Abrasion form prediction method and abrasion tester thereof

Abstract

The invention discloses a wear form prediction method and a wear testing machine thereof. The prediction method comprises the following steps: carrying out a wear experiment to obtain three kinds of test data: friction coefficient, temperature and noise; carrying out label coding on various wear forms; respectively extracting time domain features of the three test data, converting the time domain features into frequency domain data, and finally extracting frequency domain features of the frequency domain data; the method comprises the following steps: firstly forming a data set and generating a plurality of training subsets, then constructing decision trees and repeatedly obtaining a plurality of decision trees, finally selecting a majority prediction category as a final result, and obtaining a preliminary prediction function; judging whether modes exist in the three preliminary prediction functions or not, if yes, judging that the modes are final prediction functions, and if not, establishing a weighting function to determine the final prediction functions; and wear form prediction is carried out according to the final prediction function. According to the invention, data analysis can be carried out in real time while a friction wear test is carried out, and all wear forms of a material in the whole working condition process are predicted.

Description

Technical Field

[0001] The present invention relates to a prediction method in the technical field of mechanical wear detection, in particular to a wear form prediction method, and also to a wear testing machine. Background Art

[0002] Currently, to investigate the potential wear patterns of a specific material under given operating conditions, simulation experiments are typically conducted using a friction and wear testing machine. However, existing friction and wear testing machines themselves lack the ability to predict wear patterns. After the test, wear scar morphology analysis using a scanning electron microscope (SEM) or other observation equipment is required to determine the specific wear pattern. This method is cumbersome and does not allow for direct and convenient identification of wear patterns during or after the test.

[0003] Furthermore, in real-world operating environments, material wear patterns often evolve dynamically over time. Different wear patterns can occur at different stages, such as one type of wear in the initial stage triggering other wear patterns in subsequent stages. Post-test observation using existing friction and wear testing machines only captures the final wear morphology at the end of the test, failing to fully reveal and record the various wear patterns that occur sequentially or alternately throughout the friction and wear process, as well as their evolutionary patterns. Summary of the Invention

[0004] In order to solve the technical problem that the existing friction and wear test cannot predict the wear form and needs to use scanning electron microscope and other equipment to judge the wear form, the present invention provides a wear form prediction method and a wear testing machine.

[0005] The present invention is implemented by the following technical solution: a wear form prediction method, which includes the following steps:

[0006] Wear experiments were conducted to obtain three types of test data: friction coefficient, temperature, and noise;

[0007] Label and encode various wear forms;

[0008] First, the time domain features of the three test data are extracted respectively, then the time domain features are converted into frequency domain data, and finally the frequency domain features of the frequency domain data are extracted;

[0009] First, the time domain / frequency domain features of the three experimental data and the wear form labels are encoded to form a data set and generate multiple training subsets. Then, a decision tree is constructed for each training subset and the process is repeated to obtain multiple decision trees. Each decision tree is recursively split from the root node until the node meets a preset stopping condition and is marked as a leaf node. Finally, the category with the majority of predictions is selected as the final result, and a preliminary prediction function is obtained.

[0010] Determine whether the three preliminary prediction functions have a mode, if so, determine the mode of the three preliminary prediction functions as the final prediction function, otherwise establish a weighted function to determine the final prediction function;

[0011] Wear form prediction is performed according to the final prediction function.

[0012] The present invention first obtains three types of test data, namely friction coefficient, temperature, and noise, through experiments. It then extracts features from each set of test data using time-domain and frequency-domain feature extraction methods, and corresponds each set to a wear form. A data set is then constructed based on the data features and their corresponding wear forms. A decision tree is then constructed and trained on the data to obtain three preliminary prediction functions. A final prediction function is then obtained based on whether the three preliminary prediction functions have a mode, and a corresponding final prediction model is obtained. Finally, the wear form is predicted using the final prediction model. This solves the technical problem that existing friction and wear tests cannot predict wear forms and require the use of equipment such as scanning electron microscopes to determine wear forms. The present invention can collect test data in real time, perform data analysis in real time while conducting friction and wear tests, and predict all wear forms of the material throughout the entire operating process. This can reduce the use of test equipment, save time and cost, and provide a reference for product design, manufacturing, and engineering applications.

