Method for testing mechanical durability of whole vehicle and parts based on fuel cell characteristics

Abstract

The invention provides a method for testing mechanical durability of a whole vehicle and parts based on fuel cell characteristics, and relates to the technical field of traffic transportation. Aiming at the problem of poor reliability of a test method in the prior art, test channels are arranged at a plurality of key nodes of a target vehicle type; acquiring an actually measured durability load under a target working condition, and deducing a full-life-cycle complete durability load to determine target false damage of each test channel; collecting test durability load of a test field, and calculating test pseudo damage of each test channel unit according to the test durability load; establishing an equal damage model according to the target false damage and the unit test false damage, determining the cycle index of each road condition in the test field, performing test on the test field according to the cycle index, and determining a final test durability load; and according to the final test durability load, carrying out multi-physics coupling environment bench test to obtain a verification durability load, and verifying the final test durability load. The test method provided by the invention has relatively high reliability.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of transportation, in particular to a fuel cell characteristic vehicle and component mechanical durability test method. BACKGROUND

[0002] With the fuel cell vehicle industry entering the key stage from import dependence to independent technology breakthrough, the demonstration operation scale continues to expand, and the upstream and downstream industry chains are maturing. However, the mechanical durability reliability of the whole vehicle and core components is still the core bottleneck restricting commercialization, and the performance degradation law directly determines the product competitiveness and user confidence.

[0003] At present, the reliability research of fuel cell vehicles at home and abroad is mostly concentrated on the stack system level, and various component-level test standards have been formed around material aging, power attenuation and cycle life, but these specifications focus on single working condition or independent vibration test of the stack, lack of system research on the coupling of road excitation and internal load under the service environment of the whole vehicle, and it is difficult to fully reveal the durability degradation mechanism in actual use, which leads to the difficulty of reflecting the product reliability under real working conditions.

[0004] Therefore, developing a fuel cell characteristic vehicle and component mechanical durability test method has important significance for improving the reliability of test results. SUMMARY

[0005] In order to solve the problem of poor reliability of the test method in the prior art, the present application proposes a fuel cell characteristic vehicle and component mechanical durability test method, which specifically includes the following steps: S1, arranging test channels at a plurality of key nodes of a target vehicle, the key nodes including suspension nodes of a fuel cell stack, an air compressor and a vehicle-mounted hydrogen system, and vehicle structure nodes; S2, arranging corresponding acquisition devices at the test channels, acquiring measured durability loads under target working conditions through the acquisition devices, and extending and deducing complete durability loads of the target vehicle for the whole life cycle according to the measured durability loads; S3, determining target pseudo-damages of each test channel according to the complete durability loads; S4, collecting test durability loads of test field road sections, and calculating unit test pseudo-damages of each test channel according to the test durability loads; S5, establishing an equal damage model according to the target pseudo-damages and the unit test pseudo-damages, and determining cycle numbers of each road condition in the test field according to the equal damage model; S6, performing test field test according to the determined cycle numbers, and determining final test durability loads; S7, verifying the final test durability load by bench test verification, the bench test verification comprising simulating a multi-physical field coupling environment to obtain a verification durability load, so as to verify the final test durability load.

[0006] Further, corresponding acquisition devices are arranged at the test channels, and the measured durability load under the target working condition is obtained through the acquisition devices, comprising: comprehensively surveying the route, paving conditions and safety of the test site, determining the road combination covering the target working condition and arranging the test sequence; arranging acceleration sensors, strain sensors and temperature and humidity and pressure sensors at each test channel, calibrating in turn according to the channels, and then setting the fixed sampling frequency, range and trigger condition; starting the vehicle after checking abnormal data by test running verification, synchronously collecting the acceleration, strain, temperature and humidity and pressure signals and vehicle positioning data of each test channel, and after the collection is completed, formatting the data, correcting the time label and generating the metadata.

[0007] Further, according to the measured durability load, the complete durability load of the target vehicle model in the whole life cycle is derived, comprising: processing the measured durability load by using the rain flow cycle counting method to obtain the amplitude and mean value two-dimensional matrix information of the load cycle; based on the two-dimensional diffusion kernel density estimation method, the two-dimensional matrix information is extended, and the complete durability load of the target vehicle model in the whole life cycle is derived.

