Wheel hub vertical force acquisition method and device for suspension transient durability working condition

Abstract

This application discloses a method and apparatus for acquiring wheel-center vertical input force under transient durability conditions of suspension. The method includes: constructing a suspension transient input force prediction model and a full-condition error compensation parameter library; acquiring wheel-end vertical acceleration parameters, suspension state parameters, and vehicle condition identification parameters; classifying conditions according to the wheel-end vertical acceleration parameters, suspension state parameters, and vehicle condition identification parameters to obtain the current transient condition; obtaining the current compensation parameters according to the full-condition error compensation parameter library; and obtaining the wheel-center vertical input force corresponding to the current transient condition through the suspension transient input force prediction model, based on the current transient condition, wheel-end vertical acceleration parameters, suspension state parameters, and current compensation parameters. This application can improve the accuracy of wheel-center vertical input force acquisition and can cover all high-frequency transient high-damage conditions, and can be widely used in the field of automotive testing technology.

Description

Technical Field

[0001] This application relates to the field of automotive testing technology, and in particular to a method and device for obtaining wheel center vertical input force for transient durability conditions of suspension. Background Technology

[0002] Current technologies for acquiring vertical input force at the wheel center of air suspension mainly cover three mainstream approaches: direct measurement of wheel center load, multibody dynamics simulation decomposition, and general steady-state calibration based on air spring pressure / height sensors. However, these technologies have drawbacks such as inherent deviations between the acquired input force and the actual force experienced by the vehicle, and the inability to cover low-frequency quasi-steady-state basic operating conditions, making them unsuitable for high-frequency transient and high-damage scenarios. Summary of the Invention

[0003] The main objective of this application is to propose a method and apparatus for obtaining wheel center vertical input force under transient durability conditions of suspension, which can improve the accuracy of wheel center vertical input force acquisition and can cover all high-frequency transient high-damage conditions.

[0004] To achieve the above objectives, one aspect of this application proposes a method for obtaining the wheel center vertical input force under transient durability conditions of suspension, comprising the following steps: Through bench calibration tests, a suspension transient input prediction model and a full-condition error compensation parameter library were constructed. Acquire wheel-end vertical acceleration parameters, suspension status parameters, and vehicle operating condition identification parameters; The current transient operating condition is obtained by classifying the operating conditions based on the wheel-end vertical acceleration parameters, the suspension state parameters, and the vehicle operating condition identification parameters. Based on the full-condition error compensation parameter library, the current compensation parameters corresponding to the current transient condition are obtained; The suspension transient input prediction model obtains the wheel center vertical input corresponding to the current transient condition based on the current transient condition, the wheel end vertical acceleration parameters, the suspension state parameters, and the current compensation parameters.

[0005] In some embodiments, the step of constructing a suspension transient input prediction model and a full-condition error compensation parameter library through bench calibration tests specifically includes: Build a suspension assembly servo test bench; Based on the full impact peak range, full damping gear and full preload range, the target working condition range of the air suspension is divided into several transient working condition points to obtain the working condition index map; For each of the aforementioned transient operating conditions, the suspension assembly servo bench is adjusted to the target operating condition, and then several sets of wheel center vertical excitation forces are input respectively to obtain the corresponding vertical acceleration measurement value and suspension state measurement value. The vertical excitation force at the wheel center, the measured value of vertical acceleration, and the measured value of suspension status for each group are assigned to the transient working condition point corresponding to the working condition index map to obtain a training sample set; The suspension transient input force prediction model is obtained by training the preset working condition index distributed mapping model based on the training sample set. Based on the suspension transient input prediction model and the training sample set, the full-condition error compensation parameter library is constructed.

[0006] In some embodiments, the step of assigning each group of wheel center vertical excitation force, vertical acceleration measurement value, and suspension state measurement value to the transient condition point corresponding to the condition index map to obtain a training sample set specifically includes: Each set of wheel center vertical excitation force, vertical acceleration measurement value, and suspension state measurement value is assigned to the transient working condition point corresponding to the working condition index map to obtain the first sample set; Zero-point drift calibration is performed on the vertical acceleration measurements in the first sample set to obtain the second sample set; Outlier removal is performed on the second sample set, and then the second sample set after outlier removal is filtered and denoised to obtain the third sample set; The third sample set is time-aligned to obtain the training sample set.

[0007] In some embodiments, the working condition index distributed mapping model includes a transient impact encoder network and several local mapping networks. The step of training the preset working condition index distributed mapping model based on the training sample set to obtain the suspension transient input prediction model specifically includes: The transient impulse encoder network is constructed based on a multi-head self-attention mechanism; Construct a corresponding local mapping network for each transient operating point in the operating condition index map; The parameters of the transient shock encoder network are fixed, and each of the local mapping networks is trained based on the training sample set; The parameters of each of the local mapping networks are fixed, and the transient shock encoder network is trained based on the training sample set; When the model performance of the transient impact encoder network and the local mapping network reaches the preset training target, training stops, and the suspension transient input prediction model is obtained.

