Method and device for predicting fatigue life of swing arm

Abstract

The invention provides a method and a device for predicting the fatigue life of a swing arm. The method comprises the following steps: acquiring a time-history excitation load of the swing arm; inputting the time-history excitation load into a stress response prediction model to obtain a stress response output by the stress response prediction model; the stress response prediction model is obtained based on sample time-history excitation load and sample stress response training; and predicting the fatigue life of the swing arm based on the stress response. According to the swing arm fatigue life prediction method and device provided by the invention, the stress response can be quickly and accurately obtained based on the stress response prediction model, and then the fatigue life of the swing arm can be quickly and accurately predicted based on the stress response.

Description

Technical Field

[0001] The present invention relates to the technical field of fatigue analysis, and in particular to a method and device for predicting fatigue life of a swing arm. Background Art

[0002] Automobile swing arm fatigue analysis is the fatigue life evaluation and analysis process for the swing arm (also called control arm or suspension arm) in the automobile suspension system. In the automobile suspension system, the swing arm plays an important role in connecting the wheels and the body structure. It needs to withstand the dynamic load and stress from the wheel movement under various road conditions. Therefore, fatigue analysis of the swing arm is the key to ensure the safe and reliable operation of the automobile suspension system.

[0003] At present, fatigue analysis of automobile swing arms mainly relies on component experiments and CAE simulation calculations. Component experiments mainly occur in the structural verification stage, and in the digital design stage, they mainly rely on CAE calculations of finite element software. The fatigue performance analysis of the swing arm through CAE simulation calculation is mainly divided into two stages. The first stage includes three parts: meshing, finite element model establishment, and structural stress response analysis. The second stage mainly relies on the stress results of finite element analysis plus material fatigue performance settings, rain flow counting statistics, and fatigue damage statistics to complete the fatigue life analysis of the structure. However, this method is cumbersome, which greatly increases the cost of calculation time, affects the design efficiency, and has large deviations in the results of finite element analysis, which seriously affects the accuracy of fatigue life calculations. Summary of the invention

[0004] The present invention provides a method and device for predicting fatigue life of a swing arm, so as to solve the defects existing in the prior art.

[0005] The present invention provides a method for predicting fatigue life of a swing arm, comprising the following steps: Obtain the time history excitation load of the swing arm; Inputting the time history excitation load into a stress response prediction model to obtain a stress response output by the stress response prediction model; the stress response prediction model is obtained based on sample time history excitation load and sample stress response training; Based on the stress response, the fatigue life of the swing arm is predicted.

[0006] According to a method for predicting fatigue life of a swing arm provided by the present invention, the stress response prediction model is trained based on the following steps: Obtaining historical time-history excitation loads and historical stress responses corresponding to the historical time-history excitation loads; Based on a preset ratio, the historical time history excitation load is divided into the sample time history excitation load and the test time history excitation load, and the proportion of the sample time history excitation load is greater than the proportion of the test time history excitation load; Performing model training based on the sample time-history excitation load and the sample stress response corresponding to the sample time-history excitation load to obtain an initial model; Determining the accuracy of the initial model based on the test time history excitation load and the test stress response corresponding to the test time history excitation load; When the accuracy is greater than a threshold, the initial model is used as the stress response prediction model.

[0007] According to a method for predicting fatigue life of a swing arm provided by the present invention, the accuracy of the initial model is determined based on the test time history excitation load and the test stress response corresponding to the test time history excitation load, including: Inputting the test time history excitation load into the initial model to obtain a predicted stress response output by the initial model; The accuracy is determined based on the predicted stress response, and the tested stress response.

[0008] According to a method for predicting fatigue life of a swing arm provided by the present invention, the accuracy is determined based on the predicted stress response and the tested stress response, including: Determining a root mean square error and a correlation coefficient based on the predicted stress response and the tested stress response; The accuracy is determined based on the root mean square error or the correlation coefficient.

[0009] According to a method for predicting fatigue life of a swing arm provided by the present invention, predicting fatigue life of the swing arm based on the stress response comprises: Performing periodic statistics on the stress history corresponding to the stress response to determine the number of cycles and cycle amplitude of each cycle; Determine the fatigue damage of each cycle based on the number of cycles and cycle amplitude of each cycle; Based on the fatigue damage in each cycle, the fatigue life of the swing arm is predicted.