[0013] As a further improvement of the above solution, the time domain features include the peak value f X,1 , mean f X,2 , root mean square error f X,3 , standard deviation f X,4 , kurtosis f X,5 , skewness f X,6 , margin index f X,7 , Waveform Index f X,8 , pulse index f X,9 ; The frequency domain features include the frequency mean F S,1 , frequency center of gravity F S,2 , frequency root mean square F S,3 , frequency variance F S,4 ; Wherein, X∈{μ, Te, No}, μ is the friction coefficient, Te is the temperature, and No is the noise.

[0014] Furthermore, the calculation formula of the time domain feature is:

[0015] f X，1 =max(|X(t)|)

[0016]

[0017] Wherein, X(t) represents the time domain data of the test data, t represents time, T represents the end time point, and n represents the number of valid samples.

[0018] Furthermore, the frequency domain data is S(k), and the calculation formula is:

[0019]

[0020] The calculation formula of the frequency domain feature is:

[0021]

[0022]

[0023] Among them, f k is the frequency value of the kth spectral line.

[0024] Furthermore, the wear form is one of adhesive wear, abrasive wear, fatigue wear and oxidative wear, and the four wear forms are coded as 1, 2, 3 and 4;

[0025] Bootstrap sampling is used to generate H training subsets from the data set, each training subset accounts for 2 / 3 of the original data, and the remaining 1 / 3 of the original data is used as out-of-bag data; the decision tree construction method includes the following steps: first, randomly selecting m features when splitting each node of the tree, then selecting the best feature for node splitting based on the Gini index as a splitting criterion, and finally performing recursive splitting, repeating feature selection and optimal splitting point selection until the number of node samples is less than a set threshold or reaches a maximum depth.

[0026] Furthermore, the three datasets of experimental data are:

[0027]

[0028] Among them, D μ 、D Te 、D No data sets representing the friction coefficient, the temperature, and the noise respectively; represents a 13-dimensional feature vector composed of the time domain / frequency domain features; Yi∈{1, 2, 3, 4} represents the category labels of four wear forms.

[0029] Furthermore, the method for determining the final prediction function by the weighting function includes the following steps:

[0030] make Calculate the deviation: in, Represent three preliminary prediction functions respectively;

[0031] Calculate the weighting factors:

[0032] Let the weighting function be Then the final prediction function is Here, round(·) indicates rounding to the nearest integer.

[0033] Furthermore, the preliminary prediction function of the friction coefficient is:

[0034]

[0035] Among them, T b represents the b-th decision tree, B represents the total number of decision trees, and mode(·) represents the mode.

[0036] Furthermore, the correction model predicts in time periods: a time period cycle ΔT is set, and the prediction result of the correction model is checked to see whether it is the same as the prediction result of the previous time period; if the prediction results of the two models are different in the time period hΔT~(h+1)ΔT, the correction model updates the prediction data of the final prediction model, and takes the hΔT moment when the prediction result of the correction model begins to change as the starting moment of the updated final prediction model, and the final prediction model is recalculated; when outputting, the prediction result of the final prediction model before the hΔT moment is output as one of the wear form results, and the prediction result after the hΔT moment is output as another wear form result, and multiple wear forms in time periods during the entire working process are predicted.

[0037] The present invention further provides a wear testing machine, which is applied to any of the above-mentioned wear form prediction methods, and the wear testing machine includes:

[0038] frame;

[0039] a power mechanism mounted on the frame;

[0040] a tooling mounted on the frame;

[0041] a grinding mechanism fixed to the frame; the power mechanism drives the tooling to perform reciprocating motion on the frame, and the tooling drives the test sample to contact the grinding mechanism, causing the test sample to wear;

[0042] a friction sensor mounted on the frame and configured to detect friction generated by wear of the test specimen; the wear testing machine calculating a friction coefficient based on the friction;

[0043] a temperature sensor mounted on the frame and used to detect the temperature generated by wear of the test sample;

[0044] A noise sensor is mounted on the frame and is used to detect noise generated by wear of the test sample.

