Vehicle frame multi-dimensional load spectrum accelerated compilation method based on WLTC working condition

Abstract

The invention discloses a multi-dimensional load spectrum acceleration compilation method for a vehicle frame based on a WLTC working condition, and the method comprises the following steps: setting rigid connection of the vehicle frame, and building a whole vehicle dynamics model; creating a WLTC event working condition, carrying out whole vehicle dynamics simulation, and extracting a load at a connection point; performing finite element analysis by using an inertial release method; carrying out fatigue life analysis on the frame, and extracting fatigue hot spot numbers; reconstructing an analysis process, and extracting stress time sequence data; carrying out pseudo damage analysis on the stress time spectrum to intercept an original load spectrum; carrying out smooth connection processing on the intercepted acceleration spectrum fragments; and performing power spectrum density analysis on the acceleration spectrum and the original spectrum, screening the acceleration spectrum according to a result, and finally obtaining the multi-dimensional load acceleration spectrum of the vehicle frame based on the WLTC working condition, thereby shortening the bench test time and reducing the development cost of the vehicle frame.

Description

Technical Field

[0001] The present invention relates to a method for compiling a load spectrum of a vehicle frame under a WLTC working condition, and in particular to a method for compiling a multi-dimensional load spectrum of a vehicle frame from a random road surface under a WLTC working condition. Background Art

[0002] With the development of the automobile industry, the durability of the frame, as an important load-bearing structure, directly affects the safety and service life of the vehicle. The traditional frame load spectrum compilation method mainly relies on standard working condition experiments, but it cannot fully reflect the multi-dimensional loads under complex road conditions. The WLTC working condition simulates a variety of road conditions that are closer to actual driving. The frame is subjected to random loads from multiple directions during actual driving. The existing accelerated durability test method does not fully consider these complex factors, resulting in insufficient test accuracy and difficulty in reflecting the actual service life of the frame within a limited time.

[0003] Therefore, a method for accelerating the compilation of multidimensional load spectra of random road surfaces under WLTC conditions is needed to improve the accuracy and efficiency of frame durability testing, optimize the design and shorten the test cycle. In the process of compiling the accelerated life load spectrum, it is necessary to follow the basic criterion of damage equivalence before and after the load spectrum is accelerated.

[0004] How to accelerate the compilation of random road load spectra from the vehicle frame under WLTC conditions to shorten the bench test time and reduce development costs has become a problem that needs to be solved at present. Summary of the invention

[0005] In view of the defects existing in the prior art, the purpose of the present invention is to achieve the above purpose. The technical solution adopted by the present invention is: the method for accelerating the compilation of a multi-dimensional load spectrum of a frame under WLTC working conditions is characterized by being applied to fatigue life calculation and bench test of automobile frames under WLTC working conditions, and comprises the following steps:

[0006] Step 1: Set the rigid connection points of the frame, output the MNF flexible body file through finite element simulation, and create a vehicle dynamics model. The rigid connection points include: frame and subframe, front shock absorber, rear leaf spring, rear shock absorber, and rear buffer block;

[0007] Step 2: Create a WLTC event condition, perform vehicle dynamics simulation analysis, and extract the loads at the connection points;

[0008] Step 2.1: In the dynamic simulation analysis, the WLTC standard data condition is imported through the event editor to generate the WLTC event condition, and the road surface file generator function is used to generate a random road surface. The road surface parameters are set according to the international standards to meet the spectrum requirements of the road surface;

[0009] Step 2.2: Extract the excitation load spectrum files in the three directions of X, Y, and Z of each connection point through vehicle dynamics simulation to obtain the original load spectrum compiled by the accelerated life load spectrum;

[0010] Step 3: Perform finite element analysis using the inertia release method;

[0011] Step 3.1: Use the finite element analysis software to set the X, Y, and Z forces at the connection points and set the corresponding load steps as the load spectrum connection channels. The load force is set to 1N, and the number of channels is consistent with the original spectrum dimension;

[0012] Step 3.2: Set up the control card for finite element analysis (1) automatically constrain unconstrained nodes (2) define inertia release (3) generate output result file op2;

[0013] Step 3.3: Use the inertia release method to perform finite element calculations to obtain the inertia release method analysis results;

[0014] Step 4: Perform fatigue life analysis on the frame and extract fatigue hotspot numbers;