[0010] According to a method for predicting fatigue life of a swing arm provided by the present invention, the fatigue life of the swing arm is predicted based on fatigue damage in each cycle, comprising: The fatigue damage of all cycles is accumulated to obtain the cumulative fatigue damage; Based on the accumulated fatigue damage, the fatigue life of the swing arm is predicted.

[0011] The present invention also provides a swing arm fatigue life prediction device, comprising the following modules: An acquisition unit is used to acquire the time history excitation load of the swing arm; An input unit, used for inputting the time history excitation load into a stress response prediction model to obtain a stress response output by the stress response prediction model; the stress response prediction model is obtained based on sample time history excitation load and sample stress response training; A prediction unit is used to predict the fatigue life of the swing arm based on the stress response.

[0012] The present invention also provides an electronic 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 program, a method for predicting fatigue life of a swing arm as described above is implemented.

[0013] The present invention also provides a non-transitory computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the method for predicting the fatigue life of a swing arm as described in any one of the above is implemented.

[0014] The present invention also provides a computer program product, comprising a computer program, wherein when the computer program is executed by a processor, the method for predicting fatigue life of a swing arm as described above is implemented.

[0015] The method and device for predicting the fatigue life of a swing arm provided by the present invention can quickly and accurately obtain the stress response based on the stress response prediction model, and further can quickly and accurately predict the fatigue life of the swing arm based on the stress response. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0017] Figure 1 It is a flow chart of the swing arm fatigue life prediction method provided by the present invention.

[0018] Figure 2 It is a structural diagram of the LSTM network provided by the present invention.

[0019] Figure 3 It is a flow chart of another swing arm fatigue life prediction method provided by the present invention.

[0020] Figure 4 It is a structural schematic diagram of the swing arm fatigue life prediction device provided by the present invention.

[0021] Figure 5 It is a structural schematic diagram of the electronic device provided by the present invention. DETAILED DESCRIPTION

[0022] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0023] At present, automobile swing arm fatigue analysis mainly relies on component experiments and CAE simulation calculations. Component experiments mainly occur in the structural verification stage, and in the digital design stage, they mainly rely on CAE calculations of finite element software. CAE simulation calculation of swing arm fatigue performance analysis is mainly divided into two stages. The first stage includes three parts: mesh division, finite element model establishment, and structural stress response analysis. The second stage mainly relies on the stress results of finite element analysis plus material fatigue performance setting, rain flow counting statistics, and fatigue damage statistics to complete the fatigue life analysis of the structure. However, this method has the following defects: First, in the initial stage of design, the design scheme of the structure changes greatly, and timely finite element performance analysis and calculation feedback is needed to guide the direction of design optimization. The traditional fatigue calculation method is cumbersome, which greatly increases the time cost of calculation and seriously affects the efficiency of design.

[0024] Second, the current fatigue analysis methods based on finite element analysis are mainly divided into two types: time domain analysis method and quasi-static analysis method. The time domain fatigue analysis method is not widely used due to its long calculation time and high requirements for computing equipment; the quasi-static method has been widely used due to its high calculation efficiency. However, the quasi-static analysis method cannot consider the influence of the excitation vibration sequence during the calculation process, and ignores the resonance of the structure, resulting in a large deviation in the finite element analysis results, which seriously affects the accuracy of fatigue life calculation.

[0025] The present invention provides a method for predicting the fatigue life of a swing arm, which changes the traditional fatigue analysis method based on finite element calculation into a deep learning fatigue analysis calculation driven by a data plus mechanism, which can quickly and efficiently complete the fatigue calculation of the swing arm and greatly improve the speed of product development.

[0026] Figure 1 : is a schematic diagram of the process of predicting the fatigue life of the swing arm provided by the present invention, such as Figure 1 As shown, the method includes step 110 , step 120 and step 130 .

[0027] Step 110: Obtain the time-history excitation load of the swing arm.

[0028] Here, the time-history excitation load of the swing arm refers to the force or load that the swing arm is subjected to that varies with time during actual use or testing. The time-history excitation load includes, but is not limited to, the impact force on the swing arm due to uneven road surface, the longitudinal force on the swing arm when the vehicle accelerates or brakes, the lateral force on the swing arm when the vehicle turns, and the vibration load caused by engine vibration, road vibration, etc.

[0029] Optionally, when the car is driving on an actual road, sensors and data acquisition equipment can be installed to record various forces and loads on the swing arm to obtain actual time-history excitation load data.