[0045] Compared with the prior art, the wear form prediction method and wear testing machine of the present invention have the following beneficial effects:

[0046] 1. This wear form prediction method first obtains three types of test data: friction coefficient, temperature, and noise. It then extracts features from each set of test data using time-domain and frequency-domain feature extraction methods, corresponding to wear forms. A data set is then constructed based on the data features and their corresponding wear forms. A decision tree is then constructed and trained on the data to obtain three preliminary prediction functions. A final prediction function is then determined based on whether the three preliminary prediction functions have a mode, and a corresponding final prediction model is obtained. Finally, the wear form is predicted using the final prediction model. This method addresses the technical issue of existing friction and wear tests being unable to predict wear forms and requiring the use of equipment such as scanning electron microscopes to determine wear forms. The method can collect test data in real time, perform data analysis in real time while conducting friction and wear tests, and predict all wear forms of the material throughout the entire operating process. This can reduce the use of test equipment, save time and costs, and provide a reference for product design, manufacturing, and engineering applications.

[0047] 2. The wear testing machine has a fixed grinding mechanism, and the test sample reciprocates with the tooling. The temperature sensor and noise sensor are both aligned with the contact part between the grinding mechanism and the test sample, collecting friction coefficient, temperature and noise data in real time. Finally, the wear form that will occur in a certain material under a certain working condition is predicted through the wear form prediction method. The wear form can be predicted while the test is in progress, improving the prediction efficiency and timeliness. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Figure 1 This is a flow chart of the wear form prediction method according to Example 1 of the present invention.

[0049] Figure 2 This is a flow chart of the wear form prediction method of Example 2 of the present invention.

[0050] Figure 3 This is a schematic structural diagram of a wear testing machine according to Example 3 of the present invention.

[0051] Figure 4 for Figure 3 Schematic diagram of the main structure of the wear testing machine.

[0052] Explanation of symbols:

[0053] 1 Frame 5 Temperature sensor

[0054] 2 Power mechanism 6 Noise sensor

[0055] 3 Tooling 7 Test sample

[0056] 4 grinding mechanism DETAILED DESCRIPTION

[0057] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0058] Example 1

[0059] See also Figure 1 This embodiment provides a wear form prediction method. This method primarily implements wear form prediction for a test specimen through the following steps. It should be noted that in actual application, the test specimen can be the finished product to be tested. This means that the method of this embodiment can be used not only in testing but also in direct product testing.

[0060] The first step is to conduct wear experiments to obtain three types of test data: friction coefficient, temperature, and noise. In the first step of the wear form prediction method, test specimens made of different materials are used. This allows the collection of test data from different materials to eliminate the impact of material differences on the accuracy of the prediction results when extracting features to train the model, thereby developing a wear form prediction method applicable to any material. Similarly, multiple operating conditions are set to eliminate the impact of operating conditions on the accuracy of the prediction results, making the wear form prediction method applicable to any operating condition. Operating condition types include load, relative speed, and test time.

[0061] The second step is to label and encode various wear forms. In this embodiment, the test sample can be observed by an electron microscope or other equipment. According to the morphology of the sample, the wear form is one of adhesive wear, abrasive wear, fatigue wear and oxidative wear, and the four wear forms are labeled and coded as 1, 2, 3, and 4. Among them, adhesive wear refers to the irregular adhesive nodules visible on the surface of the sample, and the local presence of torn edges and pits caused by plastic deformation. Abrasive wear refers to the parallel furrows visible on the surface of the sample, and the appearance of directional scratches. Fatigue wear refers to the appearance of shell-like or fish-scale flaking areas on the surface of the sample, with fatigue cracks on the edges. Oxidative wear refers to the formation of a dense oxide layer on the surface of the sample, with local ruptures. The four wear forms are labeled and coded: adhesive wear → 1, abrasive wear → 2, fatigue wear → 3, and oxidative wear → 4.

[0062] The third step is to extract the time domain features of the three test data respectively, then convert the time domain features into frequency domain data, and finally extract the frequency domain features of the frequency domain data. In this embodiment, the time domain data X(t) is converted into frequency domain data S(k) by Fourier transform. t represents time, T represents the end time point, k represents the kth frequency component, k=0 represents the DC component, k=1 represents the fundamental wave, j represents the imaginary unit, Represents the rotation factor, which is a complex exponential basis function that projects the time domain signal into the frequency domain. The time domain features include the peak value f X,1 , mean f X,2 , root mean square error f X,3 , standard deviation f X,4 , kurtosis f X,5 , skewness f X,6 , margin index f X,7 , Waveform Index f X,8 , pulse index f X,9 The frequency domain features include the frequency mean F S,1 , frequency center of gravity F S,2 , frequency root mean square F S,3 , frequency variance F S,4 Where, X∈{μ, Te, No}, μ is the friction coefficient, Te is the temperature, and No is the noise.

