A method, system, device and storage medium for evaluating fatigue life of hanging bolts

Abstract

The present invention discloses a method, system, device, and storage medium for evaluating the fatigue life of a hanger bolt. S1: Establishing a hanger bolt fatigue life model; S2: Constructing a cycle life dataset of the hanger bolt under different reasonable variable combinations based on the fatigue life model; S3: Constructing three types of fused characteristic indicators: average mass, stress ratio, and dynamic load; S4: Constructing a multilayer perceptron network model, using the three types of fused characteristic indicators as inputs to the multilayer perceptron network model, and taking the logarithm of fatigue life as label data for the multilayer perceptron network model; S5: Using samples from the filtered and selected cycle life dataset, training and testing the multilayer perceptron network model to evaluate the fatigue life of the hanger bolt. The method has high solution accuracy and strong applicability, and can accurately calculate the cycle life of the bolt under different variables.

Description

Technical Field

[0001] The present invention belongs to the field of life prediction of rotating parts, and relates to a method, system, equipment and storage medium for evaluating the fatigue life of a hanging bolt. Background Art

[0002] As a key component in the operation of high-speed trains, undercarriage suspension equipment is rigidly connected to the train body using suspension bolts. However, the service life of suspension bolts is an urgent issue that needs to be addressed.

[0003] Due to track irregularities, high-speed trains generate irregular vibrations during operation, causing high-cycle fatigue in the suspension bolts. If suspension bolts are not replaced promptly, they can fracture due to fatigue, causing the undercarriage equipment to fall off, compromising driving safety and even endangering passenger life and property. Frequent suspension bolt replacement requires constant disassembly and assembly of undercarriage equipment, resulting in high maintenance costs. Therefore, accurately assessing the service life of suspension bolts for different undercarriage equipment, while ensuring their safety and feasibility, is crucial.

[0004] Against the backdrop of continued optimization of EMU component repair schedules, there's a potential for extended maintenance of high-speed train suspension equipment. Currently, only typical off-board equipment has been shown to have a lifespan extension, but feasibility analysis of extending the service life of suspension bolts is lacking. However, current research primarily focuses on fatigue performance studies of single bolts, which lacks general applicability. It's difficult to quickly determine the fatigue life of the same type of bolts under different sizes, strengths, and loads. Summary of the Invention

[0005] The purpose of the present invention is to overcome the shortcomings of the above-mentioned prior art and provide a method, system, equipment and storage medium for evaluating the fatigue life of a hanging bolt, which has high solution accuracy and strong applicability and can accurately solve the cycle life of the bolt under different variables.

[0006] In order to achieve the above object, the present invention adopts the following technical solutions:

[0007] A method for evaluating the fatigue life of a hanging bolt includes the following steps:

[0008] S1, establish the fatigue life model of the hanging bolt;

[0009] S2, constructs a cycle life dataset of hanging bolts under different reasonable variable combinations based on the fatigue life model;

[0010] S3, construct three types of integrated characteristic indicators: average mass, stress ratio, and dynamic load;

[0011] S4, constructing a multi-layer perceptron network model, using the three types of fusion characteristic indicators as inputs of the multi-layer perceptron network model, and taking the logarithm of fatigue life as label data of the multi-layer perceptron network model;

[0012] S5, uses samples from the filtered and selected cycle life dataset to train and test the multi-layer perceptron network model to evaluate the fatigue life of the hanging bolts.

[0013] Preferably, in S1, the hanger bolt accelerated fatigue life test data is used to establish a hanger bolt fatigue life model.

[0014] Preferably, in S1, the fatigue life model of the hanging bolt is:

[0015] NS m =A

[0016] Where S is the cyclic stress, N is the number of cycles, and m and A are parameters to be solved. After equation transformation, the functional relationship of lgN-lgS is:

[0017]

[0018] According to different equivalent cyclic stresses [S1, S2, S3, ...., S n ] and its corresponding fatigue life [N1, N2, N3, ...., N n ], the linear regression algorithm is used to fit the linear model of lgN-lgS; and then the two parameters m and A of the fatigue life model are solved.