[0030] Step 120: input the time history excitation load into the stress response prediction model to obtain the stress response output by the stress response prediction model; the stress response prediction model is obtained based on the sample time history excitation load and the sample stress response training.

[0031] Specifically, the time history excitation load obtained in step 110 is used as an input to a pre-established stress response prediction model, and the stress response prediction model performs analysis based on the time history excitation load to obtain a stress response. The stress response can be understood as the stress state of each point in the swing arm structure in each time step or a specified time period, and the stress state includes but is not limited to tension, pressure, shear force, etc.

[0032] In addition, the stress response prediction model is trained based on sample time-history excitation loads and sample stress responses. The sample time-history excitation loads can be understood as real or simulated load data, and the sample stress response is the actual stress data corresponding to each moment or time period, that is, the real label corresponding to the sample time-history excitation load.

[0033] In the training phase, a large amount of known sample time-history excitation load data and corresponding sample stress response data are input into the model for learning, so that the trained stress response prediction model can accurately predict stress response. Among them, the stress response prediction model can be constructed based on the Long Short-Term Memory DeepSurrogate Model (LSTM).

[0034] Step 130: predict the fatigue life of the swing arm based on the stress response.

[0035] Specifically, the fatigue life of the swing arm is predicted using the stress response obtained in step 120. The fatigue life of the swing arm refers to the time or mileage that the swing arm can be used normally without fatigue damage under repeated driving loads and vibrations, which measures the durability of the swing arm in long-term use and ensures the safety and stability of the vehicle under normal operating conditions.

[0036] Optionally, fatigue life prediction methods can be used to predict the fatigue life of the swing arm, such as the SN curve (stress-cycle number curve) based on the material, or the Miner linear accumulation criterion, etc. The fatigue life of the swing arm can also be predicted by the rain flow counting method and damage accumulation calculation.

[0037] The swing arm fatigue life prediction method provided by the embodiment of the present invention can quickly and accurately obtain the stress response based on the stress response prediction model, and then can quickly and accurately predict the fatigue life of the swing arm based on the stress response. Experiments have shown that the traditional method takes 4 hours to complete the swing arm fatigue life prediction, while the method provided by the embodiment of the present invention only takes 2 minutes to complete the swing arm fatigue life prediction, greatly improving the calculation efficiency.

[0038] Based on the above embodiment, the stress response prediction model is trained based on the following steps: Obtain historical time-history excitation loads and historical stress responses corresponding to historical time-history excitation loads; Based on the preset ratio, the historical time history excitation load is divided into the sample time history excitation load and the test time history excitation load, and the proportion of the sample time history excitation load is greater than that of the test time history excitation load; Model training is performed based on the sample time history excitation load and the sample stress response corresponding to the sample time history excitation load to obtain an initial model; Determine the accuracy of the initial model based on the test time history excitation load and the test stress response corresponding to the test time history excitation load; When the accuracy is greater than the threshold, the initial model is used as the stress response prediction model.

[0039] Specifically, the historical time-history excitation load refers to the time-varying force or load historically applied to the swing arm structure. The historical stress response corresponding to the historical time-history excitation load refers to the stress state of each point in the swing arm structure under the action of the historical time-history excitation load.

[0040] After obtaining the historical time history excitation load, the historical time history excitation load is divided into a sample time history excitation load and a test time history excitation load based on a preset ratio. The sample time history excitation load can be understood as a training set, and the test time history excitation load can be understood as a test set. Among them, the proportion of the sample time history excitation load is greater than that of the test time history excitation load.

[0041] Optionally, the preset ratio may be 8:2, that is, 80% of the data sets in the historical time history excitation load are used as the training set (ie, the sample time history excitation load), and 20% of the data sets are used as the test set (ie, the test time history excitation load).

[0042] After obtaining the sample time history excitation load, model training is performed based on the sample time history excitation load and the sample stress response corresponding to the sample time history excitation load to obtain an initial model. The initial model can be constructed based on an LSTM network.

[0043] Figure 2 It is a structural diagram of the LSTM network provided by the present invention, such as Figure 2 As shown in the figure, the LSTM network consists of an input layer (Input layer), two LSTM layers (LSTM1 and LSTM2), a fully connected layer (FC Layer) and an output layer (Output Layer).