[0008] In some embodiments, constructing the full-condition error compensation parameter library based on the suspension transient input prediction model and the training sample set specifically includes: For each of the aforementioned transient operating points, the vertical acceleration measurement values ​​and suspension state measurement values ​​of each group are sequentially input into the suspension transient input force prediction model, and the corresponding wheel center vertical input force prediction value is output. Based on the predicted value of the vertical input force at the wheel center and the vertical excitation input force at the wheel center, calculate the residual set corresponding to each transient operating point; Calculate the mean curve of the residual set, and then determine the compensation parameters corresponding to each transient operating point based on the mean curve; The compensation parameters are mapped to the working condition index map to obtain the full working condition error compensation parameter library.

[0009] In some embodiments, obtaining the wheel center vertical input force corresponding to the current transient condition through the suspension transient input force prediction model, based on the current transient condition, the wheel end vertical acceleration parameters, the suspension state parameters, and the current compensation parameters, specifically includes: The wheel-end vertical acceleration parameters and the suspension state parameters are input into the transient impact encoder network in the suspension transient input prediction model, and the working condition global feature vector is output. Determine the local mapping network in the suspension transient input force prediction model that corresponds to the current transient condition, and then input the global feature vector of the condition into the local mapping network to output the original wheel center vertical input force; The original wheel center vertical input force is corrected based on the current compensation parameters to obtain the wheel center vertical input force.

[0010] In some embodiments, the method further includes: Statistical analysis was performed on the vertical input force at the wheel center to obtain the load spectrum under different transient conditions; The load spectrum was used as the vertical load input for bench testing to perform transient durability verification of the air suspension under the corresponding working conditions.

[0011] To achieve the above objectives, another aspect of this application proposes a wheel center vertical input force acquisition device for suspension transient durability conditions, comprising: The model and parameter library construction module is used to build a suspension transient input prediction model and a full-condition error compensation parameter library through bench calibration tests. The data acquisition module is used to acquire wheel-end vertical acceleration parameters, suspension status parameters, and vehicle operating condition identification parameters; The working condition identification module is used to classify the working conditions based on the wheel end vertical acceleration parameters, the suspension state parameters, and the vehicle working condition identification parameters to obtain the current transient working condition. The compensation parameter determination module is used to obtain the current compensation parameters corresponding to the current transient condition based on the full-condition error compensation parameter library. The input force acquisition module is used to obtain the wheel center vertical input force corresponding to the current transient condition through the suspension transient input force prediction model, based on the current transient condition, the wheel end vertical acceleration parameters, the suspension state parameters, and the current compensation parameters.

[0012] To achieve the above objectives, another aspect of this application provides an electronic device, which includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the method described above.

[0013] To achieve the above objectives, another aspect of the embodiments of this application proposes a computer program product, including a computer program that, when executed by a processor, implements the aforementioned method.

[0014] The embodiments of this application include at least the following beneficial effects: The method and apparatus for obtaining wheel center vertical input force for suspension transient durability conditions of this application first constructs a suspension transient input force prediction model and a full-condition error compensation parameter library through bench calibration tests; then, it obtains wheel end vertical acceleration parameters, suspension state parameters, and vehicle condition identification parameters; next, it classifies the conditions according to the wheel end vertical acceleration parameters, suspension state parameters, and vehicle condition identification parameters to obtain the current transient condition; then, it obtains the current compensation parameters corresponding to the current transient condition according to the full-condition error compensation parameter library; finally, it obtains the wheel center vertical input force corresponding to the current transient condition through the suspension transient input force prediction model, based on the current transient condition, wheel end vertical acceleration parameters, suspension state parameters, and current compensation parameters. This application trains a suspension transient input prediction model that covers all high-frequency transient and high-damage conditions. Using the vertical acceleration data collected by the wheel-end accelerometers that are standard on mass-produced vehicles as the core input, it accurately inversely calculates the wheel-center vertical input under high-frequency transient conditions. Furthermore, it combines a full-condition error compensation parameter library calibrated on the bench to perform error compensation, thereby improving the accuracy of wheel-center vertical input acquisition and covering all high-frequency transient and high-damage conditions. Attached Figure Description

[0015] To more clearly illustrate the technical solutions in the embodiments of this application, the drawings used in the embodiments of this application are described below. It should be understood that the drawings described below are only for the purpose of clearly illustrating some embodiments of the technical solutions in this application. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0016] Figure 1 A flowchart illustrating the steps of a method for obtaining wheel center vertical input force under transient durability conditions of suspension provided in one embodiment of this application; Figure 2This is a schematic diagram of the processing flow of a suspension transient input force prediction model provided in one embodiment of this application; Figure 3 This is a schematic diagram of the structure of a wheel center vertical input force acquisition device for suspension transient durability conditions provided in one embodiment of this application; Figure 4 This is a schematic diagram of the hardware structure of an electronic device provided in one embodiment of this application. Detailed Implementation

[0017] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of this application and are not intended to limit it. In the following description, when referring to the accompanying drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with those of this application; they are merely examples of apparatuses and methods consistent with some aspects of the embodiments of this application as detailed in the appended claims.

[0018] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing embodiments of this application only and is not intended to limit this application.

