CN-121786983-B - Automobile leaf spring fatigue analysis method based on simulation model optimization

CN121786983BCN 121786983 BCN121786983 BCN 121786983BCN-121786983-B

Abstract

The application relates to the technical field of simulation analysis, in particular to an automobile leaf spring fatigue analysis method based on simulation model optimization, which comprises the steps of sampling a multichannel load spectrum under a test working condition to obtain a load sequence; the method comprises the steps of carrying out grid division on a plate spring model to obtain a plurality of grid nodes, applying load to the plate spring model according to a load sequence, carrying out finite element analysis to obtain stress sequences of the grid nodes under a test working condition, carrying out real-time statistics on accumulated fatigue damage and damage saturation of the grid nodes on the stress sequences by using a rain flow counting method, and constructing a damage cloud picture according to the accumulated fatigue damage of the grid nodes and a second-order differential sequence of the stress sequences at the termination time of the load sequence. According to the technical scheme, the fatigue analysis efficiency and precision of the automobile leaf spring can be improved.

Inventors

LU XIAOJIANG
WANG XIAOHUI
FU YONG

Assignees

山东森德数控机械有限公司
郑州新交通汽车板簧有限公司

Dates

Publication Date: 20260508
Application Date: 20260305

Claims (9)

1. A fatigue analysis method of an automobile leaf spring based on simulation model optimization is characterized by comprising the steps of sampling a multichannel load spectrum under a test working condition to obtain a load sequence; Performing grid division on the plate spring model to obtain a plurality of grid nodes; applying load to the plate spring model according to the load sequence, and performing finite element analysis to obtain stress sequences of grid nodes under a test working condition; the method comprises the steps of utilizing a rain flow counting method to count accumulated fatigue damage and damage saturation of grid nodes on a stress sequence in real time, utilizing the rain flow counting method to extract stress circulation of any grid node in each active segment, calculating the ratio of actual times of stress circulation of each level of stress amplitude to fatigue life limit of a material under corresponding stress amplitude to obtain a single-level damage value, summing the single-level damage values of each level of stress amplitude to obtain damage increment of the grid node in each active segment, wherein the active segment is a time interval when the grid node is not in a dormant node state, counting accumulated fatigue damage of the grid node in real time based on the damage increment in each active segment, dividing the accumulated fatigue damage by a failure threshold to obtain the damage saturation, setting the corresponding grid node as the dormant node in response to the increase rate of the damage saturation, suspending calculation of the rain flow counting method, monitoring the load change of a load sequence, and activating each dormant node in response to the load change is greater than a second threshold to execute the rain flow counting method again; And constructing a damage cloud picture according to the second-order differential sequence of the accumulated fatigue damage and stress sequence of each grid node at the termination moment of the load sequence.
2. The method for analyzing fatigue of an automobile leaf spring based on simulation model optimization according to claim 1, wherein the sampling the multichannel load spectrum under the test condition to obtain the load sequence comprises: acquiring load variation of each moment in a multichannel load spectrum, and calculating average load variation; Comparing the load variation quantity and the average load variation quantity at any moment, and dividing the multichannel load spectrum into a working condition stable section and a working condition abrupt change section; sampling at a first step length in the working condition stabilizing section, and sampling at a second step length in the working condition abrupt section to obtain a load sequence, wherein the second step length is smaller than the first step length.
3. The method for analyzing fatigue of an automobile leaf spring based on simulation model optimization according to claim 1, wherein the step of meshing the leaf spring model to obtain a plurality of mesh nodes comprises: Carrying out grid division on the leaf spring model according to a preset grid size; calculating geometric sensitivity factors of all position points in the plate spring model; And in response to the geometric sensitivity factor being greater than the sensitivity threshold, dividing the grid where the corresponding position point is located again to realize local grid encryption.
4. The method for analyzing fatigue of the automobile leaf spring based on simulation model optimization according to claim 3, wherein the geometric sensitivity factor is positively correlated with the surface principal curvature and the local thickness of the corresponding position point.
5. The automobile leaf spring fatigue analysis method based on simulation model optimization according to claim 1 is characterized in that the finite element analysis is conducted to obtain stress sequences of grid nodes under test conditions, wherein the stress sequences comprise the steps of constructing a nonlinear model from load vectors to node stress through krifen Jin Suanfa, inputting each load vector in the load sequences into the nonlinear model to obtain stress of each grid node under each load vector, and further obtaining stress sequences of each grid node.
6. The method for analyzing fatigue of the automobile leaf spring based on simulation model optimization according to claim 5, wherein the construction process of the nonlinear model comprises the following steps: performing Latin hypercube sampling in a hypercube space consisting of six-way loads to obtain characteristic sample points, wherein the characteristic sample points correspond to one load vector; The method comprises the steps of obtaining stress of each grid node under characteristic sample points as sample data, and splitting the sample data into a training set and a verification set; and in response to the root mean square error not being smaller than a preset error upper limit, adding sample data to continue training the nonlinear model.
7. The automobile leaf spring fatigue analysis method based on simulation model optimization according to claim 1, wherein the method for acquiring the accumulated fatigue damage of any grid node comprises the following steps: acquiring the damage increment of the grid node in each active segment in real time; And accumulating the damage increment counted by the grid node in the current active segment in real time with the damage increment of each previous active segment to obtain updated accumulated fatigue damage.
8. The method for analyzing fatigue of an automobile leaf spring based on simulation model optimization according to claim 1, wherein the multi-channel load spectrum is a time series of six-directional loads at a leaf spring seat, a front lifting lug and a rear lifting lug, and the loads comprise three axial acting forces and three axial torques.
9. The method for analyzing the fatigue of the automobile leaf spring based on the simulation model optimization according to claim 1 is characterized in that constructing the damage cloud picture according to the accumulated fatigue damage of each grid node and the second-order differential sequence of the stress sequence comprises the steps of counting the target number of which the second-order differential value is larger than a differential threshold value in the second-order differential sequence of the stress sequence of any grid node, and calculating the final damage value of the grid node, wherein the final damage value is positively correlated with the accumulated fatigue damage and the target number.