[0063] The calculation formula of time domain characteristics is:

[0064] f X，1 =max(|X(t)|)

[0065]

[0066]

[0067] Where X(t) represents the time domain data of the test data, t represents time, T represents the end time point, and n represents the number of valid samples.

[0068] The frequency domain data is S(k), and the calculation formula is:

[0069]

[0070] The calculation formula of frequency domain characteristics is:

[0071]

[0072] Among them, f k is the frequency value of the kth spectrum line. In this embodiment, 13 features of each of the three types of data are obtained through the third step of the wear form prediction method.

[0073] The fourth step is to establish three prediction models for the three types of data. In this step, the time domain / frequency domain features of the three experimental data and the wear form label encoding are first used to form a data set and generate multiple training subsets. Then, a decision tree is constructed for each training subset and repeated to obtain multiple decision trees. Each decision tree is recursively split from the root node until the node meets a preset stop condition and is marked as a leaf node. Finally, the voting method is used to select the category with the majority prediction as the final result, and a preliminary prediction function is obtained, that is, three preliminary prediction models are obtained.

[0074] In this embodiment, Bootstrap sampling (random sampling with replacement) is used to generate H training subsets from the data set. Each training subset accounts for 2 / 3 of the original data, and the remaining 1 / 3 of the original data is used as out-of-bag data (OOB). For each training subset, a decision tree is constructed. The construction method of the decision tree includes the following steps: first, randomly select m features (m=3) when splitting each node of the tree, and then select the best feature to split the node according to the Gini index as the splitting criterion. where p g is the proportion of samples of the gth class in the node. Finally, recursive splitting is performed, and feature selection and optimal split point selection are repeated until the number of node samples is less than a set threshold or the maximum depth is reached.

[0075] The three experimental data sets are:

[0076]

[0077] Among them, D μ 、D Te 、D No The data sets represent friction coefficient, temperature, and noise respectively; Represents a 13-dimensional feature vector composed of time domain / frequency domain features; Yi∈{1, 2, 3, 4} represents the category labels of four wear forms.

[0078] In this embodiment, the above process is repeated to generate H decision trees. Each decision tree begins at the root node and repeats the recursive splitting process until a stopping condition is met. At internal nodes, samples are split based on the selected optimal features and thresholds. When a node meets the stopping condition, it is marked as a leaf node, and the prediction result is determined based on the majority class.

[0079] The fifth step is to determine whether the three preliminary prediction functions have a mode. If so, the mode of the three preliminary prediction functions is determined as the final prediction function. Otherwise, a weighted function is established to determine the final prediction function and generate the final prediction model. If the result has a mode, the final prediction function is: The results are different, and a weighting function is established to determine the final prediction function.

[0080] In this embodiment, the method for determining the final prediction function by using the weighting function includes the following steps:

[0081] (1) Order Calculate the deviation: in, They represent three preliminary prediction functions respectively. The preliminary prediction function of the friction coefficient is:

[0082]

[0083] Among them, T b represents the b-th decision tree, B represents the total number of decision trees, and mode(·) represents the mode.

[0084] (2) Calculate the weighting factor:

[0085] (3) Let the weighting function be The final prediction function is In the expression, round(·) means rounding to the nearest integer.

[0086] The sixth step is to predict the wear form based on the final prediction model.

[0087] In summary, the wear form prediction method of this embodiment has the following beneficial effects:

[0088] This wear form prediction method first obtains three types of test data, namely friction coefficient, temperature, and noise, through experiments. It then extracts features from each set of test data using time domain feature extraction methods and frequency domain feature extraction methods, and corresponds them to wear forms. A data set is then formed based on the data features and their corresponding wear forms. A decision tree is constructed and trained on the data to obtain three preliminary prediction functions. A final prediction function is then obtained based on whether the three preliminary prediction functions have a mode, and a corresponding final prediction model is obtained. Finally, the wear form is predicted using the final prediction model. This solves the technical problem that existing friction and wear tests cannot predict wear forms and require the use of equipment such as scanning electron microscopes to determine wear forms. The method can collect test data in real time, perform data analysis in real time while conducting friction and wear tests, and predict all wear forms of the material throughout the entire operating process. This can reduce the use of test equipment, save time and cost, and provide a reference for product design, manufacturing, and engineering applications.