[0019] Current technologies for acquiring vertical input force at the wheel center of air suspension systems mainly cover three mainstream approaches: direct measurement of wheel center load, multibody dynamics simulation decomposition, and general steady-state calibration based on air spring pressure / height sensors. Among these, the general steady-state calibration approach can only estimate the basic load under low-frequency quasi-steady-state conditions and has become the industry-standard basis for mass production applications. The above-mentioned technical solutions have the following drawbacks: First, the direct measurement scheme for wheel center load has high hardware costs, requires special modifications to the test vehicle, and can only be carried out in limited scenarios at the test site. It cannot realize the full-scenario, large-scale input force data collection for mass-produced civilian vehicles. Second, the simulation load decomposition scheme relies on an idealized model, which cannot reproduce the nonlinear characteristics in actual vehicle use. The obtained input force has an inherent deviation from the actual force on the actual vehicle, which can easily lead to the performance disconnect between bench tests and actual vehicle use. Third, the general steady-state calibration scheme only covers low-frequency quasi-steady-state basic working conditions and cannot adapt to high-frequency transient high-damage scenarios. The signal is severely distorted under high-frequency impact, and this type of working condition is the main cause of fatigue failure of the core components of the suspension. Fourth, the standard wheel-end vertical acceleration sensor in mass production is only used for road surface recognition and damping adjustment. There is no mass-producible high-precision vertical input reverse thrust solution for suspension durability verification, and the industry cannot achieve complete coverage of durability loads under all working conditions.

[0020] In view of this, this application proposes a method for obtaining wheel center vertical input force for suspension transient durability conditions. First, a suspension transient input force prediction model and a full-condition error compensation parameter library are constructed through bench calibration tests. Next, wheel-end vertical acceleration parameters, suspension state parameters, and vehicle condition identification parameters are obtained. Then, the operating conditions are classified according to the wheel-end vertical acceleration parameters, suspension state parameters, and vehicle condition identification parameters to obtain the current transient operating condition. Subsequently, the current compensation parameters corresponding to the current transient operating condition are obtained based on the full-condition error compensation parameter library. Finally, the wheel center vertical input force corresponding to the current transient operating condition is obtained using the suspension transient input force prediction model, based on the current transient operating condition, wheel-end vertical acceleration parameters, suspension state parameters, and current compensation parameters. This application trains a suspension transient input prediction model that covers all high-frequency transient and high-damage conditions. Using the vertical acceleration data collected by the wheel-end accelerometers that are standard on mass-produced vehicles as the core input, it accurately inversely calculates the wheel-center vertical input under high-frequency transient conditions. Furthermore, it combines a full-condition error compensation parameter library calibrated on the bench to perform error compensation, thereby improving the accuracy of wheel-center vertical input acquisition and covering all high-frequency transient and high-damage conditions.

[0021] Reference Figure 1 , Figure 1 This is a flowchart illustrating the steps of a method for obtaining wheel center vertical input force under transient durability conditions for suspension, according to an embodiment of this application. This application proposes a method for obtaining wheel center vertical input force under transient durability conditions for suspension, which may include, but is not limited to, the following steps S101 to S105: Step S101: Construct a suspension transient input prediction model and a full-condition error compensation parameter library through bench calibration tests; Specifically, the suspension transient input prediction model is used to establish the mapping relationship between wheel-end vertical acceleration parameters, suspension state parameters and wheel-center vertical input; the full-condition error compensation parameter library is used to correct model deviations under high-frequency transient conditions such as different impact boundaries, damping gears, and preloads.

[0022] As an optional implementation, step S101 can be further divided into the following steps S1011 to S1016: Step S1011: Build the suspension assembly servo bench; Step S1012: Based on the full impact peak range, full damping gear and full preload range, the target working condition range of the air suspension is divided into several transient working condition points to obtain the working condition index map. Specifically, a suspension assembly servo bench was constructed to replicate the topology and hardpoint coordinates of the actual vehicle's suspension at a 1:1 scale. Wheel-end vertical accelerometers and suspension height sensors were fixed in mounting positions identical to those on the production vehicle, and a force sensor was installed at the wheel center as the true reference for the vertical excitation force. Then, based on the target air suspension system's full impact peak range, full damping settings, and full preload range, the air suspension's full high-frequency transient operating range was divided into several groups of transient operating points. Using the full damping settings and full preload range as the first-level index key, and the full impact peak range as the second-level index key, a multi-level operating point index map was constructed. Each terminal leaf node of this map corresponds to a unique transient operating point.

[0023] Among them, the full impact peak range refers to the range of vertical relative velocity at the wheel center when subjected to transient impact, covering the range from smooth driving to extreme impact; the full damping range refers to all control states of the adjustable damping shock absorber from the softest to the hardest; the full preload range refers to all initial force positions of the suspension within the design travel range, from zero load to full load, and from tensile limit to compression limit.

[0024] For example, taking a certain transient operating point as an example, its complete coordinates in the operating condition index map can be represented as: [Preload range: 0mm (design position), damping level: 1.0A (motion level), impact peak: 1.8m / s (high-speed impact)].

