Search

CN-121980691-A - Double front axle steering point sensitivity analysis and joint optimization method and system

CN121980691ACN 121980691 ACN121980691 ACN 121980691ACN-121980691-A

Abstract

The invention discloses a double front axle steering point sensitivity analysis and joint optimization method and a system, and relates to the technical field of steering point optimization. The method comprises the steps of S1, collecting corner data, carrying out data preprocessing, S2, carrying out frequency domain analysis on corners of front axle steering vertical arms, judging the non-constant-speed disturbance intensity of universal joints, analyzing and removing periodic disturbance components in the corners of the wheels to obtain wheel corner trends, S3, establishing ideal steering relations, evaluating deviation of the wheel corner trends and the ideal steering relations, identifying steering geometric error degrees, carrying out single-axis step adjustment on coordinates of key structural points, analyzing and screening optimized candidate positions, S4, constructing a joint optimization target, and carrying out iterative updating by combining the non-constant-speed disturbance intensity of the universal joints to obtain the coordinates of the optimized key structural points. The method solves the problem that structural point position adjustment misjudgment is caused by the fact that the differential speed disturbance of the universal joint and steering geometric deviation are difficult to distinguish in the point position optimization process of the double front axle steering system.

Inventors

  • XIA FUGEN
  • WANG BO
  • YAN DAPENG

Assignees

  • 成都壹为新能源汽车有限公司

Dates

Publication Date
20260505
Application Date
20260408

Claims (10)