[0089] Example 2

[0090] See also Figure 2This embodiment provides a wear form prediction method, which adds some steps to the first embodiment. This embodiment also establishes a correction model, which predicts in different time periods: a time period ΔT is set, and the correction model prediction result is checked to see if it is the same as the prediction result of the previous time period; if the prediction results of the two models are different in the time period hΔT to (h+1)ΔT, the correction model updates the prediction data of the final prediction model, and the time hΔT when the prediction result of the correction model begins to change is used as the starting time of the updated final prediction model, and the final prediction model is recalculated; when outputting, the prediction result of the final prediction model before the time hΔT is output as one of the wear form results, and the prediction result after the time hΔT is output as another wear form result, thereby predicting multiple wear forms in different time periods during the entire working process.

[0091] When predicting wear patterns using actual prediction methods, we consider that materials may exhibit different wear patterns at different times. Therefore, we add a correction model. The correction model uses the same calculation method as the prediction model, but the correction model predicts wear patterns in different time periods. During the prediction phase, sensors are used to collect three types of test data in real time for a specific material and operating condition. The correction model is then used to predict wear patterns in different time periods. The correction model's prediction results are then checked to see if they change. If so, the final prediction model updates the data and outputs the results for the different time periods. Otherwise, the final prediction model continues to calculate and output the results, ultimately yielding the wear pattern results for the entire operating condition.

[0092] Example 3

[0093] See also Figure 3 as well as Figure 4 , this embodiment provides a wear testing machine, which is applied to the wear form prediction method in embodiment 1 or 2. Among them, the wear testing machine includes a frame 1, a power mechanism 2, a tooling 3, a grinding mechanism 4, a friction sensor, a temperature sensor 5 and a noise sensor 6. The power mechanism 2 is installed on the frame 1, the tooling 3 is installed on the frame 1, and the grinding mechanism 4 is fixed on the frame 1. The power mechanism 2 drives the tooling 3 to perform reciprocating motion on the frame 1, and the tooling 3 drives the test block 7 to contact the grinding mechanism 4, causing the test block 7 to wear. The friction sensor is installed on the frame 1 and is used to detect the friction force generated by the wear of the test block 7. The wear testing machine calculates the friction coefficient based on the friction force, and the temperature sensor 5 is installed on the frame 1 and is used to detect the temperature generated by the wear of the test block 7. The noise sensor 6 is installed on the frame 1 and is used to detect the noise generated by the wear of the test block 7.

[0094] The wear testing machine has a fixed grinding mechanism 4, a test sample 7 that reciprocates with the tooling 3, and a temperature sensor 5 and a noise sensor 6 that are aligned with the contact portion between the grinding mechanism 4 and the test sample 7. The friction coefficient, temperature, and noise data are collected in real time. Finally, the wear form that will occur to a certain material under a certain working condition is predicted through a wear form prediction method. The wear form can be predicted while the test is in progress, thereby improving prediction efficiency and timeliness.

[0095] Example 4

[0096] This embodiment provides a computer terminal, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, the steps of the wear form prediction method of embodiment 1 are implemented.

[0097] The method of Example 1 can be implemented in the form of software, such as a standalone program installed on a computer terminal, which can be a computer, a smartphone, a control system, or other IoT device. The method of Example 1 can also be implemented as an embedded program installed on a computer terminal, such as a single-chip microcomputer.

[0098] Example 5

[0099] This embodiment provides a computer-readable storage medium having a computer program stored thereon. When the program is executed by a processor, the steps of the wear form prediction method of embodiment 1 are implemented.

[0100] When the method of Example 1 is applied, it can be applied in the form of software, such as a program designed as a computer-readable storage medium that can run independently. The computer-readable storage medium can be a USB flash drive designed as a USB shield, and the USB flash drive is designed to start the program of the entire method through external triggering.

[0101] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A wear form prediction method, characterized in that: It includes the following steps: Wear experiments were conducted to obtain three types of test data: friction coefficient, temperature, and noise; Label and encode various wear forms; First, the time domain features of the three test data are extracted respectively, then the time domain features are converted into frequency domain data, and finally the frequency domain features of the frequency domain data are extracted; First, the time domain / frequency domain features of the three experimental data and the wear form labels are encoded to form a data set and generate multiple training subsets. Then, a decision tree is constructed for each training subset and the process is repeated to obtain multiple decision trees. Each decision tree is recursively split from the root node until the node meets a preset stopping condition and is marked as a leaf node. Finally, the category with the majority of predictions is selected as the final result, and a preliminary prediction function is obtained. Determine whether the three preliminary prediction functions have a mode. If so, determine the mode of the three preliminary prediction functions as the final prediction function. Otherwise, establish a weighted function to determine the final prediction function and generate the final prediction model. Wear form prediction is performed based on the final prediction model.