[0025] Step S1013: For each transient working condition point, adjust the suspension assembly servo test bench to the target working condition, and then input several sets of wheel center vertical excitation forces respectively to obtain the corresponding vertical acceleration measurement value and suspension state measurement value. Step S1014: Assign the measured values ​​of vertical excitation force, vertical acceleration and suspension status of each wheel center to the corresponding transient working point in the working condition index map to obtain the training sample set; Specifically, for each transient operating point in the operating condition index diagram, the preload position of the suspension assembly servo test bench and the damper damping current are adjusted to achieve the target preload range and target damping level defined for that transient operating point. A known standard wheel center vertical excitation force F is input to the wheel center via the wheel center-end servo actuator. w Simultaneously, the wheel-end vertical acceleration measurement value 'a' at the air spring end and suspension state parameters (including suspension relative displacement 's' and shock absorber damping gear parameter 'G') are collected. The above excitation and acquisition process is repeated multiple times for this transient operating point to obtain multiple sets of calibration data. Finally, based on the collected calibration data, a training sample set is constructed to train the suspension transient input prediction model.

[0026] As an optional implementation, step S1014 can be further divided into the following steps S10141 to S10144: Step S10141: Assign the measured values ​​of vertical excitation force, vertical acceleration and suspension status of each group of wheel center to the transient working point corresponding to the working condition index map to obtain the first sample set; Specifically, each complete transient time-domain sequence containing the wheel-center vertical excitation input, vertical acceleration measurement, and suspension state measurement is treated as an independent data sample and assigned to the terminal leaf node of the transient operating condition point to which the sample actually belongs in the operating condition index map. After the assignment is completed, multiple sets of data samples for that transient operating condition point are gathered under each terminal node, and the collection of all node data samples constitutes a tree-structured first sample set organized by operating condition.

[0027] Step S10142: Perform zero-point drift calibration on the vertical acceleration measurements in the first sample set to obtain the second sample set; Specifically, by using the zero-point offset of the vertical acceleration measurement value extracted from the preset standard no-impact working condition (such as the static preload holding section before the start of the bench test), the zero-point correction is performed on all vertical acceleration measurement values ​​under this working condition node to obtain the data set after eliminating sensor bias error, i.e., the second sample set.

[0028] Step S10143: Remove outliers from the second sample set, and then filter and denoise the second sample set after removing outliers to obtain the third sample set; Step S10144: Perform temporal alignment on the third sample set to obtain the training sample set.

[0029] Specifically, the physical rationality of each group of data in the second sample set is checked, and values ​​that exceed the sensor range are removed. Then, a low-pass filter is used to filter out the noise. Finally, the vertical excitation input force at each wheel center is time-aligned with the measured values ​​of vertical acceleration at the wheel end and the suspension state parameters to correct the signal transmission or response delay between different sensors, thus obtaining a training sample set for training the suspension transient input force prediction model.

[0030] Step S1015: Train the preset working condition index distributed mapping model according to the training sample set to obtain the suspension transient input prediction model. It should be noted that, through bench calibration tests, this application constructs a working condition index map covering the entire peak impact range, all damping gears, and the entire preload range. Then, based on the training sample set under the working condition index map, a suspension transient input prediction model covering all high-frequency transient high-damage working conditions is trained, which can accurately predict the wheel center vertical input under different high-frequency transient working conditions.

[0031] As an optional implementation, the working condition index distributed mapping model includes a transient impact encoder network and several local mapping networks. Step S1015 can be further divided into the following steps S10151 to S10155: Step S10151: Construct a transient impulse encoder network based on a multi-head self-attention mechanism; Step S10152: Construct a corresponding local mapping network for each transient working condition point in the working condition index map; Specifically, the model constructed in this application adopts a hybrid architecture of a shared feature extraction layer and a condition-specific output layer. First, a transient impact encoder network based on a multi-head self-attention mechanism is constructed, and this transient impact encoder network is used as the shared feature extraction layer. Using wheel-end vertical acceleration and suspension state parameters as inputs, the dynamic dependencies between signals at different time steps within an impact event are automatically learned through the self-attention mechanism, outputting the corresponding global feature vector for that condition. Following the constructed transient impact encoder network, a local mapping network with the same structure but independent parameters is initialized for each terminal condition node in the condition index map. Each local mapping network takes the global feature vector for that condition as input and outputs the estimated value of the wheel-center vertical input force under that condition.

[0032] Step S10153: Fix the parameters of the transient shock encoder network and train each local mapping network according to the training sample set; Step S10154: Fix the parameters of each local mapping network and train the transient shock encoder network according to the training sample set; Step S10155: When the model performance of the transient impact encoder network and the local mapping network reaches the preset training target, stop training and obtain the suspension transient input prediction model.

[0033] Specifically, the model is trained iteratively using the training dataset obtained from the aforementioned calibration experiments. First, the parameters of the transient impact encoder network are frozen, and the parameters of each local mapping network are independently trained and updated using the sample data of each node. Then, the parameters of all local mapping networks are frozen, and the data of all nodes are mixed and input. The parameters of the transient impact encoder network are fine-tuned with the goal of minimizing the joint prediction error of all working conditions. The above training steps are executed alternately until the performance of the working condition index distributed mapping model composed of the transient impact encoder network and each local mapping network reaches the corresponding training objective (such as the loss function value converging to the stable interval and the number of alternating iterations reaching the preset number of training times). Training is then stopped, and the trained suspension transient input calibration inverse model is obtained.