[0008] Further, according to the complete durability load, the target pseudo-damage of each test channel is determined, comprising: based on the complete durability load covering the load characteristics in the whole life cycle, combining the fatigue parameters of the parts materials, calculating the cumulative pseudo-damage of each test channel in the whole life cycle to determine the target pseudo-damage.

[0009] Further, the test durability load of the test site section is collected, and the unit test pseudo-damage of each test channel is calculated according to the test durability load, comprising: selecting n basic road sections in the test site, controlling the vehicle to travel at a specified speed under a specific load, and collecting the test durability load of each road section; processing the collected load data of each road section to calculate the unit pseudo-damage of each test channel generated by each road section, wherein the unit pseudo-damage is the fatigue damage of each test channel generated when the vehicle travels the corresponding road section once.

[0010] Further, when calculating the unit test pseudo-damage of each test channel, the rain flow cycle counting method is used to process the test durability load of each road section, and after extracting the load cycle characteristic parameters, the unit pseudo-damage calculation is completed by combining the fatigue characteristic parameters of the parts materials.

[0011] Further, the equal damage model is established according to the target pseudo-damage and the unit test pseudo-damage, and the cycle number of each road condition in the test field is determined according to the equal damage model, including: constructing a target pseudo-damage vector by using the target pseudo-damage of each test channel, constructing a pseudo-damage matrix by using the unit pseudo-damage of each test channel for each road section, and establishing an over-determined equation set based on the vector and the matrix; and obtaining an optimal solution by using a least square method to solve the over-determined equation set.

[0012] Further, the test field test is performed according to the determined cycle number, and the final test durability load is determined, including: determining the core basic road section to be tested in the test field and the corresponding cycle number according to the determined cycle number of each road condition; performing cycle test on the core basic road section for the corresponding number of times; and recording the final test durability load of the fuel cell stack, the air compressor and the vehicle-mounted hydrogen system suspension node synchronously.

[0013] Further, when the final test durability load is determined, the collected test data needs to be checked for integrity to ensure that the data collection dimension is consistent with the data collection dimension of the actual measured durability load.

[0014] Further, the bench test verification is performed according to the final test durability load, and the bench test verification includes simulating a multi-physical field coupling environment to obtain a verification durability load, including: after pretreatment of the final test durability load, frequency domain conversion is performed, fatigue damage spectrum and impact response spectrum are calculated, and three-dimensional vibration power spectrum density is prepared; a multi-physical field coupling environment including three-dimensional random vibration, temperature and humidity control and hydrogen atmosphere is constructed on a bench, and load is applied according to the power spectrum density; acceleration, current, voltage and hydrogen flow data of the test piece are collected synchronously to obtain the verification durability load; and verification is completed by comparing the actual fatigue damage spectrum calculated based on the verification durability load with the target fatigue damage spectrum calculated based on the final test durability load.

[0015] Compared with the prior art, the beneficial effects of the present application are that: The application arranges test channels at multiple key nodes of the target vehicle, the key nodes including the fuel cell stack, the air compressor and the suspension nodes of the vehicle-mounted hydrogen system and the vehicle structure nodes; corresponding acquisition devices are arranged at the test channels to obtain the measured durability load under the target working condition through the acquisition devices; and according to the measured durability load, the complete durability load of the target vehicle in the whole life cycle is derived, the target pseudo-damage of each test channel is determined according to the complete durability load; the test durability load of the test field section is collected, and the unit test pseudo-damage of each test channel is calculated according to the test durability load; the equal-damage model is established according to the target pseudo-damage and the unit test pseudo-damage, and the cycle number of each road condition in the test field is determined according to the equal-damage model; the test field test is carried out according to the determined cycle number, and the final test durability load is determined; the bench test verification is carried out according to the final test durability load, the bench test verification includes simulating the multi-physical field coupling environment to obtain the verification durability load, so as to verify the final test durability load. Through the three-level joint test of the target working condition measured durability load collection, the test field equal-damage model construction and test and the bench multi-physical field coupling verification, the real working condition measured data is taken as the basis, the whole life cycle damage is mapped to the test field road condition cycle number through the equal-damage model, the damage equivalence of the test field test and the real use scene is ensured, and finally the accuracy of the final test durability load is verified through the simulation and verification of the bench multi-physical field coupling environment, so that the data is continuously transmitted and supported in each link, and the authenticity and reliability of the test result are greatly improved.