[0015] Step 4.1: Set the SN curve of the corresponding material in the software analysis module, connect the connection channel of the inertia release method analysis result with the original load spectrum one by one, set the stress correction method to Goodman, and obtain the fatigue hot spot number;

[0016] Step 5: Reconstruct the analysis process and extract stress time series data;

[0017] Step 5.1: Re-establish the SN fatigue analysis process for the fatigue hotspot to obtain a new analysis process independent of the original fatigue life analysis;

[0018] Step 5.2: For the newly created analysis process, copy the material parameters of the original fatigue analysis, input the fatigue hot spot number as the extraction point number, and recalculate to obtain the upper and lower surface stress time series of the specified node;

[0019] Step 5.3: For the stress time series of the upper and lower surfaces of the specified node, select the stress time spectrum with the largest absolute value and amplitude change as the stress time spectrum for subsequent analysis;

[0020] Step 6: Intercept the original load spectrum by performing pseudo damage analysis on the stress-time spectrum;

[0021] Step 6.1: By setting different window lengths, the stress-time spectrum and the original load spectrum are respectively segmented to obtain stress spectrum and original load spectrum segments under different windows;

[0022] Step 6.2: Calculate the damage value of the stress spectrum segment using a pseudo-damage equivalent method, set a damage cutoff value, and mark the stress spectrum segment below the damage cutoff value;

[0023] Step 6.3: The original load spectrum and the stress-time spectrum should have a one-to-one correspondence in the time domain. The original load spectrum segments are marked according to the correspondence, and the marked original load spectrum segments are deleted to obtain the remaining acceleration spectrum segments after the segments with different window lengths are cut;

[0024] Step 7: Smoothly connect the captured acceleration spectrum segments;

[0025] Step 7.1: Use equation (1) to calculate the connection signal y1 of the acceleration spectrum fragments remaining after the fragments of different window lengths are cut, and use the connection signal to smoothly connect the head and tail of the remaining window fragments to obtain the acceleration spectra under different window lengths;

[0026]

[0027] Where α1 is the last data of the previous segment that is retained, and α2 is the first data of the next segment that is retained;

[0028] Step 8: Perform power spectral density analysis on the acceleration spectrum and the original spectrum, and select the acceleration spectrum according to the results;

[0029] Step 8.1: Use formula (2) to perform Fourier transform on the original load spectrum and the acceleration spectrum under different window lengths to obtain discrete frequency domain signal X(k), and use formula (3) to calculate the discrete frequency domain signal X(k) to obtain the original load spectrum and the power spectrum density under different window lengths;

[0030]

[0031] Where x[n] is the original load spectrum and the acceleration spectrum of different window lengths, N is the length of the signal, j is the imaginary unit, and k is the kth signal;

[0032] Step 8.2: Screen the acceleration spectra under the different windows, and the screening criterion is to select the power spectrum density curve that best fits the power spectrum density curve of the original load spectrum, so as to obtain the final accelerated life load spectrum.

[0033] In step 8.2, the power spectrum density curve is screened as follows:

[0034] The power spectral density of the different window lengths and the power spectral density of the original load spectrum are grouped according to different window values. Each group is the power spectral density under a window value and the power spectral density of the original load spectrum. The Euclidean distance d of each group is calculated using formula (4).k , choose the Euclidean distance d k The smallest group has the power spectrum density under the corresponding window length that best fits the power spectrum density of the original load spectrum;

[0035]

[0036] Where x ki is the ith data of the power spectrum density under a window length in the kth group, y i is the i-th data of the power spectrum density of the original load spectrum.

[0037] On the basis of the above technical solution, compared with the prior art, the advantages of the present invention are:

[0038] 1. The accelerated compilation method of the multi-dimensional load spectrum of the frame under the WLTC working condition proposed in the present invention is the original spectrum excitation obtained through the whole vehicle dynamics simulation under the WLTC working condition. The traditional frame load spectrum compilation method is mostly based on finite element simulation and simple standard working condition experiments, but these methods are usually difficult to fully reflect the complex conditions of real roads. In particular, with the increasingly diverse working condition requirements of passenger cars and commercial vehicles, such as the World Light Vehicle Test Cycle (WLTC), its simulated working conditions are closer to the actual road driving environment, covering a variety of traffic scenarios such as urban roads, highways, and rural roads. Under such complex working conditions, the frame is subjected to multi-dimensional loads from different directions, such as vertical, lateral, and longitudinal. At the same time, the irregularity of the random road surface increases the diversity and randomness of the frame load.