[0034] It should be noted that the embodiments of this application adopt a hybrid architecture of shared feature extraction layer and working condition specific output layer, and train the transient impact encoder network and local mapping network based on alternating iterative training strategy, which can improve the training efficiency and training accuracy of the model, thereby improving the prediction accuracy of the vertical input force of the wheel center under different transient working conditions.

[0035] Step S1016: Construct a full-condition error compensation parameter library based on the suspension transient input prediction model and training sample set.

[0036] It should be noted that the embodiments of this application estimate the input force based on real vehicle operating data and establish a full-condition error compensation parameter library based on the working condition index map. Error compensation is performed on the wheel center vertical input force under different transient working conditions, which can improve the accuracy of wheel center vertical input force acquisition and solve the inherent deviation problem between the simulation scheme and the actual vehicle working conditions.

[0037] As an optional implementation, step S1016 can be further divided into steps S10161 to S10164: Step S10161: For each transient operating point, input the vertical acceleration measurement value and suspension state measurement value of each group into the suspension transient input force prediction model in sequence, and output the corresponding wheel center vertical input force prediction value. Step S10162: Calculate the residual set corresponding to each transient working point based on the predicted value of the vertical input force at the wheel center and the vertical excitation input force at the wheel center; Step S10163: Calculate the mean curve of the residual set, and then determine the compensation parameters corresponding to each transient operating point based on the mean curve. Step S10164: Map the compensation parameters to the working condition index map to obtain the full working condition error compensation parameter library.

[0038] Specifically, the aforementioned steps, for each transient operating point under the operating condition index map, adjust the test bench to the target operating condition, input the known wheel-center vertical excitation force through the wheel-center servo actuator, and simultaneously collect wheel-center vertical acceleration measurements and suspension state measurements. Here, for each transient operating point, the collected sets of vertical acceleration measurements and suspension state measurements are input point by point into the trained suspension transient input prediction model to obtain multiple sets of wheel-center vertical input prediction values.

[0039] For any transient operating point in the operating condition index map It has a total of Group of training samples. For the first training sample under this transient operating condition. The residual is obtained by calculating the difference between the predicted vertical excitation force at the wheel center and the actual vertical excitation force at the wheel center using the following formula: ; in, Indicates the first Under the transient operating condition point, the first Group training samples at time The residual, Indicates the first Under the transient operating condition point, the first Group training samples at time The vertical excitation force input to the wheel center, Indicates the first Under the transient operating condition point, the first Group training samples at time Predicted value of vertical input force at the wheel center. .

[0040] The first All under each transient operating condition point The residuals of the training samples are summarized to form the residual set for that transient operating condition. : ; For the first Calculate the residual set for each transient operating point. mean curve : ; Then, the mean curve is obtained through calculation. Design corresponding compensation parameters for each transient operating point. (e.g., directly retain the complete mean curve) As a time-varying compensation curve, the model output is subsequently corrected point-by-point according to time steps. Finally, the compensation parameters corresponding to each transient condition point are... The parameters are stored in the terminal leaf node corresponding to the transient operating condition point in the operating condition index map to obtain the full operating condition error compensation parameter library.

[0041] Step S102: Obtain wheel-end vertical acceleration parameters, suspension state parameters, and vehicle operating condition identification parameters; In some optional embodiments, wheel-end vertical acceleration parameters are collected based on wheel-end vertical accelerometers standard on mass-produced vehicles; suspension status parameters are collected based on suspension height sensors standard on mass-produced vehicles. The suspension status parameters include relative suspension displacement and shock absorber damping gear parameters; vehicle operating condition identification parameters include at least one or a combination of vehicle speed, vehicle longitudinal acceleration, vehicle lateral acceleration, braking signal, drive torque signal, road gradient signal, and steering wheel angle signal.

[0042] It should be noted that the embodiments of this application reuse the standard sensors in mass-produced vehicles, which can reduce hardware costs and vehicle modification requirements, and enable large-scale acquisition of input forces in all scenarios of mass-produced vehicles.

[0043] Step S103: Classify the operating conditions based on the wheel-end vertical acceleration parameters, suspension state parameters, and vehicle operating condition identification parameters to obtain the current transient operating condition; In some optional embodiments, the relative suspension displacement in the suspension state parameters is filtered by moving average to obtain the current static preload level, which is then matched to the closest preload interval node in the operating condition index map. Simultaneously, the shock absorber damping gear parameter in the suspension state parameters is matched to the damping gear node in the operating condition index map, completing the location of the first-level index key in the operating condition index map. Next, using the vehicle lateral acceleration, steering wheel angle rate, vehicle longitudinal acceleration, braking signal, drive torque, and road slope signal in the vehicle operating condition identification parameters, gating judgment is performed on the vehicle's current overall driving state. Furthermore, using the wheel-end vertical acceleration parameter and vehicle speed signal, feature recognition is performed on the currently occurring transient impact event. Then, event detection and segmentation are performed on the wheel-end vertical acceleration signal, extracting the peak speed and main frequency of the impact event. The extracted impact features are matched with each second-level index node under the current first-level index node to find the closest impact peak, completing the location of the second-level index key. Finally, the current transient operating condition corresponding to the vehicle is determined, completing the current transient operating condition classification.