[0019] Preferably, in S2, the variables include bolt nominal diameter, bolt strength grade, bolt pre-tightening torque, hanging equipment mass, bolt group quantity and load amplitude.

[0020] Preferably, in S3, the average mass α, stress ratio β, and dynamic load γ are defined as:

[0021]

[0022]

[0023] γ=Q

[0024] Where M is the mass of the hanging equipment, n is the number of bolt groups, T is the tightening torque, D is the nominal diameter of the bolt, K is the tightening torque coefficient, σ is the tensile strength corresponding to the bolt strength grade, and Q is the bolt load amplitude.

[0025] Preferably, in S5, the specific process of screening and selecting samples in the cycle life data set is: retaining samples with a stress ratio index less than 0.6 as the screening principle, then randomly selecting multiple groups of samples, and performing standardization processing on three types of feature labels.

[0026] Preferably, in S5, 70% of the samples are used for training and 30% of the samples are used for prediction, and the network structure of the multi-layer perceptron network model is two linear layers.

[0027] A suspension bolt fatigue life assessment system, comprising:

[0028] A hanging bolt fatigue life model establishment module is used to establish a hanging bolt fatigue life model;

[0029] The cycle life dataset construction module is used to construct the cycle life dataset of the hanging bolt under different reasonable variable combinations based on the fatigue life model;

[0030] Fusion characteristic index construction module, used to construct three types of fusion characteristic indicators: average mass, stress ratio, and dynamic load;

[0031] A multi-layer perceptron network model construction module is used to construct a multi-layer perceptron network model, taking three types of fusion characteristic indicators as inputs of the multi-layer perceptron network model and taking the logarithm of fatigue life as label data of the multi-layer perceptron network model;

[0032] The fatigue life assessment module is used to train and test the multi-layer perceptron network model by screening and selecting samples from the cycle life dataset to evaluate the fatigue life of the hanging bolts.

[0033] A computer device includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps of the method for evaluating fatigue life of a hanger bolt are implemented.

[0034] A computer-readable storage medium stores a computer program, wherein the computer program, when executed by a processor, implements the steps of the method for evaluating fatigue life of a hanger bolt.

[0035] Compared with the prior art, the present invention has the following beneficial effects:

[0036] The present invention constructs a cycle life dataset by combining different variables, and then inputs it into a multi-layer perceptron network model through three types of fusion characteristic indicators. After screening and selecting samples from the cycle life dataset, the multi-layer perceptron network model is trained and tested to evaluate the fatigue life of the hanging bolts. The solution has high accuracy and strong applicability, and can accurately solve the cycle life of the bolts under different variables. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1 This is a flow chart of the method for quickly evaluating the fatigue life of hanging bolts described in an example of the present invention.

[0038] Figure 2 This is the linear fitting result diagram of the M20 hanging bolt lgN-lgS in the example of the present invention.

[0039] Figure 3 This is the S-IgN curve model diagram of the M20 hanging bolt in the example of the present invention.

[0040] Figure 4 This is a numerical graph of the multi-layer perceptron model training loss in an example of the present invention.

[0041] Figure 5 This is a numerical graph of the test loss of the multi-layer perceptron model in the example of the present invention. DETAILED DESCRIPTION

[0042] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments; based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0043] It should be noted that the words "front", "rear", "left", "right", "up" and "down" used in the following description refer to directions in the accompanying drawings, and the words "inside" and "outside" refer to directions toward or away from the geometric center of a specific component, respectively.

[0044] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the art to which this invention pertains. The terms used herein in the specification of the present invention are for the purpose of describing specific embodiments only and are not intended to limit the present invention. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.

[0045] like Figure 1 As shown in the figure, the fatigue life assessment method of the hanging bolts according to the present invention is firstly established by using the accelerated fatigue life test data of the hanging bolts; then, for six types of variables, namely bolt nominal diameter, bolt strength grade, bolt pre-tightening torque, hanging equipment mass, number of bolt groups, and load amplitude, the cycle life data set of the hanging bolts under different reasonable variable combinations is calculated according to the fatigue life model; then, three types of fusion characteristic indicators, namely average mass, stress ratio, and dynamic load, are constructed and used as the input of the multi-layer perceptron model, and the logarithm of fatigue life is taken as the label data of the model; finally, the cycle life data set is used to train and test the multi-layer perceptron network model to quickly evaluate the fatigue life of the hanging bolts with good prediction accuracy.