[0039] 2. The present invention converts the compilation of multidimensional load spectrum into one-dimensional load spectrum compilation (stress time series spectrum at damage hotspot) based on the principle of loss equivalence, and uses the one-to-one correspondence between the one-dimensional load spectrum and the original spectrum in the time dimension to compile the multidimensional load spectrum, which not only takes into account the interference problem between the load spectra of each dimension in a macroscopic sense, but also greatly reduces the difficulty and cost of compilation. For the selection of the window, the frequency domain characteristics of the load spectrum are considered as much as possible, which ensures the accuracy and efficiency of the load spectrum acceleration compilation process. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] Figure 1 The present invention is a flow chart of the method for accelerating the compilation of multi-dimensional load spectra of a vehicle frame under WLTC conditions.

[0041] Figure 2 It is the original load spectrum of the frame connection point in the Z direction in the example of the present invention.

[0042] Figure 3 It is a time series diagram of the top and bottom layer stresses at the damage hotspot in an example of the present invention.

[0043] Figure 4It is a label diagram of the stress time series at the damage hotspot when the window value is 5 seconds in the present invention.

[0044] Figure 5a It is the power spectrum density diagram under the original load spectrum in the example of the present invention.

[0045] Figure 5b It is the power spectrum density diagram of the acceleration spectrum when the window length is 5 seconds in the example of the present invention;

[0046] Figure 6a This is a before and after comparison diagram of the X-direction load spectrum in the example of the present invention.

[0047] Figure 6b A before and after comparison diagram of the Y-direction load spectrum in the example of the present invention.

[0048] Figure 6c A before and after comparison diagram of the Z-direction load spectrum in the example of the present invention. DETAILED DESCRIPTION

[0049] In this embodiment, the accelerated compilation method of the frame multidimensional load spectrum under the WLTC condition is applied to the B-level random road WLTC condition. The frame multidimensional load spectrum data is extracted through vehicle dynamics simulation, and the multidimensional load spectrum is accelerated to ensure the efficiency of acceleration while ensuring that the frame fatigue failure hotspots before and after acceleration are consistent and the frame fatigue damage before and after acceleration is equivalent.

[0050] Figure 1 The flowchart of the method for accelerating the compilation of the multi-dimensional load spectrum of the vehicle frame under the WLTC working condition in this embodiment is shown, which includes the following steps:

[0051] Step 1: Set the rigid connection points of the frame, output the MNF flexible body file through finite element simulation, and create a vehicle dynamics model. The rigid connection points include: frame and subframe, front shock absorber, rear leaf spring, rear shock absorber, and rear buffer block;

[0052] Step 2: Create a WLTC event condition, perform vehicle dynamics simulation analysis, and extract the loads at the connection points;

[0053] Step 2.1: In the dynamic simulation analysis, the WLTC standard data condition is imported through the event editor to generate the WLTC event condition, and the road surface file generator function is used to generate a random road surface. The road surface parameters are set according to the international standards to meet the spectrum requirements of the road surface;

[0054] Step 2.2: Extract the excitation load spectrum files of each connection point in the X, Y, and Z directions through the vehicle dynamics simulation, and obtain the original load spectrum compiled by the accelerated life load spectrum. The original load spectrum of the frame connection point in the Z direction is as follows: Figure 2 As shown;

[0055] Step 3: Perform finite element analysis using the inertia release method in HyperWorks software;

[0056] Step 3.1: Use the finite element analysis software to set the X, Y, and Z forces at the connection points and set the corresponding load steps as the load spectrum connection channels. The load force is set to 1N, and the number of channels is consistent with the original spectrum dimension;

[0057] Step 3.2: Set up the control card for finite element analysis (1) automatically constrain unconstrained nodes (2) define inertia release (3) generate output result file op2;

[0058] Step 3.3: Use the inertia release method to perform finite element calculations to obtain the inertia release method analysis results;

[0059] Step 4: Perform fatigue life analysis on the frame and extract fatigue hotspot numbers;

[0060] Step 4.1: Set the SN curve of the corresponding material in the software analysis module, connect the connection channel of the inertia release method analysis result with the original load spectrum one by one, set the stress correction method to Goodman, and obtain the fatigue hot spot number;

[0061] Step 5: Reconstruct the analysis process and extract stress time series data;

[0062] Step 5.1: Re-establish the SN fatigue analysis process for the fatigue hotspot to obtain a new analysis process independent of the original fatigue life analysis;