[0044] After obtaining the initial model, the test time history excitation load and its corresponding test stress response are used to verify the performance of the initial model, that is, the accuracy of the initial model is evaluated based on the test time history excitation load and its corresponding test stress response. The higher the accuracy, the better the performance of the initial model.

[0045] If the accuracy is greater than the threshold, it indicates that the performance of the initial model is good, and the initial model can be used as the stress response prediction model. If the accuracy is less than or equal to the threshold, it indicates that the performance of the initial model is poor. At this time, the training set can be incrementally obtained and the parameters of the initial model can be updated to ensure that the accuracy of the initial model obtained by the final training is greater than the threshold.

[0046] Based on any of the above embodiments, determining the accuracy of the initial model based on the test time history excitation load and the test stress response corresponding to the test time history excitation load includes: Input the test time history excitation load into the initial model to obtain the predicted stress response output by the initial model; Based on the predicted stress response, as well as the tested stress response, the accuracy is determined.

[0047] Specifically, after the initial model is obtained, the performance of the initial model needs to be verified. In the embodiment of the present invention, the performance of the initial model is verified based on the accuracy of the initial model. The higher the accuracy, the better the performance.

[0048] Optionally, the test time history excitation load is input into the initial model for prediction, and the predicted stress response output by the initial model is obtained. The predicted stress response (i.e., the predicted value) is compared with the test stress response (i.e., the actual value) to determine the accuracy of the initial model. For example, the accuracy of the initial model can be determined by calculating the error between the predicted value and the actual value. The accuracy can include the root mean square error (RMSE), the mean absolute error (MAE), the correlation coefficient (R²), etc.

[0049] Based on any of the above embodiments, determining the accuracy based on the predicted stress response and the tested stress response includes: Determine the root mean square error and correlation coefficient based on the predicted stress response and the tested stress response; Determine accuracy based on root mean square error or correlation coefficient.

[0050] Specifically, the smaller the root mean square error, the higher the accuracy. The larger the correlation coefficient, the higher the accuracy. Optionally, the root mean square error can be used as the accuracy, or the correlation coefficient can be used as the accuracy. The root mean square error (RMSE) and the correlation coefficient (R²) can be determined based on the following formula:

[0051] in, Indicates The predicted stress response corresponding to the excitation load of the test time history, represents the total number of excitation loads in the test time history, represents the test stress response; Represents the average of all tested stress responses.

[0052] Based on any of the above embodiments, predicting the fatigue life of the swing arm based on the stress response includes: Perform periodic statistics on the stress history corresponding to the stress response to determine the number of cycles and cycle amplitude of each cycle; Determine the fatigue damage of each cycle based on the number of cycles and cycle amplitude of each cycle; Predict the fatigue life of the swing arm based on the fatigue damage per cycle.

[0053] Specifically, the stress response is analyzed periodically to determine the number of cycles and cycle amplitude of each cycle. The number of cycles refers to the number of stress cycles experienced by the material in one cycle. The cycle amplitude refers to the amplitude of a stress cycle, which is usually the difference between the maximum stress and the minimum stress. Optionally, the rain flow counting method can be used to perform cycle statistics to determine the number of cycles and cycle amplitude of each cycle.

[0054] According to the number of cycles and cycle amplitude of each cycle, the fatigue damage of each cycle can be calculated using a suitable fatigue damage model (such as Miner's linear cumulative damage theory). Among them, the fatigue damage of each cycle can be calculated by the following formula: :

[0055] in, Indicates The number of cycles per cycle, Indicates The cycle amplitude of a cycle.

[0056] The total fatigue life of the swing arm is predicted by accumulating the fatigue damage of all cycles. Usually, the fatigue life of the swing arm is determined by comparing the accumulated damage degree with the maximum allowable damage degree (usually 1).

[0057] Based on any of the above embodiments, the fatigue life of the swing arm is predicted based on the fatigue damage of each cycle, including: The fatigue damage of all cycles is accumulated to obtain the cumulative fatigue damage; Predict the fatigue life of the swing arm based on the accumulated fatigue damage.

[0058] Specifically, the fatigue damage of all cycles is accumulated to obtain the cumulative fatigue damage. When the cumulative fatigue damage reaches or exceeds a threshold value (such as reaching or exceeding 1), fatigue failure is expected to occur. Based on the cumulative fatigue damage, the fatigue life of the swing arm is predicted.