  1. 1. The double front axle steering point sensitivity analysis and joint optimization method is characterized by comprising the following steps of: S1, acquiring the rotation angle data of a steering wheel, a front axle steering vertical arm and each wheel in real time, carrying out synchronous processing, abnormal correction and normalization processing on each rotation angle data, and establishing a corresponding relation between the rotation angle of the steering wheel and the rotation angle of each wheel; s2, carrying out frequency domain analysis based on a corner sequence of a front axle steering vertical arm, judging the non-constant-velocity disturbance intensity of the universal joint according to a frequency domain analysis result, analyzing periodic disturbance components by combining the frequency domain analysis result and the non-constant-velocity disturbance intensity of the universal joint, and removing the periodic disturbance components in wheel corners to obtain wheel corner trends; S3, establishing an ideal steering relation according to geometric parameters of the vehicle, carrying out deviation evaluation on a wheel turning angle trend and the ideal steering relation, and identifying the steering geometric error degree according to a deviation evaluation result; And S4, taking the optimized candidate positions as point position optimization variables, constructing a joint optimization target based on the steering geometric error degree, and carrying out iterative updating by combining the non-constant velocity disturbance intensity constraint of the universal joint to obtain the optimized key structure point coordinates.
  2. 2. The dual front axle steering point sensitivity analysis and joint optimization method according to claim 1, wherein the specific processes of acquiring the steering angle data of the steering wheel, the front axle steering vertical arm and each wheel in real time, performing synchronous processing, anomaly correction and normalization processing on each steering angle data, and establishing the corresponding relationship between the steering wheel steering angle and each wheel steering angle are as follows: In the running process of a steering system simulation model, steering motion data are acquired in real time, wherein the steering motion data comprise steering wheel corners, front axle steering vertical arm angles and four wheel corners, and the four wheel corners comprise a first front axle left wheel corner, a first front axle right wheel corner, a second front axle left wheel corner and a second front axle right wheel corner; Resampling steering motion data according to a uniform sampling frequency, taking steering wheel corners as uniform independent variables, sorting all sampling points in ascending order according to the numerical value of the steering wheel corners, calculating the wheel corners under the corresponding steering wheel angles by using a linear interpolation method to form a one-to-one correspondence between the four wheel corners and the steering wheel corners, calculating the mean value and standard deviation of each steering motion data in a sliding time window, identifying abnormal points based on the corresponding mean value and standard deviation and correcting by using a linear interpolation method, smoothing each steering motion data by using median filtering, executing minimum and maximum normalization processing, establishing a steering point optimization database, and storing the original and preprocessed steering motion data.
  3. 3. The dual front axle steering point sensitivity analysis and joint optimization method according to claim 1, wherein the specific process of determining the differential speed disturbance intensity of the universal joint according to the frequency domain analysis result is as follows: extracting a front axle steering vertical arm angle sequence, segmenting the front axle steering vertical arm angle sequence according to the length of a fixed window, and dividing the front axle steering vertical arm angle sequence into time windows according to a fixed step length moving window function; Performing short-time Fourier transform on the front axle steering vertical arm angle sequence in each time window, converting the front axle steering vertical arm angle sequence into a time-frequency domain signal to obtain frequency spectrum distribution of the front axle steering vertical arm angles in different time windows, and calculating spectral energy corresponding to different frequencies in each time window based on the modular square of a frequency spectrum coefficient; The method comprises the steps of integrating and accumulating spectrum energy in a corresponding disturbance frequency band range in each time window to obtain disturbance energy, integrating and accumulating the spectrum energy in a range from 0 to Nyquist frequency based on the spectrum energy in a full frequency band, adding an extremely small constant value to obtain total energy, calculating the ratio of the disturbance energy to the total energy, carrying out natural index operation by taking the opposite number, and subtracting a natural index operation result to obtain the non-constant disturbance intensity value of the universal joint.
  4. 4. The dual front axle steering point sensitivity analysis and joint optimization method according to claim 1, wherein the specific process of analyzing the periodic disturbance component by combining the frequency domain analysis result and the universal joint non-constant velocity disturbance intensity is as follows: Calculating the non-constant-velocity disturbance intensity value of the universal joint in each time window, and judging that the non-constant-velocity disturbance of the universal joint exists in the corresponding time window when the non-constant-velocity disturbance intensity value of the universal joint is larger than the disturbance identification threshold value; For each time window, respectively acquiring a first front axle left wheel corner, a first front axle right wheel corner, a second front axle left wheel corner and a second front axle right wheel corner sequence according to sampling intervals corresponding to the time windows, constructing a band-pass filter by taking a corresponding disturbance frequency band range as a passband, carrying out band-pass filtering on each wheel corner sequence to obtain periodic disturbance components of each wheel corner in the time windows, combining the periodic disturbance components of each time window according to time sequences, mapping the periodic disturbance components of each time window to a unified time sequence corresponding to the wheel corners, and recording the periodic disturbance components of the corresponding wheel corners as zero for the time windows without universal joint non-constant speed disturbance.
  5. 5. The dual front axle steering point sensitivity analysis and joint optimization method according to claim 1, wherein the specific process of removing the periodic disturbance component in the wheel corner to obtain the wheel corner trend is as follows: The method comprises the steps of carrying out point-by-point subtraction on periodic disturbance components of wheel corners and corresponding wheel corners at the same sampling time to obtain wheel corner trend components after the non-constant speed disturbance of the universal joint is removed, extracting the wheel corner trend components, constructing a wheel corner trend component sequence, taking steering wheel corners as independent variables, and establishing a one-to-one correspondence relationship between the wheel corner trend components and the steering wheel corners.