[0046] The specific steps include:

[0047] 1) Process the accelerated fatigue life test data of the hanging bolts. The tightening torque of the hanging bolts is M t , n different dynamic loads {F d1 , F d2 , F d3 ,....,F dn} level of fatigue life {N 11 , N 12 ,...,N 1n ,....,N 51 , N 52 ,....,N 5n Each load level has five test data. For the same dynamic axial load, remove the data points with large relative deviation, average the remaining data points, and obtain the bolt tightening torque M at this time. t Under the F dk Number of cycles under dynamic load:

[0048]

[0049] Where N kn is the kth dynamic axial force F dk Valid test data under this condition, n is the number of valid test data. Then obtain the bolt tightening torque M t Under the equally spaced alternating axial force {F d1 , F d2 , F d3 ,...,F dn}The corresponding number of cycles {N1, N2, N3, ...N n}.

[0050] 2) Solve the cyclic stress of the bolt under different dynamic loads. Calculate the bolt preload according to the preload-preload torque formula

[0051]

[0052] Where P is the preload force, T t is the tightening torque, d is the nominal diameter of the bolt, and K is the tightening torque coefficient. d Tensile cyclic stress, according to the bolt preload P0, alternating axial force F d , calculate the cyclic stress amplitude S a and mean stress S m for:

[0053]

[0054]

[0055] According to the Goodman mean stress correction equation, the equivalent cyclic stress S can be solved as:

[0056]

[0057] Where S b is the tensile stress corresponding to the strength grade of the bolt specimen.

[0058] 3) Establish a fatigue life model for hanging bolts. The fatigue life SN model for hanging bolts is:

[0059] NS m =A

[0060] Where S is the cyclic stress, N is the number of cycles, and m and A are parameters to be solved. After equation transformation, the functional relationship of lgN-lgS is:

[0061]

[0062] According to different equivalent cyclic stresses [S1, S2, S3, ..., S n ] and its corresponding fatigue life [N1, N2, N3, ..., N n ] and used the linear regression algorithm to fit the linear model of lgN-lgS.

[0063] Convert the functional relationship of lgN-lgS to:

[0064] y=w*x+b

[0065] In the formula, y=lgS, x=lgN,

[0066] For the number of cycles N, the cyclic stress S is directly introduced into the linear model y = w*x + b, and the number of cycles and the cyclic stress are both taken as logarithms:

[0067] y=lgS=[lgS1, lgS2, lgS3,..., lgS n ]

[0068] x=lgN=[lgN1, lgN2, lgN3,..., lgN n ]

[0069] Use the LinearRegression model to fit the data y and x and solve for w and b. The specific solution process is as follows:

[0070] Assume w = w′, b = b′. Substitute all samples to get the predicted label data:

[0071] y′1=w′x′1+b′

[0072] y′2=w′x′2+b′

[0073] …

[0074] y′ n =w′x′ n +b′

[0075] At this time, y′1, y′2, …, y′ n are all label data obtained by assuming w′ and b′, ​​then the real label data y1, y2, ..., y n There is a discrepancy between the labeled data and the hypothesis derived ones.

[0076] Solve this equation using the least squares method, minimizing the mean squared error:

[0077]

[0078] Expand the quadratic terms in the brackets to obtain the following equation:

[0079]

[0080] Then, the J(w, b) function is partial derivative with respect to the parameter b:

[0081]

[0082] make Then we get the following equation:

[0083]

[0084] Take the partial derivative of J(w,b) with respect to w:

[0085]

[0086] make and will Substituting into the formula, we get

[0087]

[0088] After the above derivation, w0 and b0 can be solved. Thus, the two unknown parameters m0 and A0 in the bolt fatigue life model can be solved, namely:

[0089]

[0090]

[0091] 4) For the six variables of hanging bolt nominal diameter, strength grade, pre-tightening torque, hanging equipment mass, number of bolt groups, and load amplitude, reasonable parameter values ​​are selected and combined to construct a cycle life data set.