[0016] In addition, by accurately selecting the fuel cell stack, the air compressor, the suspension nodes of the vehicle-mounted hydrogen system and the vehicle structure nodes as the test channel arrangement positions, these nodes are directly related to the core components of the fuel cell vehicle and the key stress parts of the vehicle, the acquisition devices are arranged to obtain the measured load, the stress characteristics of the core components and the vehicle under the real working condition can be accurately captured, accurate data basis is provided for the derivation of the complete durability load in the whole life cycle, the determination of the target pseudo-damage and the subsequent test field and bench test, the data one-sidedness caused by improper selection of test nodes is avoided, the mechanical durability characteristics of the core components and the vehicle can be fully reflected through the test, and the authenticity and reliability of the test result are improved. BRIEF DESCRIPTION OF DRAWINGS

[0017] In order to more clearly illustrate the specific embodiments of the present application or the technical solutions in the prior art, the drawings needed to be used in the specific embodiments or the prior art description will be briefly introduced below. Obviously, the drawings in the following description are some embodiments of the present application, and those skilled in the art can also obtain other drawings according to these drawings without creating any inventive labor.

[0018] Figure 1A flow chart of a fuel cell characteristic vehicle and part mechanical durability test method is provided by the embodiment of the present application. DETAILED DESCRIPTION

[0019] To make the objectives, technical solutions and advantages of the present application clearer, the technical solutions of the present application will be described below clearly and completely. Obviously, the described embodiments are only some of the embodiments of the present application, but not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work fall within the scope of the present application.

[0020] The specific embodiments of the present application will be described below.

[0021] In view of the poor reliability of the test method in the prior art, the present application arranges test channels at multiple key nodes of a target vehicle; obtains actual durability load under a target working condition, and deduces complete durability load in a whole life cycle to determine target pseudo-damage of each test channel; collects test field test durability load, calculates unit test pseudo-damage of each test channel according to the test durability load; establishes an equal-damage model according to the target pseudo-damage and the unit test pseudo-damage, and determines the number of cycles of each road condition in the test field, and performs test field test according to the number of cycles to determine the final test durability load; performs multi-physical field coupling environmental bench test according to the final test durability load to obtain calibration durability load, and calibrates the final test durability load. The test method of the present application has high reliability.

[0022] The embodiment of the present application provides a fuel cell characteristic vehicle and part mechanical durability test method, Figure 1 A flow chart of a fuel cell characteristic vehicle and part mechanical durability test method is provided by the embodiment of the present application, as shown in Figure 1 The specific steps include the following steps: S1, test channels are arranged at multiple key nodes of a target vehicle, and the key nodes include suspension nodes of a fuel cell stack, an air compressor and a vehicle-mounted hydrogen system, and vehicle structure nodes.

[0023] Critical nodes refer to the core locations in a fuel cell vehicle that are directly related to the stress on core components and the durability of the overall vehicle structure. Test channels are dedicated data acquisition paths used to collect load and operating condition data; each channel corresponds to the acquisition of specific physical signals at a specific measurement point. The fuel cell stack is the core power component of a fuel cell vehicle, and its mounting nodes are the installation locations where the fuel cell stack connects to the vehicle. The air compressor is a key auxiliary component of the fuel cell system; its mounting nodes are the installation points where the air compressor directly bears the vibration and impact loads during operating cycles. The onboard hydrogen system is the core system for storing and supplying hydrogen; its mounting nodes are the installation points where the hydrogen storage tanks are mounted. Mounting nodes are the installation locations where core components connect to the vehicle, serving as the key interface for transmitting stress from components to the vehicle and the direct point of application for vibration and impact loads. Vehicle structural nodes are the parts of the vehicle structure that bear critical loads, and correspondingly, Global Positioning System (GPS) sensor channels are deployed to collect background data on the vehicle's operating conditions.

[0024] For example, a total of 40 test channels are set up, including 4 test points set up in front of, behind, to the left and right of the fuel cell stack suspension node, each test point collecting x, y, and z acceleration signals, for a total of 12 channels; 4 test points set up in front of, behind, to the left and right of the air compressor suspension node, each test point collecting x, y, and z acceleration signals, for a total of 12 channels; 4 test points set up in front of, behind, to the left and right of the on-board hydrogen system suspension node, each test point collecting x, y, and z acceleration signals, for a total of 12 channels; and GPS sensing channels are deployed at the vehicle structure nodes, corresponding to 4 test channels, to measure vehicle speed, longitude, latitude, and altitude data respectively. By deploying test channels at multiple key nodes, the system accurately focuses on the core components of the fuel cell vehicle and the key stress-bearing parts of the vehicle structure, ensuring that the collected data directly reflects the actual stress state of the core components and the entire vehicle.