2. The wear form prediction method according to claim 1, characterized in that: The time domain features include peak value f X,1 , mean f X,2 , root mean square error f X,3 , standard deviation f X,4 , kurtosis f X,5 , skewness f X,6 , margin index f X,7 , Waveform Index f X ,8.,Pulse index f X,9 ; The frequency domain features include the frequency mean F S,1 , frequency center of gravity F S,2 , frequency root mean square F S,3 , frequency variance F S,4 ; Wherein, X∈{μ, Te, No}, μ is the friction coefficient, Te is the temperature, and No is the noise.

3. The wear form prediction method according to claim 2, characterized in that: The calculation formula of the time domain feature is: f X，1 =max(|X(t)|) Wherein, X(t) represents the time domain data of the test data, t represents time, T represents the end time point, and n represents the number of valid samples.

4. The wear form prediction method according to claim 3, wherein: The frequency domain data is S(k), and the calculation formula is: The calculation formula of the frequency domain feature is: Among them, f k is the frequency value of the kth spectral line.

5. The wear form prediction method according to claim 4, characterized in that: The wear form is one of adhesive wear, abrasive wear, fatigue wear and oxidative wear, and the four wear forms are coded as 1, 2, 3 and 4; Bootstrap sampling is used to generate H training subsets from the data set, each training subset accounts for 2 / 3 of the original data, and the remaining 1 / 3 of the original data is used as out-of-bag data; The decision tree construction method includes the following steps: first, randomly selecting m features when splitting each node of the tree, then selecting the best feature for node splitting based on the Gini index as a splitting criterion, and finally performing recursive splitting, repeating feature selection and optimal splitting point selection until the number of node samples is less than a set threshold or reaches a maximum depth.

6. The wear form prediction method according to claim 5, characterized in that: The three experimental data sets are: Among them, D μ 、D Te 、D No data sets representing the friction coefficient, the temperature, and the noise respectively; represents a 13-dimensional feature vector composed of the time domain / frequency domain features; Yi∈{1, 2, 3, 4} represents the category labels of four wear forms.

7. The wear form prediction method according to claim 6, characterized in that: The method for determining the final prediction function by the weighting function comprises the following steps: make Calculate the deviation: in, Represent three preliminary prediction functions respectively; Calculate the weighting factors: Let the weighting function be Then the final prediction function is Here, round(·) indicates rounding to the nearest integer.

8. The wear form prediction method according to claim 7, characterized in that: The preliminary prediction function of the friction coefficient is: Among them, T b represents the b-th decision tree, B represents the total number of decision trees, and mode(·) represents the mode.

9. The wear form prediction method according to claim 8, characterized in that: A correction model is also established, which predicts in different time periods: a time period period ΔT is set, and the prediction result of the correction model is checked to see if it is the same as the prediction result of the previous time period; if the prediction results of the two models are different in the time period hΔT~(h+1)ΔT, the correction model is used to update the prediction data of the final prediction model, and the moment hΔT when the prediction result of the correction model begins to change is used as the starting moment of the updated final prediction model, and the final prediction model is recalculated; when outputting, the prediction result of the final prediction model before the moment hΔT is output as one of the wear form results, and the prediction result after the moment hΔT is output as another wear form result, and multiple wear forms in different time periods during the entire working process are predicted.

10. A wear testing machine, characterized in that: It is applied to the wear form prediction method according to any one of claims 1 to 9, wherein the wear testing machine comprises: frame; a power mechanism mounted on the frame; a tooling mounted on the frame; a grinding mechanism fixed to the frame; the power mechanism drives the tooling to perform reciprocating motion on the frame, and the tooling drives the test sample to contact the grinding mechanism, causing the test sample to wear; a friction sensor mounted on the frame and configured to detect friction generated by wear of the test specimen; the wear testing machine calculating a friction coefficient based on the friction; a temperature sensor mounted on the frame and used to detect the temperature generated by wear of the test sample; A noise sensor is mounted on the frame and is used to detect noise generated by wear of the test sample.

Images