[0092] 5) The dynamic load Q in the actual vehicle load spectrum is the vibration acceleration, which is converted into the alternating axial force as follows:

[0093]

[0094] Where M is the mass of the hanging equipment, n is the number of bolt groups, and Q is the dynamic load amplitude. Repeat step (2) to calculate the cyclic stress of the bolts under different variable combinations. Substitute the cyclic stress corresponding to each sample into the hanging bolt fatigue life model to solve the fatigue life corresponding to each sample.

[0095] 6) For each sample, three types of integration indicators are constructed: average mass, stress ratio, and dynamic load. The average mass α, stress ratio β, and dynamic load γ are defined as follows:

[0096]

[0097]

[0098] γ=Q

[0099] Where M is the mass of the hanging equipment, n is the number of bolt groups, T is the tightening torque, D is the nominal diameter of the bolt, K is the tightening torque coefficient, σ is the tensile strength corresponding to the bolt strength grade, and Q is the bolt load amplitude. The fatigue life of the variable combination in the data set is taken as the neural network label data, that is:

[0100] l=lgN

[0101] 7) The samples were retained based on the stress ratio index less than 0.6 as the screening principle. Then 10,000 groups of samples were randomly selected as the cycle life data set and standardized for three types of feature labels, namely:

[0102]

[0103] In the formula, mean is the mean value of the eigenvalue in the data set, std is the standard deviation of the eigenvalue in the data set, X is the original data, and X1 is the data after standardization.

[0104] 8) Construct a multi-layer perceptron network model, use the three types of fusion indicators of 10,000 samples as input data, l as label data, 70% of the data for training, and 30% of the data for prediction. The network structure consists of two layers of linear fully connected network layers, 3*5 and 5*1 respectively. The final network accuracy performance is good.

[0105] The following is a practical example to further illustrate:

[0106] First, the test data are shown in Table 1.

[0107] Table 1 M20 hanging bolt accelerated fatigue life test data

[0108]

[0109]

[0110] Numerous factors can influence the results of accelerated fatigue testing of bolts. Therefore, data that deviated significantly from the average were deleted. These data were 24,028 from Test 1, 27,081 from Test 2, 110,269 and 72,496 from Test 3, and 35,354 from Test 5. These data were not included in the subsequent data analysis. The remaining data were averaged according to the dynamic load, and the processed results are shown in Table 2.

[0111] Table 2 Results of M20 hanging bolt accelerated fatigue life test data processing

[0112]

[0113] The tensile stress corresponding to different bolt strength grades is shown in Table 3.

[0114] Table 3 Correspondence between strength grade and tensile stress of hanging bolts

[0115]

[0116] The cyclic stress of the bolt under different dynamic loads can be solved according to the Goodman stress correction equation:

[0117]

[0118] Where, F d is the alternating axial force of the bolt, P is the bolt tightening torque 170Nm, D b is the tensile stress corresponding to the bolt strength grade, and d is the nominal diameter of the bolt.

[0119] The calculation results of cyclic stress under different alternating axial forces are shown in Table 4.

[0120] Table 4 Cyclic stress data of M20 hanging bolts under different dynamic loads

[0121]

[0122] In order to accurately obtain the unknown parameters m and A of the bolt fatigue life model, the accelerated fatigue life test data are logarithmically operated, and the results are shown in Table 5.

[0123] Table 5 M20 hanging bolt accelerated fatigue test lgN and lgS data

[0124]

[0125] At this time, the functional relationship of lgN-lgS is:

[0126]

[0127] Use the linear regression algorithm to fit this linear model lgS = w*lgN + b, such as Figure 2 As shown, the results are as follows:

[0128]

[0129]

[0130] Finally, the unknown parameters of the bolt fatigue life model are solved as follows:

[0131] m=1.6250

[0132] A=2.1768*10 18

[0133] Therefore, the mathematical model of bolt fatigue life (lgN-S) is as follows Figure 3 As shown, the result is:

[0134] N*S 1.6250 =2.1768*10 18

[0135] When the bolt selection (bolt thread, pitch and other parameters) is constant, the variables affecting the bolt service life include: bolt diameter, bolt material, bolt strength grade, hanging equipment quality, number of bolt groups, tightening torque, load amplitude, load frequency, hanging equipment vibration characteristics, etc.