[0025] S2. Set up corresponding data acquisition equipment at the test channel to obtain the measured durability load under the target working condition; and based on the measured durability load, deduce the complete durability load of the target vehicle throughout its entire life cycle.

[0026] The data acquisition equipment refers to the signal acquisition components adapted to the test channel, including accelerometers, strain sensors, temperature, humidity, and pressure sensors, and GPS sensors, used to capture physical signals and operating condition data related to mechanical durability. The target operating condition refers to typical motion scenarios in the actual service of a fuel cell vehicle, including cyclic operating conditions such as idling, acceleration from a standstill, constant speed driving, and braking deceleration, covering different road surfaces, altitudes, and other environmental conditions. The measured durability load refers to the raw physical signals directly acquired by the acquisition equipment, reflecting the stress state of core components and the entire vehicle under the target operating condition, including data on acceleration, strain, temperature, humidity, pressure, vehicle speed, and altitude. The entire life cycle refers to the complete service life of the target vehicle model from its introduction into service until it reaches its design fatigue life, covering the cumulative effects of all typical operating conditions and environments. The complete durability load refers to the complete load spectrum covering all load characteristics throughout the entire life cycle of the target vehicle model, obtained by extending and deriving the measured durability load through specific mathematical methods, including key information such as the amplitude and frequency of various loads in actual use.

[0027] Specifically, corresponding data acquisition equipment is deployed at the test lanes to obtain the measured durability loads under the target working conditions. This includes: conducting a comprehensive survey of the test site's route, paving conditions, and safety to determine the road combinations covering the target working conditions and arranging the test sequence; deploying acceleration sensors, strain sensors, and temperature, humidity, and pressure sensors at each test lane, calibrating them sequentially according to the lanes, and setting fixed sampling frequencies, ranges, and trigger conditions; after conducting test runs to verify and check for abnormal data, starting the vehicle and simultaneously acquiring acceleration, strain, temperature, humidity, and pressure signals and vehicle positioning data from each test lane; and formatting the data, calibrating the time stamps, and generating metadata after data acquisition.

[0028] First, a comprehensive survey of the test site's route, paving conditions, and safety was conducted to identify the road combinations that could cover the target operating conditions. The test sequence was then rationally arranged based on the operating condition level and mileage requirements. Subsequently, accelerometers, strain sensors, and temperature, humidity, and pressure sensors were deployed in each test channel. After calibration was completed for each channel, fixed sampling frequencies, ranges, and trigger conditions were set. Once the equipment was installed, a trial run was conducted to verify its integrity. Waveform integrity and communication reliability were checked in real time, and anomalies such as data drift, breakpoints, and noise were promptly investigated. After confirming there were no problems, the vehicle was started. Acceleration, strain, temperature, humidity, and pressure signals, as well as vehicle positioning data, were simultaneously collected from each test channel. After data acquisition, the data was formatted, time-stamped, and metadata containing acquisition parameters and data integrity identifiers was generated, providing the original basis for subsequent durability spectrum derivation.

[0029] Through comprehensive preliminary surveys and reasonable test planning, it was ensured that the collected load data fully covered the typical operating conditions of the target vehicle model in actual service, avoiding data bias due to missing operating conditions. The process of sensor channel calibration, setting unified acquisition parameters, and test runs effectively guaranteed the accuracy and reliability of data acquisition, reducing the interference of abnormal data on subsequent analysis. The formatting, time-stamping, and metadata generation of the collected data formed a standardized data system for the measured durability loads, providing accurate and standardized raw data support for subsequent core aspects such as the extended derivation of complete lifecycle durability loads and the calculation of target pseudo-damage, laying the foundation for the reliability of the entire testing method.

[0030] Specifically, based on the measured durability load, the complete durability load of the target vehicle throughout its entire life cycle is derived, including: processing the measured durability load using the rainflow cycle counting method to obtain two-dimensional matrix information of the amplitude and mean of the load cycles; and extending the two-dimensional matrix information based on the two-dimensional diffusion kernel density estimation method to derive the complete durability load of the target vehicle throughout its entire life cycle.