  6. 6. The method for analyzing and jointly optimizing the sensitivity of the steering point positions of the double front axles according to claim 1, wherein the specific process of establishing an ideal steering relation according to the geometric parameters of the vehicle and evaluating the deviation of the steering angle trend of the wheels from the ideal steering relation is as follows: The method comprises the steps of reading vehicle geometric parameters from a steering system simulation model, wherein the vehicle geometric parameters comprise a vehicle wheelbase and a front axle wheelbase, calculating a first front axle left wheel ideal corner and a first front axle right wheel ideal corner based on an Ackermann steering geometric relation according to the vehicle wheelbase and the front axle wheelbase by taking a steering wheel corner as independent variables, and calculating a second front axle left wheel ideal corner and a second front axle right wheel ideal corner according to a steering proportion relation of a double front axle steering mechanism to obtain four wheel ideal corners; and reading corresponding wheel angle trend components and ideal wheel angles at the corner positions of each steering wheel, and calculating the difference value between the wheel angle trend components and the ideal wheel angles to obtain wheel angle trend deviation.
  7. 7. The dual front axle steering point sensitivity analysis and joint optimization method according to claim 1, wherein the specific process of identifying the steering geometric error degree according to the deviation evaluation result is as follows: For each wheel, integrating and accumulating the squares of the trend turning angle deviations of the wheels at all turning angle positions of the steering wheel in the turning angle range of the steering wheel to obtain turning angle deviation accumulated values of the wheels; Calculating the square of the change rate of the ideal turning angle of the corresponding wheel to the turning angle of the steering wheel in the same turning angle range of the steering wheel, and integrating and accumulating the squares of the change rates of all turning angle positions of the steering wheel to obtain an ideal turning angle change accumulated value; Dividing the angle deviation cumulative value by the sum of the ideal angle variation cumulative value and the minimum constant value to obtain a trend geometric deviation intensity value of the corresponding wheel, and calculating the average value of the trend geometric deviation intensity values of the four wheels to obtain a comprehensive trend geometric deviation value.
  8. 8. The method for analyzing and jointly optimizing the sensitivity of the steering point positions of the double front axles according to claim 1, wherein the method for adjusting the single-axle step length of the coordinates of the key structure points is characterized in that the method for analyzing the change characteristics of the steering geometric error degree, analyzing the sensitivity of the key structure points according to the change characteristics, and screening out the optimized candidate positions comprises the following specific processes: Reading three-dimensional coordinates of key structural points from a steering system simulation model, wherein the key structural points comprise a trapezoid mechanism connecting point, a middle steering mechanism connecting point and a steering pull rod connecting point; Under the condition of keeping other structural parameters unchanged, aiming at each key structural point, carrying out uniaxial coordinate increment adjustment on the corresponding three-dimensional coordinates according to the minimum coordinate adjustment step length, namely changing only one direction coordinate at a time and keeping the other two coordinates unchanged; Calculating a difference value between the new comprehensive trend geometric deviation value and the original comprehensive trend geometric deviation value, dividing the difference value by a corresponding minimum coordinate adjustment step length to obtain sensitivity values of the key structure points in three coordinate directions, combining the sensitivity values in the three directions to form a point position sensitivity vector of the key structure points, and calculating a modulus value of the point position sensitivity vector as the comprehensive sensitivity value of the key structure points; and sequencing the comprehensive sensitivity values of all the key structure points, and selecting the key structure points with the comprehensive sensitivity values positioned at the first N to construct a front axle steering point position optimizing candidate position set.
  9. 9. The dual front axle steering point sensitivity analysis and joint optimization method according to claim 1, wherein the optimization candidate position is used as a point optimization variable, a joint optimization target is constructed based on the steering geometric error degree, iterative updating is performed by combining the non-constant velocity disturbance intensity constraint of the universal joint, and the specific process of obtaining the optimized key structure point coordinates is as follows: The method comprises the steps of obtaining three-dimensional coordinates of a front axle steering point position optimization candidate position, taking a comprehensive trend geometric deviation value as an optimization index, taking the minimum comprehensive trend geometric deviation value as an optimization target, constructing a point position joint optimization objective function, and updating the three-dimensional coordinates in an iterative search mode; Updating the optimized three-dimensional coordinate to a steering system simulation model, rerun the steering system simulation model, comparing the recalculated comprehensive trend geometrical deviation value with the comprehensive trend geometrical deviation value obtained by the previous round of optimization calculation, updating the three-dimensional coordinate and entering the next round of optimization iterative calculation when the comprehensive trend geometrical deviation value is reduced, and stopping the three-dimensional coordinate updating and outputting the current three-dimensional coordinate as the optimized steering point when the comprehensive trend geometrical deviation value is not reduced or reaches the optimization termination condition.
  10. 10. The double front axle steering point position sensitivity analysis and joint optimization system is characterized by comprising: The steering data acquisition and processing module is used for acquiring the steering wheel, the front axle steering vertical arm and the corner data of each wheel in real time, carrying out synchronous processing, abnormal correction and normalization processing on the corner data, and establishing a corresponding relation between the steering wheel corner and the wheel corner; The non-constant-speed disturbance separation module is used for carrying out frequency domain analysis based on a corner sequence of the front axle steering vertical arm, judging the non-constant-speed disturbance intensity of the universal joint according to the frequency domain analysis result, analyzing periodic disturbance components by combining the frequency domain analysis result and the non-constant-speed disturbance intensity of the universal joint, and removing the periodic disturbance components in the wheel corners to obtain the wheel corner trend; The front axle steering point position analysis module is used for establishing an ideal steering relation according to the geometric parameters of the vehicle, carrying out deviation evaluation on the steering angle trend of the wheels and the ideal steering relation, and identifying the steering geometric error degree according to the deviation evaluation result; the point position joint optimization module is used for constructing a joint optimization target based on the steering geometric error degree by taking the optimization candidate position as a point position optimization variable, and carrying out iterative updating by combining the non-constant disturbance intensity constraint of the universal joint to obtain the optimized key structure point coordinates.