[0136] The bolt material determines the bolt strength. Currently, there are 8.8-grade strength bolts, 6.8-grade strength bolts, and 4.8-grade strength bolts on the market. They correspond to different bolt materials, but bolts of the same strength are made of the same material. Therefore, the bolt strength factor can be considered in the modeling process, regardless of the bolt material factor.

[0137] The bolt load frequency of the actual vehicle suspension equipment is often lower than the first-order modal resonance frequency, so the bolt load frequency factor can be ignored.

[0138] The vibration characteristics of suspended equipment are reflected in its load spectrum. The load spectrum of vibrating equipment exhibits a high frequency of vibration acceleration, while the load spectrum of non-vibrating equipment exhibits a low frequency of vibration acceleration. However, if the same bolts are used for both vibrating and non-vibrating equipment, their fatigue life is the same under a single load. This chapter focuses on single-load analysis of different suspended equipment, therefore, the vibration characteristics of the equipment do not need to be considered.

[0139] The factors that affect the service life of bolts are as follows:

[0140] a. Nominal diameter of bolt;

[0141] b. Bolt strength grade;

[0142] c. Tightening torque;

[0143] d. Quality of hanging equipment;

[0144] e. Number of bolt groups for hanging equipment;

[0145] f. Load amplitude;

[0146] The reference values ​​of factors affecting the service life of bolts are shown in Table 6.

[0147] Table 6 Numerical parameter table of six types of parameters of hanging bolts

[0148]

[0149]

[0150] Various variables at different levels were cross-combined, resulting in a total of 6*3*6*6*5*8 = 25,920 samples. Based on the mathematical model of bolt fatigue life, the fatigue life of the samples corresponding to different variable parameter combinations was calculated. Three integrated performance indicators were constructed: stress ratio, average mass, and dynamic load. To prevent the bolts from being directly broken under the combined action of preload and alternating axial force, a stress ratio less than 0.6 was used as a screening criterion. A total of 10,368 samples were retained that met the criteria. Next, 10,000 samples were randomly selected from the retained samples as the data set.

[0151] The three fusion indicators of each sample, stress ratio, average mass, and dynamic load, are used as neural network inputs, and the logarithm of the fatigue life of each sample is used as the neural network output. 70% of the data set is randomly divided into a training set, totaling 7,000 samples; the remaining 30% of the data is used as a test set, totaling 3,000 samples. For the input feature data, standardization is completed, namely:

[0152]

[0153] In the formula, mean is the mean value of the eigenvalue in the data set, std is the standard deviation of the eigenvalue in the data set, X is the original data, and X1 is the data after standardization.

[0154] Build a multi-layer perceptron neural network model. The network model structure is as follows:

[0155] 1) 3*5 linear fully connected layer;

[0156] 2) 5*1 linear fully connected layer;

[0157] The network training loss function is set to mean square error loss function (MSE), the optimizer is set to stochastic gradient descent optimizer (SGD), the learning rate is set to 0.02, the number of training iterations epoch is set to 50, and the final training set loss error is as follows Figure 4 As shown, the test set loss error is Figure 5 As shown in the figure, the mean square error is less than 0.1, and the model accuracy is good.

[0158] The following are device embodiments of the present invention, which can be used to perform the method embodiments of the present invention. For details not disclosed in the device embodiments, please refer to the method embodiments of the present invention.

[0159] In another embodiment of the present invention, a hanging bolt fatigue life assessment system is provided. The hanging bolt fatigue life assessment system can be used to implement the above-mentioned hanging bolt fatigue life assessment method. Specifically, the hanging bolt fatigue life assessment system includes a hanging bolt fatigue life model establishment module, a cycle life data set construction module, a fusion feature index construction module, a multi-layer perceptron network model construction module and a fatigue life assessment module.

[0160] Among them, the hanging bolt fatigue life model establishment module is used to establish the hanging bolt fatigue life model.

[0161] The cycle life dataset construction module is used to construct the cycle life dataset of the hanging bolt under different reasonable variable combinations based on the fatigue life model.