[0044] Step S104: Obtain the current compensation parameters corresponding to the current transient condition based on the full-condition error compensation parameter library; Specifically, after completing the working condition classification, the current transient working condition index is obtained. Using this index, the previously constructed full-working-condition error compensation parameter library is queried, and the current compensation parameter corresponding to the current transient working condition is obtained by looking up the table. .

[0045] Step S105: Using the suspension transient input prediction model, based on the current transient operating condition, wheel-end vertical acceleration parameters, suspension state parameters, and current compensation parameters, obtain the wheel-center vertical input corresponding to the current transient operating condition.

[0046] As an optional implementation, step S105 can be further divided into the following steps S1051 to S1053: Step S1051: Input the wheel-end vertical acceleration parameters and suspension state parameters into the transient impact encoder network in the suspension transient input prediction model, and output the global feature vector of the working condition; Step S1052: Determine the local mapping network in the suspension transient input force prediction model that corresponds to the current transient working condition, and then input the global feature vector of the working condition into the local mapping network to output the original wheel center vertical input force; Step S1053: Correct the original wheel center vertical input force according to the current compensation parameters to obtain the wheel center vertical input force.

[0047] Specifically, in real-vehicle applications, the current transient condition is identified based on real-time collected parameters. This identified transient condition serves as the index key, and the corresponding local mapping network is invoked within the suspension transient input force prediction model. For example... Figure 2 The diagram illustrates the processing flow of the suspension transient input force prediction model. The collected wheel-end vertical acceleration and suspension state parameters are input into the transient impact encoder network to extract the global feature vector of the operating condition. This feature vector is then fed into the invoked local mapping network to calculate and output the original wheel-center vertical input force at the current moment. Finally, the current compensation parameters obtained above are used... The original wheel center vertical input force output by the model Make corrections, that is The vertical input force at the wheel center under this transient working condition is obtained. .

[0048] As a further optional implementation, the method for obtaining the wheel center vertical input force for transient durability conditions of the suspension may also include the following steps S106 to S107: Step S106: Perform statistical analysis on the vertical input force at the wheel center to obtain the load spectrum under different transient conditions; Step S107: Use the load spectrum as the vertical load input for the bench test to perform transient durability verification of the air suspension under the corresponding working conditions.

[0049] Specifically, the obtained vertical input force at the wheel center is statistically analyzed to generate a load spectrum for the corresponding transient conditions. Simultaneously, large-scale statistical analysis of load spectra collected from multiple vehicles can be performed to obtain a typical transient load spectrum for the entire user's lifecycle. The generated load spectrum is then imported into the air suspension bench transient durability test system as the vertical load input for the bench test, enabling durability verification under the corresponding transient conditions.

[0050] The above describes the method for obtaining the wheel center vertical input force for suspension transient durability conditions according to embodiments of this application. It can be recognized that, compared with existing methods for obtaining wheel center vertical input force, embodiments of this application have the following advantages: 1. Through bench calibration tests, a working condition index map covering the entire impact peak range, all damping gears, and the entire preload range is constructed. Then, based on the training sample set under the working condition index map, a suspension transient input prediction model covering the entire high-frequency transient high-damage working condition is trained, which can accurately predict the wheel center vertical input under different high-frequency transient working conditions.

[0051] Second, by adopting a hybrid architecture of shared feature extraction layer and condition-specific output layer, and by training the transient impact encoder network and local mapping network based on an alternating iterative training strategy, the training efficiency and accuracy of the model can be improved, thereby enhancing the prediction accuracy of the vertical input force of the wheel center under different transient conditions.

[0052] Third, the input force is estimated based on the actual operating data of the real vehicle, and a full-condition error compensation parameter library is established based on the working condition index map. Error compensation is performed on the wheel center vertical input force under different transient working conditions, which can improve the accuracy of wheel center vertical input force acquisition and solve the inherent deviation problem between the simulation scheme and the actual vehicle working conditions.

[0053] Fourth, reusing standard sensors in mass-produced vehicles can reduce hardware costs and vehicle modification requirements, and enable large-scale force acquisition in all scenarios of mass-produced vehicles.

[0054] Fifth, the acquired input force can be directly adapted to the industry-standard air suspension test bench system without modifying existing test bench equipment, and can be directly imported into the application, demonstrating strong compatibility.

[0055] Reference Figure 3 This application embodiment also provides a wheel center vertical input force acquisition device for suspension transient durability conditions, including: The model and parameter library construction module is used to build a suspension transient input prediction model and a full-condition error compensation parameter library through bench calibration tests. The data acquisition module is used to acquire wheel-end vertical acceleration parameters, suspension status parameters, and vehicle operating condition identification parameters; The working condition identification module is used to classify working conditions based on wheel-end vertical acceleration parameters, suspension state parameters, and vehicle working condition identification parameters to obtain the current transient working condition. The compensation parameter determination module is used to obtain the current compensation parameters corresponding to the current transient condition based on the full-condition error compensation parameter library. The input force acquisition module is used to obtain the wheel center vertical input force corresponding to the current transient condition by using the suspension transient input force prediction model, based on the current transient condition, wheel end vertical acceleration parameters, suspension state parameters, and current compensation parameters.