Description

Double front axle steering point sensitivity analysis and joint optimization method and system Technical Field The invention relates to the technical field of steering point position optimization, in particular to a dual front axle steering point position sensitivity analysis and joint optimization method and system. Background With the increasing demands of heavy vehicles and special engineering vehicles for low-speed maneuverability, tire wear control and steering stability, the application of dual front axle steering structures in multi-axle vehicles is increasing. The double front axle steering system is generally composed of a steering gear, a steering connecting rod mechanism and a multi-stage transmission part, and the steering performance matching of the vehicle under different working conditions is realized by coordinating the rotation angles of wheels of all front axles. In the design and development process of a steering system, the geometric relationship, transmission characteristics and wheel turning angle change rules of a steering mechanism are generally required to be analyzed by means of a multi-body dynamics simulation model, and the steering characteristics are optimized by adjusting key structure point position parameters so as to enable the wheel turning angle relationship to be matched with the steering geometric requirement of a vehicle, thereby improving the overall steering performance of the vehicle. For example, chinese patent CN120337408A discloses a modeling and KC analysis method of a double front axle balanced suspension, wherein the modeling method comprises the steps of determining the structural form of the double front axle balanced suspension, constructing a plate spring model of the double front axle balanced suspension, determining hard point coordinates in the double front axle balanced suspension under the structural form, constructing the hard point coordinates in the plate spring model to obtain a corresponding balanced connecting rod model, and assembling the balanced connecting rod model to obtain an assembly model of the double front axle balanced suspension. The KC analysis method comprises the steps of newly adding a two-bridge reverse run-out working condition in a KC analysis project, and carrying out KC analysis on an assembly model of the double front axle balanced suspension based on the two-bridge reverse run-out working condition to obtain KC performance indexes of the assembly model under the two-bridge reverse run-out working condition. The application is used for overcoming the defects of the modeling and KC analysis modes of the vehicle suspension for the double front axle balanced suspension in the prior art. For example, chinese patent CN114154251B discloses a steering knuckle and a design method of the steering knuckle, and discloses the steering knuckle, which comprises a ring topology structure, a disc body and a front shaft, wherein the ring topology structure and the front shaft are respectively connected to two sides of the disc body, two pin holes are arranged on the ring topology structure and are positioned on the same axis, a curved surface formed by encircling the outer side end part of the ring topology structure is a saddle-shaped hyperboloid, and the two pin holes are arranged at the protruding positions of the saddle-shaped hyperboloid. The invention also discloses a design method of the steering knuckle. The steering knuckle obtained by the invention has reasonable structure, and can avoid the problem of integral failure of parts due to crack expansion caused by the broken root of the stud on the steering knuckle in the prior art. However, the point location optimization with the dual front axle steering system generally compares the wheel corner relationship curve obtained by simulation with the ideal corner relationship curve, and adopts curve point-to-point deviation or unified error index to drive hard point coordinate iterative adjustment. However, in a steering drive chain comprising a plurality of universal joints, the non-constant velocity of the universal joints can introduce periodic angular velocity fluctuation, so that a simulated wheel corner curve overlaps periodic oscillation outside systematic geometric deviation, and when the oscillation and geometric error sources are not distinguished in the optimization process, point-to-point errors can 'equivalently' the periodic oscillation as structural point position deviation, thereby triggering error compensation of hard point coordinates. The misjudgment can lead to that the optimization result is locally similar to a target in a curve, but the precision and stability of the overall steering angle relation are reduced, so that the tire bias wear risk is increased, the consistency of low-speed maneuvering performance is reduced, and the simulation-test alignment period is prolonged and reworked. Therefore, in order to solve the above problems, a method and a system for dua