[0063] Step 5.2: For the new analysis process, copy the material parameters of the original fatigue analysis, enter the fatigue hotspot number as the extraction point number, and recalculate the upper and lower surface stress time series of the specified node as follows: Figure 3 As shown;

[0064] Step 5.3: For the stress time series of the upper and lower surfaces of the specified node, select the stress time spectrum with the largest absolute value and amplitude change as the stress time spectrum for subsequent analysis;

[0065] Step 6: Intercept the original load spectrum by performing pseudo damage analysis on the stress-time spectrum;

[0066] Step 6.1: By setting different window lengths, the window lengths range from 1 second to 10 seconds, and the interval is 1 second, the stress time spectrum and the original load spectrum are respectively segmented to obtain the stress spectrum and the original load spectrum segments under different windows;

[0067] Step 6.2: Use the pseudo-damage equivalent method to calculate the damage value of the stress spectrum segment, set the damage cutoff value, and mark the stress spectrum segments below the damage cutoff value. When the window value is 5 seconds, the marking diagram of the stress time series at the damage hotspot is as follows: Figure 4 As shown;

[0068] Step 6.3: The original load spectrum and the stress-time spectrum should have a one-to-one correspondence in the time domain. The original load spectrum segments are marked according to the correspondence, and the marked original load spectrum segments are deleted to obtain the remaining acceleration spectrum segments after the segments with different window lengths are cut;

[0069] Step 7: Smoothly connect the captured acceleration spectrum segments;

[0070] Step 7.1: Use equation (1) to calculate the connection signal y1 of the acceleration spectrum fragments remaining after the fragments of different window lengths are cut, and use the connection signal to smoothly connect the head and tail of the remaining window fragments to obtain the acceleration spectra under different window lengths;

[0071]

[0072] Where α1 is the last data of the previous segment that is retained, and α2 is the first data of the next segment that is retained;

[0073] Step 8: Perform power spectrum density analysis on the acceleration spectrum and the original spectrum. The power spectrum density of the original load spectrum is obtained as follows: Figure 5a When the window length is 5 seconds, the power spectrum density diagram of the acceleration spectrum is as follows: Figure 5b ;

[0074] Step 8.1: Use formula (2) to perform Fourier transform on the original load spectrum and the acceleration spectrum under different window lengths to obtain discrete frequency domain signal X(k), and use formula (3) to calculate the discrete frequency domain signal X(k) to obtain the original load spectrum and the power spectrum density under different window lengths;

[0075]

[0076] Where x[n] is the original load spectrum and the acceleration spectrum of different window lengths, N is the length of the signal, j is the imaginary unit, and k is the kth signal;

[0077] Step 8.2: Screen the acceleration spectra under the different windows, and the screening criterion is to select the power spectrum density curve that best fits the power spectrum density curve of the original load spectrum, so as to obtain the final accelerated life load spectrum; in this embodiment, the X-direction acceleration spectrum at the connection point is compared with the original spectrum. Figure 6a , compare the Y-axis acceleration spectrum at the connection point with the original spectrum Figure 6b, compare the Z-direction acceleration spectrum at the connection point with the original spectrum Figure 6c .

[0078] In step 8.2, the power spectrum density curve is screened as follows:

[0079] The power spectral density of the different window lengths and the power spectral density of the original load spectrum are grouped according to different window values. Each group is the power spectral density under a window value and the power spectral density of the original load spectrum. The Euclidean distance d of each group is calculated using formula (4). k , choose the Euclidean distance d k The smallest group has the power spectrum density under the corresponding window length that best fits the power spectrum density of the original load spectrum;

[0080]

[0081] Where x ki is the ith data of the power spectrum density under a window length in the kth group, y i is the i-th data of the power spectrum density of the original load spectrum.