[0162] The fusion characteristic index construction module is used to construct three types of fusion characteristic indicators: average mass, stress ratio, and dynamic load.

[0163] The multi-layer perceptron network model construction module is used to construct the multi-layer perceptron network model, taking three types of fusion feature indicators as the input of the multi-layer perceptron network model, and taking the logarithm of fatigue life as the label data of the multi-layer perceptron network model.

[0164] The fatigue life assessment module is used to train and test the multi-layer perceptron network model by screening and selecting samples from the cycle life dataset to evaluate the fatigue life of the hanging bolts.

[0165] In another embodiment of the present invention, a terminal device is provided, the terminal device including a processor and a memory, the memory being used to store a computer program, the computer program including program instructions, and the processor being used to execute the program instructions stored in the computer storage medium. The processor may be a central processing unit (CPU), or may be other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGAs), or a processor that is ... GateArray, FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc., which are the computing core and control core of the terminal, which are suitable for implementing one or more instructions, specifically suitable for loading and executing one or more instructions to realize the corresponding method flow or corresponding function; the processor described in the embodiment of the present invention can be used for the operation of the fatigue life evaluation method of the hanging bolt, including: S1, establishing a fatigue life model of the hanging bolt; S2, constructing a cycle life data set of the hanging bolt under different reasonable variable combinations according to the fatigue life model; S3, constructing three types of fusion characteristic indicators of average mass, stress ratio, and dynamic load; S4, constructing a multi-layer perceptron network model, using the three types of fusion characteristic indicators as the input of the multi-layer perceptron network model, and taking the logarithm of fatigue life as the label data of the multi-layer perceptron network model; S5, using samples in the cycle life data set after screening and selection, training and testing the multi-layer perceptron network model, and evaluating the fatigue life of the hanging bolt.

[0166] In another embodiment, the present invention further provides a computer-readable storage medium (Memory), which is a memory device in a terminal device for storing programs and data. It is understandable that the computer-readable storage medium here can include both the built-in storage medium in the terminal device and, of course, the extended storage medium supported by the terminal device. The computer-readable storage medium provides a storage space, which stores the operating system of the terminal. In addition, one or more instructions suitable for being loaded and executed by the processor are also stored in the storage space. These instructions can be one or more computer programs (including program codes). It should be noted that the computer-readable storage medium here can be a high-speed RAM memory or a non-volatile memory, such as at least one disk memory.

[0167] One or more instructions stored in a computer-readable storage medium can be loaded and executed by a processor to implement the corresponding steps of the method for evaluating the fatigue life of a hanger bolt in the above-mentioned embodiment; one or more instructions in the computer-readable storage medium are loaded by the processor and execute the following steps: S1, establishing a fatigue life model for a hanger bolt; S2, constructing a cycle life data set of a hanger bolt under different reasonable variable combinations based on the fatigue life model; S3, constructing three types of fusion characteristic indicators of average mass, stress ratio, and dynamic load; S4, constructing a multi-layer perceptron network model, using the three types of fusion characteristic indicators as inputs of the multi-layer perceptron network model, and taking the logarithm of the fatigue life as label data of the multi-layer perceptron network model; S5, using samples in the cycle life data set after screening and selection, training and testing the multi-layer perceptron network model, and evaluating the fatigue life of the hanger bolt.

[0168] Those skilled in the art will appreciate that the embodiments of the present application can be provided as methods, systems, or computer program products. Therefore, the present application can adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application can adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.

[0169] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the steps in the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0170] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0171] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0172] It should be noted that, in this document, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "comprises," "comprising," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that includes a list of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, method, article, or apparatus.

[0173] It should be understood that the above description is for illustrative purposes only and is not intended to be limiting. Many embodiments and many applications beyond the examples provided will be apparent to those skilled in the art upon reading the above description. Therefore, the scope of the present teachings should not be determined with reference to the above description, but rather with reference to the preceding claims and the full scope of equivalents to which such claims are entitled. For the purpose of completeness, all articles and references, including disclosures of patent applications and publications, are incorporated herein by reference. The omission of any aspect of the subject matter disclosed herein from the preceding claims is not a disclaimer of such subject matter, nor should it be considered that the applicants did not consider such subject matter to be part of the disclosed inventive subject matter.