[0059] Figure 3 FIG. 1 is a flow chart of another swing arm fatigue life prediction method provided by the present invention, such as Figure 3 As shown, the method includes: Obtain a long-term load spectrum and a small amount of load-stress time history data, and determine the training set and test set from the long-term load spectrum and a small amount of load-stress time history data. For example, 80% of the data can be used as the training set and 20% of the data can be used as the test set.

[0060] Based on the training set, the initial model is trained and the root mean square error or correlation coefficient of the initial model is determined based on the test set, and the accuracy is determined based on the root mean square error or correlation coefficient. If the accuracy is greater than the threshold, the initial model is used as the stress response prediction model.

[0061] The time-history excitation load of the swing arm is input into the stress response prediction model to obtain the stress response output by the stress response prediction model. The rain flow statistics of the stress response corresponding to the stress history are performed to determine the number of cycles and cycle amplitude of each cycle, and based on the number of cycles and cycle amplitude of each cycle, the fatigue damage of each cycle is determined. The fatigue damage accumulation calculation of all cycles is performed to predict the fatigue life of the swing arm.

[0062] The following is a description of the swing arm fatigue life prediction device provided by the present invention. The swing arm fatigue life prediction device described below and the swing arm fatigue life prediction method described above can be referenced to each other.

[0063] Based on any of the above embodiments, Figure 4 Schematic diagram of the structure of the swing arm fatigue life prediction device provided by the present invention. Figure 4 As shown, the device comprises: An acquisition unit 410 is used to acquire a time-history excitation load of the swing arm; An input unit 420 is used to input the time history excitation load into the stress response prediction model to obtain the stress response output by the stress response prediction model; the stress response prediction model is obtained based on the sample time history excitation load and the sample stress response training; The prediction unit 430 is used to predict the fatigue life of the swing arm based on the stress response.

[0064] Based on any of the above embodiments, the stress response prediction model is trained based on the following steps: Obtain historical time-history excitation loads and historical stress responses corresponding to historical time-history excitation loads; Based on the preset ratio, the historical time history excitation load is divided into the sample time history excitation load and the test time history excitation load, and the proportion of the sample time history excitation load is greater than that of the test time history excitation load; Model training is performed based on the sample time history excitation load and the sample stress response corresponding to the sample time history excitation load to obtain an initial model; Determine the accuracy of the initial model based on the test time history excitation load and the test stress response corresponding to the test time history excitation load; When the accuracy is greater than the threshold, the initial model is used as the stress response prediction model.

[0065] Based on any of the above embodiments, determining the accuracy of the initial model based on the test time history excitation load and the test stress response corresponding to the test time history excitation load includes: Input the test time history excitation load into the initial model to obtain the predicted stress response output by the initial model; Based on the predicted stress response, as well as the tested stress response, the accuracy is determined.

[0066] Based on any of the above embodiments, determining the accuracy based on the predicted stress response and the tested stress response includes: Determine the root mean square error and correlation coefficient based on the predicted stress response and the tested stress response; Determine accuracy based on root mean square error or correlation coefficient.

[0067] Based on any of the above embodiments, predicting the fatigue life of the swing arm based on the stress response includes: Perform periodic statistics on the stress history corresponding to the stress response to determine the number of cycles and cycle amplitude of each cycle; Determine the fatigue damage of each cycle based on the number of cycles and cycle amplitude of each cycle; Predict the fatigue life of the swing arm based on the fatigue damage per cycle.

[0068] Based on any of the above embodiments, the fatigue life of the swing arm is predicted based on the fatigue damage of each cycle, including: The fatigue damage of all cycles is accumulated to obtain the cumulative fatigue damage; Predict the fatigue life of the swing arm based on the accumulated fatigue damage.

[0069] Figure 5 is a schematic diagram of the structure of the electronic device provided by the present invention, such as Figure 5 As shown, the electronic device may include: a processor 510, a communication interface 520, a memory 530 and a communication bus 540, wherein the processor 510, the communication interface 520 and the memory 530 communicate with each other through the communication bus 540. The processor 510 may call the logic instructions in the memory 530 to execute the fatigue life prediction method of the swing arm, the method comprising: obtaining the time history excitation load of the swing arm; inputting the time history excitation load into the stress response prediction model to obtain the stress response output by the stress response prediction model; the stress response prediction model is obtained based on the sample time history excitation load and the sample stress response training; based on the stress response, predicting the fatigue life of the swing arm.