Claims

1. A method for accelerating the compilation of multi-dimensional load spectrum of a vehicle frame under WLTC working conditions, characterized in that: The specific steps include: Step 1: Set the rigid connection points of the frame, output the MNF flexible body file through finite element simulation, and create a vehicle dynamics model. The rigid connection points include: Frame and subframe, front shock absorber, rear leaf spring, rear shock absorber, rear buffer block; Step 2: Create a WLTC event condition, perform vehicle dynamics simulation analysis, and extract the loads at the connection points; Step 2.1: In the dynamic simulation analysis, import the standard data condition through the event editor to generate the WLTC event condition, use the road surface file generator function to generate a random road surface, and set its road surface parameters according to international standards to meet the spectrum requirements of its road surface; Step 2.2: Extract the excitation load spectrum files in the three directions of X, Y, and Z of each connection point through vehicle dynamics simulation to obtain the original load spectrum compiled by the accelerated life load spectrum; Step 3: Perform finite element analysis using the inertia release method; Step 3.1: Use the finite element analysis software to set the X, Y, and Z forces at the connection points and set the corresponding load steps as the load spectrum connection channels. The load force is set to 1N, and the number of channels is consistent with the original spectrum dimension; Step 3.2: Set up the control card for finite element analysis (1) automatically constrain unconstrained nodes (2) define inertia release (3) generate output result file op2; Step 3.3: Use the inertia release method to perform finite element calculations to obtain the inertia release method analysis results; Step 4: Perform fatigue life analysis on the frame and extract fatigue hotspot numbers; Step 4.1: Set the SN curve of the corresponding material in the software analysis module, connect the connection channel of the inertia release method analysis result with the original load spectrum one by one, set the stress correction method to Goodman, and obtain the fatigue hot spot number; Step 5: Reconstruct the analysis process and extract stress time series data; Step 5.1: Re-establish the SN fatigue analysis process for the fatigue hotspot to obtain a new analysis process independent of the original fatigue life analysis; Step 5.2: For the newly created analysis process, copy the material parameters of the original fatigue analysis, input the fatigue hot spot number as the extraction point number, and recalculate to obtain the upper and lower surface stress time series of the specified node; Step 5.3: For the stress time series of the upper and lower surfaces of the specified node, select the stress time spectrum with the largest absolute value and amplitude change as the stress time spectrum for subsequent analysis; Step 6: Intercept the original load spectrum by performing pseudo damage analysis on the stress-time spectrum; Step 6.1: By setting different window lengths, the stress-time spectrum and the original load spectrum are respectively segmented to obtain stress spectrum and original load spectrum segments under different windows; Step 6.2: Calculate the damage value of the stress spectrum segment using a pseudo-damage equivalent method, set a damage cutoff value, and mark the stress spectrum segment below the damage cutoff value; Step 6.3: The original load spectrum and the stress-time spectrum should have a one-to-one correspondence in the time domain. The original load spectrum segments are marked according to the correspondence, and the marked original load spectrum segments are deleted to obtain the remaining acceleration spectrum segments after the segments with different window lengths are cut; Step 7: Smoothly connect the captured acceleration spectrum segments; Step 7.1: Use equation (1) to calculate the connection signal y1 of the acceleration spectrum fragments remaining after the fragments of different window lengths are cut, and use the connection signal to smoothly connect the head and tail of the remaining window fragments to obtain the acceleration spectra under different window lengths; In formula (1), α1 is the last data of the previous segment that is retained, and α2 is the first data of the next segment that is retained; Step 8: Perform power spectral density analysis on the acceleration spectrum and the original spectrum, and select the acceleration spectrum according to the results; Step 8.1: Use formula (2) to perform Fourier transform on the original load spectrum and the acceleration spectrum under different window lengths to obtain discrete frequency domain signal X(k), and use formula (3) to calculate the discrete frequency domain signal X(k) to obtain the original load spectrum and the power spectrum density under different window lengths; In formula (2), x[n] is the original load spectrum and the acceleration spectrum of different window lengths, N is the length of the signal, j is the imaginary unit, and k is the kth signal; Step 8.2: Screen the acceleration spectra under the different windows, and the screening criterion is to select the power spectrum density that best fits the power spectrum density of the original load spectrum to obtain the final accelerated life load spectrum.

2. The method for accelerating the compilation of a multi-dimensional load spectrum of a vehicle frame under WLTC conditions according to claim 1 is characterized in that: In step 8.2, the power spectrum density curve is screened as follows: The power spectral density of the different window lengths and the power spectral density of the original load spectrum are grouped according to different window values. Each group is the power spectral density under a window value and the power spectral density of the original load spectrum. The Euclidean distance d of each group is calculated using formula (4). k , choose the Euclidean distance d k The smallest group has the power spectrum density under the corresponding window length that best fits the power spectrum density of the original load spectrum; In formula (4), x ki is the ith data of the power spectrum density under a window length in the kth group, y i is the i-th data of the power spectrum density of the original load spectrum.

Images