Claims

1. A method for evaluating the fatigue life of a hanging bolt, characterized in that: The following processes are included: S1, establish the fatigue life model of the hanging bolt; The fatigue life model of the hanging bolt is: Where S is the cyclic stress, N is the number of cycles, and m and A are parameters to be solved. After the equation is transformed, The functional relationship is: According to different equivalent cyclic stress And its corresponding fatigue life , using linear regression algorithm to fit The linear model of fatigue life model is solved. 、 ; S2, constructs a cycle life dataset of hanging bolts under different reasonable variable combinations based on the fatigue life model; S3, construct three types of integrated characteristic indicators: average mass, stress ratio, and dynamic load; Average quality , stress ratio , dynamic load is defined as: Where, For the quality of hanging equipment, is the number of bolt groups, T is the tightening torque, D is the nominal diameter of the bolt, is the tightening torque coefficient, is the tensile strength corresponding to the bolt strength grade, Q is the bolt load amplitude; S4, constructing a multi-layer perceptron network model, using the three types of fusion characteristic indicators as inputs of the multi-layer perceptron network model, and taking the logarithm of fatigue life as label data of the multi-layer perceptron network model; S5, uses samples from the filtered and selected cycle life dataset to train and test the multi-layer perceptron network model to evaluate the fatigue life of the hanging bolts.

2. The method for evaluating the fatigue life of a hanging bolt according to claim 1, characterized in that: In S1, the hanger bolt accelerated fatigue life test data is used to establish the hanger bolt fatigue life model.

3. The method for evaluating the fatigue life of a hanging bolt according to claim 1, wherein: In S2, the variables include bolt nominal diameter, bolt strength grade, bolt preload torque, hanging equipment mass, number of bolt groups, and load amplitude.

4. The method for evaluating the fatigue life of a hanging bolt according to claim 1, wherein: In S5, the specific process of screening and selecting samples from the cycle life dataset is as follows: samples are retained based on the stress ratio index less than 0.6 as the screening principle, and then multiple groups of samples are randomly selected and standardized for three types of feature labels.

5. The method for evaluating the fatigue life of a hanging bolt according to claim 1, wherein: In S5, 70% of the samples are used for training and 30% of the samples are used for prediction. The network structure of the multi-layer perceptron network model is two linear layers.

6. A system for evaluating the fatigue life of hanging bolts, characterized in that: include: A hanging bolt fatigue life model establishment module is used to establish a hanging bolt fatigue life model; The fatigue life model of the hanging bolt is: Where S is the cyclic stress, N is the number of cycles, and m and A are parameters to be solved. After the equation is transformed, The functional relationship is: According to different equivalent cyclic stress And its corresponding fatigue life , using linear regression algorithm to fit The linear model of fatigue life model is solved. 、 ; The cycle life dataset construction module is used to construct the cycle life dataset of the hanging bolt under different reasonable variable combinations based on the fatigue life model; Fusion characteristic index construction module, used to construct three types of fusion characteristic indicators: average mass, stress ratio, and dynamic load; Average quality , stress ratio , dynamic load is defined as: Where, For the quality of hanging equipment, is the number of bolt groups, T is the tightening torque, D is the nominal diameter of the bolt, is the tightening torque coefficient, is the tensile strength corresponding to the bolt strength grade, Q is the bolt load amplitude; A multi-layer perceptron network model construction module is used to construct a multi-layer perceptron network model, taking three types of fusion characteristic indicators as inputs of the multi-layer perceptron network model and taking the logarithm of fatigue life as label data of the multi-layer perceptron network model; The fatigue life assessment module is used to train and test the multi-layer perceptron network model by screening and selecting samples from the cycle life dataset to evaluate the fatigue life of the hanging bolts.

7. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the steps of the method for evaluating the fatigue life of a hanger bolt according to any one of claims 1 to 5 are implemented.

8. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the steps of the method for evaluating the fatigue life of a hanger bolt according to any one of claims 1 to 5 are implemented.

Images