[0056] It is understood that the content of the above method embodiments is applicable to the present device embodiments. The specific functions implemented by the present device embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0057] This application also provides an electronic device, which includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the above-described method. This electronic device can be any smart terminal, including tablet computers, in-vehicle computers, etc.

[0058] It is understood that the content of the above method embodiments is applicable to this device embodiment. The specific functions implemented by this device embodiment are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0059] Please see Figure 4 , Figure 4 The hardware structure of an electronic device according to another embodiment is illustrated. The electronic device includes: The processor 1001 can be implemented using a general-purpose CPU (Central Processing Unit), microprocessor, application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of this application. The memory 1002 can be implemented as a read-only memory (ROM), static storage device, dynamic storage device, or random access memory (RAM). The memory 1002 can store the operating system and other applications. When the technical solutions provided in the embodiments of this specification are implemented through software or firmware, the relevant program code is stored in the memory 1002 and is called and executed by the processor 1001 using the methods described in the embodiments of this application. Input / output interface 1003 is used to implement information input and output; The communication interface 1004 is used to enable communication and interaction between this device and other devices. Communication can be achieved through wired means (such as USB, network cable, etc.) or wireless means (such as mobile network, WIFI, Bluetooth, etc.). Bus 1005 transmits information between various components of the device (e.g., processor 1001, memory 1002, input / output interface 1003, and communication interface 1004); The processor 1001, memory 1002, input / output interface 1003 and communication interface 1004 are connected to each other within the device via bus 1005.

[0060] This application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described method.

[0061] It is understood that the content of the above method embodiments is applicable to this storage medium embodiment. The specific functions implemented in this storage medium embodiment are the same as those in the above method embodiments, and the beneficial effects achieved are also the same as those achieved in the above method embodiments.

[0062] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the above-described method.

[0063] It is understood that the content of the above method embodiments is applicable to the embodiments of this program product. The specific functions implemented by the embodiments of this program product are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0064] Memory, as a non-transitory computer-readable storage medium, can be used to store non-transitory software programs and non-transitory computer-executable programs. Furthermore, memory may include high-speed random access memory, and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some embodiments, memory may optionally include memory remotely located relative to the processor, and these remote memories can be connected to the processor via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

[0065] The embodiments described in this application are for the purpose of more clearly illustrating the technical solutions of the embodiments of this application, and do not constitute a limitation on the technical solutions provided by the embodiments of this application. As those skilled in the art will know, with the evolution of technology and the emergence of new application scenarios, the technical solutions provided by the embodiments of this application are also applicable to similar technical problems.

[0066] Those skilled in the art will understand that the technical solutions shown in the figures do not constitute a limitation on the embodiments of this application, and may include more or fewer steps than shown, or combine certain steps, or different steps.

[0067] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0068] Those skilled in the art will understand that all or some of the steps in the methods disclosed above, as well as the functional modules / units in the systems and devices, can be implemented as software, firmware, hardware, or suitable combinations thereof.

[0069] The terms “first,” “second,” “third,” “fourth,” etc. (if present) in the specification and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms “comprising” and “having,” and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0070] It should be understood that in this application, "at least one (item)" means one or more, and "more than" means two or more. "And / or" is used to describe the relationship between related objects, indicating that three relationships can exist. For example, "A and / or B" can represent three cases: only A exists, only B exists, and both A and B exist simultaneously, where A and B can be singular or plural. The character " / " generally indicates that the preceding and following related objects are in an "or" relationship. "At least one (item) of the following" or similar expressions refer to any combination of these items, including any combination of single or plural items. For example, at least one (item) of a, b, or c can represent: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, and c can be single or multiple.

[0071] In the several embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of the units described above is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0072] The units described above as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0073] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0074] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes multiple instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this application. The aforementioned storage medium includes various media capable of storing programs, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0075] The preferred embodiments of the present application have been described above with reference to the accompanying drawings, but this does not limit the scope of the claims of the present application. Any modifications, equivalent substitutions, and improvements made by those skilled in the art without departing from the scope and substance of the embodiments of the present application shall be within the scope of the claims of the present application.

Claims

1. A method for obtaining the wheel center vertical input force under transient durability conditions of suspension, characterized in that, Includes the following steps: Through bench calibration tests, a suspension transient input prediction model and a full-condition error compensation parameter library were constructed. Acquire wheel-end vertical acceleration parameters, suspension status parameters, and vehicle operating condition identification parameters; The current transient operating condition is obtained by classifying the operating conditions based on the wheel-end vertical acceleration parameters, the suspension state parameters, and the vehicle operating condition identification parameters. Based on the full-condition error compensation parameter library, the current compensation parameters corresponding to the current transient condition are obtained; The suspension transient input prediction model obtains the wheel center vertical input corresponding to the current transient condition based on the current transient condition, the wheel end vertical acceleration parameters, the suspension state parameters, and the current compensation parameters.