[0070] In addition, the logic instructions in the above-mentioned memory 530 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when it is sold or used as an independent product. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc. Various media that can store program codes.

[0071] On the other hand, the present invention also provides a computer program product, which includes a computer program. The computer program can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the swing arm fatigue life prediction method provided by the above-mentioned methods, and the method includes: obtaining the time-history excitation load of the swing arm; inputting the time-history excitation load into a stress response prediction model to obtain a stress response output by the stress response prediction model; the stress response prediction model is obtained based on sample time-history excitation loads and sample stress responses. Training; based on the stress response, predicting the fatigue life of the swing arm.

[0072] On the other hand, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, it is implemented to execute the swing arm fatigue life prediction method provided by the above-mentioned methods, the method comprising: obtaining the time-history excitation load of the swing arm; inputting the time-history excitation load into a stress response prediction model to obtain a stress response output by the stress response prediction model; the stress response prediction model is obtained based on sample time-history excitation load and sample stress response training; based on the stress response, predicting the fatigue life of the swing arm.

[0073] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.

[0074] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0075] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for predicting fatigue life of a swing arm, characterized in that: include: Obtain the time history excitation load of the swing arm; Inputting the time-history excitation load into a stress response prediction model to obtain a stress response output by the stress response prediction model; The stress response prediction model is obtained based on sample time history excitation load and sample stress response training; Based on the stress response, the fatigue life of the swing arm is predicted.

2. The swing arm fatigue life prediction method according to claim 1, characterized in that: The stress response prediction model is trained based on the following steps: Obtaining historical time-history excitation loads and historical stress responses corresponding to the historical time-history excitation loads; Based on a preset ratio, the historical time history excitation load is divided into the sample time history excitation load and the test time history excitation load, and the proportion of the sample time history excitation load is greater than the proportion of the test time history excitation load; Performing model training based on the sample time-history excitation load and the sample stress response corresponding to the sample time-history excitation load to obtain an initial model; Determining the accuracy of the initial model based on the test time history excitation load and the test stress response corresponding to the test time history excitation load; When the accuracy is greater than a threshold, the initial model is used as the stress response prediction model.

3. The method for predicting fatigue life of a swing arm according to claim 2, characterized in that: The determining the accuracy of the initial model based on the test time history excitation load and the test stress response corresponding to the test time history excitation load includes: Inputting the test time history excitation load into the initial model to obtain a predicted stress response output by the initial model; The accuracy is determined based on the predicted stress response, and the tested stress response.

4. The method for predicting fatigue life of a swing arm according to claim 3, characterized in that: The determining the accuracy based on the predicted stress response and the tested stress response comprises: Determining a root mean square error and a correlation coefficient based on the predicted stress response and the tested stress response; The accuracy is determined based on the root mean square error or the correlation coefficient.

5. The method for predicting fatigue life of a swing arm according to any one of claims 1 to 4, characterized in that: The predicting the fatigue life of the swing arm based on the stress response comprises: Performing periodic statistics on the stress history corresponding to the stress response to determine the number of cycles and cycle amplitude of each cycle; Determine the fatigue damage of each cycle based on the number of cycles and cycle amplitude of each cycle; Based on the fatigue damage in each cycle, the fatigue life of the swing arm is predicted.

6. The method for predicting fatigue life of a swing arm according to claim 5, characterized in that: The method of predicting the fatigue life of the swing arm based on the fatigue damage of each cycle includes: The fatigue damage of all cycles is accumulated to obtain the cumulative fatigue damage; Based on the accumulated fatigue damage, the fatigue life of the swing arm is predicted.

7. A swing arm fatigue life prediction device, characterized in that: include: An acquisition unit is used to acquire the time history excitation load of the swing arm; An input unit, used for inputting the time history excitation load into a stress response prediction model to obtain a stress response output by the stress response prediction model; the stress response prediction model is obtained based on sample time history excitation load and sample stress response training; A prediction unit is used to predict the fatigue life of the swing arm based on the stress response.

8. An electronic 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 swing arm fatigue life prediction method as described in any one of claims 1 to 6 is implemented.

9. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method for predicting the fatigue life of a swing arm as claimed in any one of claims 1 to 6 is implemented.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the method for predicting the fatigue life of a swing arm as claimed in any one of claims 1 to 6 is implemented.

Images