Description

Automobile leaf spring fatigue analysis method based on simulation model optimization Technical Field The application relates to the technical field of simulation analysis, in particular to an automobile leaf spring fatigue analysis method based on simulation model optimization. Background As the automobile industry moves toward weight reduction and long life cycle, the structural reliability of automobile leaf springs, which are the core load bearing elements of suspension systems, is directly related to the running safety of vehicles. In the development stage, it is often necessary to evaluate the fatigue performance of the leaf springs by way of a road test. However, the road test procedure is long in period, costly, and difficult to reproduce extremely random loads. Therefore, the fatigue analysis of leaf springs by using the finite element simulation technique has become a mainstream means in the industry. At present, a common leaf spring fatigue analysis method in industry mostly adopts a unified time step to sample a multi-channel load spectrum, and utilizes a finite element analysis technology to obtain stress sequences of all grid nodes under the multi-channel load spectrum, and in the aspect of fatigue damage statistics, the stress sequences of all grid nodes are calculated in a full time domain based on a rain flow counting method, and finally, the stress circulation of all grid nodes in a model on a full time axis is extracted through the rain flow counting method, and accumulated damage is calculated by comparing with an S-N curve of a material, so that a full-field damage distribution result is generated. However, the load sampling with a fixed step length cannot achieve both the calculation efficiency and the transient impact capture, which easily results in the loss of critical fatigue extreme points and generates a large amount of data redundancy. According to the method, linear accumulated damage superposition is performed only according to the S-N curve of the material, damage of stress impact to the automobile leaf spring is ignored, and therefore the efficiency and the accuracy of fatigue analysis of the automobile leaf spring are low. Disclosure of Invention In order to solve the technical problem of low efficiency and precision of automobile leaf spring fatigue analysis, the application provides an automobile leaf spring fatigue analysis method based on simulation model optimization, which can improve the efficiency and precision of automobile leaf spring fatigue analysis. The application provides an automobile leaf spring fatigue analysis method based on simulation model optimization, which comprises the steps of sampling a multichannel load spectrum under a test working condition to obtain a load sequence, meshing a leaf spring model to obtain a plurality of grid nodes, applying load to the leaf spring model according to the load sequence, carrying out finite element analysis to obtain stress sequences of grid nodes under the test working condition, and carrying out real-time statistics on accumulated fatigue damage and damage saturation of the grid nodes on the stress sequences by using a rain flow counting method. The dormancy state of the grid nodes is judged through the damage saturation, and an activation mechanism of the grid nodes is triggered based on the load variation, so that dynamic allocation of computing resources is realized, redundant computation of a large number of non-critical areas can be eliminated, accumulated fatigue damage is corrected by utilizing a second-order differential sequence of a stress sequence, potential fatigue weak points caused by load mutation are prevented from being missed, and accuracy of fatigue life assessment is ensured. The method comprises the steps of obtaining load variation of each moment in a multichannel load spectrum, calculating average load variation, comparing the load variation at any moment with the average load variation, dividing the multichannel load spectrum into a working condition stable section and a working condition abrupt change section, sampling the working condition stable section with a first step length, and sampling the working condition abrupt change section with a second step length, wherein the second step length is smaller than the first step length. The self-adaptive sampling step length driven by the load variation is adopted, the data scale is reduced by adopting a large step length in the working condition stable section, the stress peak value is captured by adopting a small step length in the working condition abrupt change section, and the preprocessing data volume of finite element analysis is greatly reduced on the premise of ensuring that the random impact characteristic of the road surface is not lost. Preferably, the grid division of the leaf spring model to obtain a plurality of grid nodes comprises the steps of grid division of the leaf spring model according to preset grid sizes, calculation