2. The method according to claim 1, characterized in that, The process of constructing a suspension transient input prediction model and a full-condition error compensation parameter library through bench calibration tests specifically includes: Build a suspension assembly servo test bench; Based on the full impact peak range, full damping gear and full preload range, the target working condition range of the air suspension is divided into several transient working condition points to obtain the working condition index map; For each of the aforementioned transient operating conditions, the suspension assembly servo bench is adjusted to the target operating condition, and then several sets of wheel center vertical excitation forces are input respectively to obtain the corresponding vertical acceleration measurement value and suspension state measurement value. The vertical excitation force at the wheel center, the measured value of vertical acceleration, and the measured value of suspension status for each group are assigned to the transient working condition point corresponding to the working condition index map to obtain a training sample set; The suspension transient input force prediction model is obtained by training the preset working condition index distributed mapping model based on the training sample set. Based on the suspension transient input prediction model and the training sample set, the full-condition error compensation parameter library is constructed.

3. The method according to claim 2, characterized in that, The step of assigning each group of wheel center vertical excitation force, vertical acceleration measurement value, and suspension state measurement value to the transient working condition point corresponding to the working condition index map to obtain a training sample set specifically includes: Each set of wheel center vertical excitation force, vertical acceleration measurement value, and suspension state measurement value is assigned to the transient working condition point corresponding to the working condition index map to obtain the first sample set; Zero-point drift calibration is performed on the vertical acceleration measurements in the first sample set to obtain the second sample set; Outlier removal is performed on the second sample set, and then the second sample set after outlier removal is filtered and denoised to obtain the third sample set; The third sample set is time-aligned to obtain the training sample set.

4. The method according to claim 2, characterized in that, The working condition index distributed mapping model includes a transient impact encoder network and several local mapping networks. The step of training the preset working condition index distributed mapping model based on the training sample set to obtain the suspension transient input prediction model specifically includes: The transient impulse encoder network is constructed based on a multi-head self-attention mechanism; Construct a corresponding local mapping network for each transient operating point in the operating condition index map; The parameters of the transient shock encoder network are fixed, and each of the local mapping networks is trained based on the training sample set; The parameters of each of the local mapping networks are fixed, and the transient shock encoder network is trained based on the training sample set; When the model performance of the transient impact encoder network and the local mapping network reaches the preset training target, training stops, and the suspension transient input prediction model is obtained.

5. The method according to claim 2, characterized in that, The step of constructing the full-condition error compensation parameter library based on the suspension transient input prediction model and the training sample set specifically includes: For each of the aforementioned transient operating points, the vertical acceleration measurement values ​​and suspension state measurement values ​​of each group are sequentially input into the suspension transient input force prediction model, and the corresponding wheel center vertical input force prediction value is output. Based on the predicted value of the vertical input force at the wheel center and the vertical excitation input force at the wheel center, calculate the residual set corresponding to each transient operating point; Calculate the mean curve of the residual set, and then determine the compensation parameters corresponding to each transient operating point based on the mean curve; The compensation parameters are mapped to the working condition index map to obtain the full working condition error compensation parameter library.

6. The method according to claim 1, characterized in that, The step of obtaining the wheel center vertical input force corresponding to the current transient condition through the suspension transient input force prediction model, based on the current transient condition, the wheel end vertical acceleration parameters, the suspension state parameters, and the current compensation parameters, specifically includes: The wheel-end vertical acceleration parameters and the suspension state parameters are input into the transient impact encoder network in the suspension transient input prediction model, and the working condition global feature vector is output. Determine the local mapping network in the suspension transient input force prediction model that corresponds to the current transient condition, and then input the global feature vector of the condition into the local mapping network to output the original wheel center vertical input force; The original wheel center vertical input force is corrected based on the current compensation parameters to obtain the wheel center vertical input force.

7. The method according to claim 1, characterized in that, The method further includes: Statistical analysis was performed on the vertical input force at the wheel center to obtain the load spectrum under different transient conditions; The load spectrum was used as the vertical load input for bench testing to perform transient durability verification of the air suspension under the corresponding working conditions.

8. A device for acquiring vertical input force at the wheel center for transient durability conditions of suspension, characterized in that, include: The model and parameter library construction module is used to build a suspension transient input prediction model and a full-condition error compensation parameter library through bench calibration tests. The data acquisition module is used to acquire wheel-end vertical acceleration parameters, suspension status parameters, and vehicle operating condition identification parameters; The working condition identification module is used to classify the working conditions based on the wheel end vertical acceleration parameters, the suspension state parameters, and the vehicle working condition identification parameters to obtain the current transient working condition. The compensation parameter determination module is used to obtain the current compensation parameters corresponding to the current transient condition based on the full-condition error compensation parameter library. The input force acquisition module is used to obtain the wheel center vertical input force corresponding to the current transient condition through the suspension transient input force prediction model, based on the current transient condition, the wheel end vertical acceleration parameters, the suspension state parameters, and the current compensation parameters.

9. An electronic device, characterized in that, The electronic device includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the method of any one of claims 1 to 7.

10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the method of any one of claims 1 to 7.

Images