[0031] Rainflow cycle counting is a widely used load statistics method in durability and reliability engineering. It records the initial and final values ​​of load cycles and extracts key information such as the amplitude and mean of load cycles. The results correspond to the stress-strain hysteresis loop in materials mechanics. A load cycle refers to the complete process of a load changing from an initial value back to its corresponding state. It is the basic unit reflecting component fatigue damage and includes load fluctuation characteristics and intensity information. The two-dimensional matrix information of amplitude and mean is a set of load statistics data obtained after processing by the rainflow cycle counting method. It presents the amplitude and mean of all load cycles in two-dimensional matrix form and is the core data for quantifying load characteristics. The two-dimensional diffusion kernel density estimation method is a mathematical extrapolation method for load spectrum expansion. Based on limited measured load statistics data, it can reasonably supplement load characteristics not captured throughout the entire life cycle, achieving complete coverage of load range and frequency.

[0032] The rainflow cycle counting method is selected to systematically process the measured durability loads collected under the target operating conditions. This method identifies the cyclic patterns of the loads, records the start and end states of each cycle, and then extracts the amplitude and mean of all load cycles, organizing them into a two-dimensional matrix to ensure that the load characteristics are accurately quantified. Using this two-dimensional matrix information as input, the two-dimensional diffusion kernel density estimation method is used for extrapolation. This method can reasonably supplement the load types and frequencies that were not captured throughout the entire life cycle based on the statistical distribution patterns of the measured loads, and finally derive the complete durability load covering all load characteristics throughout the complete service life of the target vehicle model.

[0033] The application of the rainflow cycle counting method provides a clear physical basis for the processing of measured loads. The resulting two-dimensional matrix information of amplitude and mean values ​​can truly reflect the core characteristics of the load, providing a reliable data foundation for subsequent extended derivations and avoiding the one-sidedness of load feature extraction. The two-dimensional diffusion kernel density estimation method effectively overcomes the limitation that measured data can only cover limited working conditions, achieving a reasonable expansion from limited measurements to full life cycle coverage. This ensures that the complete durability load contains key information of various loads in real use, providing complete and effective load data support for the accurate calculation of target pseudo-damage in each test channel and damage mapping in the test field, thus guaranteeing the reliability and accuracy of the entire testing method.

[0034] S3. Determine the target pseudo-damage for each test channel based on the complete durability load. Specifically, this includes: calculating the cumulative pseudo-damage generated in each test channel over the entire life cycle based on the full life cycle load characteristics covered by the complete durability load and the fatigue parameters of the component materials, in order to determine the target pseudo-damage.

[0035] Target spurious damage refers to the quantified cumulative fatigue damage of the core components or vehicle structural parts corresponding to each test channel under complete durability loads throughout their entire life cycle. It is the core basis for establishing subsequent damage models. The fatigue parameters of component materials refer to the fatigue characteristic indicators of the materials used in the fuel cell stack, air compressor, on-board hydrogen system, and vehicle structure. They are the key input parameters for calculating target spurious damage.

[0036] First, the full life cycle load characteristics of each test channel in the complete durability load are extracted, including key data such as the amplitude, frequency, and mean of the load cycles. Then, the fatigue parameters of the component materials of the core components or vehicle structure corresponding to each test channel are matched. Finally, through fatigue damage quantification calculation, the cumulative pseudo-damage generated by each test channel in the full life cycle under the action of the complete durability load is obtained. This cumulative pseudo-damage is the target pseudo-damage of each test channel.

[0037] Calculations are performed based on the full life-cycle load characteristics of complete durability loads and the inherent fatigue parameters of component materials, rather than relying on empirical estimations. This ensures that the quantitative results of the target pseudo-damage in each test channel accurately reflect the fatigue damage pattern throughout the entire life cycle, avoiding the one-sidedness and errors in damage assessment. The target pseudo-damage in each test channel serves as core data, providing a crucial basis for establishing equal damage models and determining the number of test track road condition cycles.

