The wear of the tool replaced in the next tool change; tunneling distance between two tool change points for: in, For the first The distance the shield tunneled during the next cutterhead change compared to the last cutterhead change. For the first The tunneling distance of the shield machine during the next cutterhead change For the first The tunneling distance of the shield machine during the next cutterhead change; Rock Breaking Journey for: in, For the first During the second tool change The path of the knife breaking through the rocks, For the first The installation radius of the blade, Penetration degree; Tool wear rate : in, For a knife The wear rate of the tool that was replaced during the next tool change.

4. The method for constructing a tool wear law model according to claim 3, characterized in that, By matching the cutter change time with the tunneling mileage, the tunneling distance between two cutter change points is calculated, including: Based on the cutterhead replacement time, the tunneling distance at the two cutterhead replacement time points is obtained by looking up the table, and then the difference is calculated to obtain the tunneling distance between the two cutterhead replacement points.

5. The method for constructing a tool wear law model according to claim 4, characterized in that, Create a wear-wear rate scatter plot, including: A certain knife Wear of the tool replaced in the second tool change With the wear rate of the knife Composition data pairs ,by For independent variable, Create a scatter plot of tool change point data for all tools on the tool turret as the dependent variable.

6. The method for constructing a tool wear law model according to claim 5, characterized in that, The data is divided into n segments based on wear amount, and K-means clustering is performed on each segment to obtain cluster points. By performing a second-order fitting on the n cluster points, a fitting function is obtained, leading to a model of the actual engineering wear law, including: According to wear amount, it is further divided into For each segment of data, perform K-means clustering to obtain cluster points. The independent variable is divided into different intervals, and the representative value of the wear amount interval is found by K-means clustering in the interval. The cluster points are connected using a quadratic fitting method, and the resulting curve is a model of the actual wear law in engineering.

7. The method for constructing a tool wear law model according to claim 6, characterized in that, In step three, the cutting edge width variation model and the cutting edge yield strength variation model are added to the theoretical model to obtain the modified tool wear law model, including: By incorporating the blade width variation model and the blade yield strength variation model into the theoretical model, the modified model is obtained as follows: j={1,2,3…,n} in, The wear rate of the corrected model, This refers to the amount of wear. The abrasive wear coefficient, The distance between the blades. For rock compressive strength, For penetration, The initial blade width, It is half the cutting edge angle of the hobbing cutter. Where is the hob radius, The yield strength of the cutting edge.

Images