[0038] S4. Collect test durability loads from the test track sections and calculate the unit pseudo-damage for each test lane based on these loads. Specifically, this includes: selecting n basic road sections at the test track, controlling the vehicle to travel at a specified speed under a specific load, and collecting test durability loads from each road section; processing the collected load data for each road section to calculate the unit pseudo-damage generated by each road section for each test lane. The unit pseudo-damage is the fatigue damage generated by each test lane when the vehicle travels through the corresponding road section once. When calculating the unit pseudo-damage for each test lane, the rainflow cycle counting method is used to process the test durability loads of each road section. After extracting the load cycle characteristic parameters, the unit pseudo-damage is calculated by combining these parameters with the fatigue characteristic parameters of the component materials.

[0039] The test track section refers to a road segment within a test facility specifically used for vehicle durability testing; it is the specific scenario for collecting test durability loads. Test durability load refers to load data collected through test channels while the vehicle is traveling on the test track section, reflecting the stress state of core components and the entire vehicle. Unit test spurious damage refers to the quantified fatigue damage value generated in each test channel when the vehicle travels once on the corresponding basic road segment of the test track. n basic road segments refer to several representative road segments selected from the test track for load collection and damage calculation. Specific load refers to the preset load state maintained by the vehicle during testing; it is a key parameter to ensure the consistency of test conditions. The fatigue characteristic parameters of component materials refer to the inherent fatigue performance indicators of the materials used in the fuel cell stack, air compressor, on-board hydrogen system, and vehicle structure; they are the core basis for calculating fatigue damage.

[0040] First, n basic road sections are selected from the test track. Test vehicles are controlled to travel at specified speeds under specific loads, and the corresponding test durability loads for each road section are collected through pre-set test channels. The collected load data for each road section are systematically processed, with the calculation target clearly defined as the unit pseudo-damage generated by each road section on each test channel. Finally, when calculating the unit test pseudo-damage, the rainflow cycle counting method is used to process the test durability loads of each road section, extracting the load cycle characteristic parameters. These parameters are then combined with the fatigue characteristic parameters of the corresponding component materials for each test channel to complete the calculation of the unit test pseudo-damage for each test channel. By selecting basic road sections from the test track and controlling the vehicle with specific loads and speeds, the consistency and repeatability of the test durability load collection environment are ensured, avoiding data deviations caused by fluctuations in operating conditions and providing a reliable basis for subsequent damage calculations.

[0041] S5. Establish an equal damage model based on the target spurious damage and the unit test spurious damage, and determine the number of cycles for each road condition in the test field based on the equal damage model.

[0042] The equivalent damage model refers to a mathematical model that correlates target pseudo-damage with unit test pseudo-damage. Its core logic is to achieve equivalence between the total damage in the test range and the target pseudo-damage throughout its entire lifecycle by matching the number of cycles for each road condition in the test range with the unit test pseudo-damage. Each road condition in the test range refers to the road conditions corresponding to n basic road segments selected from the test range, and is the specific implementation object of the test range.

[0043] Specifically, an equal damage model is established based on the target pseudo-damage and the unit test pseudo-damage, and the number of cycles for each road condition in the test field is determined based on the equal damage model. This includes: constructing a target pseudo-damage vector with the target pseudo-damage of each test channel, constructing a pseudo-damage matrix with the unit pseudo-damage of each road segment to each test channel, establishing an overdetermined system of equations based on the vector and matrix, and solving the overdetermined system of equations using the least squares method to obtain the optimal solution.

[0044] Based on the above embodiments, using the measured user road load data collected from 40 test lanes of the target vehicle model, the load cycle characteristics are statistically analyzed using the rainflow cycle counting method. This is then combined with two-dimensional kernel density extrapolation technology (i.e., two-dimensional diffusion kernel density estimation method) to expand the target load spectrum to cover the entire vehicle lifecycle. Subsequently, the pseudo-damage b for the entire user lifecycle of each test lane is calculated. i (That is, the fatigue damage quantification value of this channel during the entire actual use cycle by the user), and integrate these m pseudo-damage values ​​to construct an m×1 dimensional target pseudo-damage vector D=[b1,b2,…,b m ] In the test field, n basic road sections are selected, where n ranges from 50 to 100. Vehicles are driven under specific loads and speeds, and load data for each road section is collected through 40 identical test lanes. The unit pseudo-damage A of the j-th road section to the i-th lane is calculated. i,j (i.e., fatigue damage in the i-th lane when a vehicle travels this road segment once), and all A i,j The components are integrated and constructed into an m×n dimensional pseudo-damage matrix P=[[A 11 A 12 ,…,A 1n ],[A 21 A 22 ,…,A 2n ],…,[A m1 A m2 ,…,A mn Based on the pseudo-damage matrix P and the target pseudo-damage vector D, an overdetermined system of equations is established: P×w=D, where w=[w1,w2,…,w…]. n ] Let w be an n×1 dimensional vector of the number of cycles for each basic road segment in the test field. jLet w be the number of iterations for the j-th road segment, taking a non-negative integer value. Since the number of test channels m is much larger than the number of basic road segments n, the system of equations has no unique solution. Therefore, the least squares method is needed to find the optimal iteration vector α instead of w. In the solution, first define the residual sum of squares function S(α) = ||Pα-D||², where S(α) measures the error between the damage of each channel in the test field and the user's target damage, and α is the optimal iteration vector for each basic road segment in the test field. Then, expand it as... , where r i Let r be the damage residual of the i-th test channel, where r i =(Pα) i -D i That is, the difference between the damage in the test field and the damage to the user target. Since S is a concave function, its minimum value occurs at the position where the partial derivative is 0. The normal equation (P) is derived by taking the derivative. ×P)×α=P ×D, where P Given the transpose of the pseudo-damage matrix P, the optimal cycle number vector α = (P) is finally obtained. ×P) -1 ×P ×D, where (P) ×P) -1 Let P be a matrix The inverse matrix of ×P can be calculated in a streamlined manner using the nCode software.

[0045] This embodiment achieves a scientific damage mapping between user roads and test tracks. By accurately quantifying target pseudo-damage and unit test pseudo-damage, and combining it with the least squares method, it ensures that the total damage generated by the test track through road segment cycles is highly consistent with the user's full life cycle damage. This solves the problems of inaccurate damage reproduction and difficulty in reflecting real-world conditions in traditional testing. Full life cycle damage can be equivalently reproduced through controllable road segment cycles at the test track, shortening the testing cycle and significantly improving testing efficiency, eliminating the need for long-distance, time-consuming real-vehicle road testing. This provides a reliable basis for subsequent accelerated bench testing. The determined optimal number of cycles guides the test track in accurately collecting component load data, supporting subsequent calculations of fatigue damage spectrum (FDS), shock response spectrum (SRS), and bench power spectral density (PSD).

[0046] S6. Conduct test field testing according to the determined number of cycles to determine the final test durability load. Specifically, this includes: identifying the core basic road sections to be tested in the test field and their corresponding number of cycles based on the determined number of cycles for each road condition; conducting the corresponding number of cycle tests on the core basic road sections; and simultaneously recording the final test durability loads for the fuel cell stack, air compressor, and on-board hydrogen system mounting nodes. When determining the final test durability load, the integrity of the collected test data must be verified to ensure that the data collection dimensions are consistent with the data collection dimensions of the actual measured durability loads.

[0047] The core basic road section is a key reinforced durability road surface selected from 50-100 basic road sections in the test field. It is capable of efficiently replicating real-world user damage and is the core object of the test field. The final test durability load refers to the quantified form of the fatigue load borne by the fuel cell stack, air compressor, and onboard hydrogen system suspension nodes under coupled real-world operating conditions during test field testing. Specifically, it is represented by x, y, and z-axis acceleration signals. Data acquisition dimensions refer to the unified standards for test channels, signal types, sampling frequency, range, and triggering conditions.

[0048] Based on the optimal number of cycles calculated in the above embodiments, core road segments are selected from 50-100 basic road segments in the test field. The specific number of cycles for each core road segment is determined to ensure that the road segment combination can reproduce the damage throughout the user's entire life cycle. Cyclic tests are conducted on the core road segments at the test field with the corresponding number of cycles. The entire test uses 40 fixed test channels to ensure that the data acquisition process is consistent with the user's actual road tests. The x, y, and z-axis acceleration signals of the three key components' suspension nodes are collected and stored in real time. Simultaneously, auxiliary data such as vehicle speed and altitude from the vehicle's GPS are recorded to avoid data loss or deviation. The integrity of the collected data is verified, including checking for drift, breakpoints, and abnormal noise. The data acquisition dimensions, such as test channels, signal types, and parameter settings, are checked for consistency with the user's actual road durability load. Finally, the data is integrated to form the final test durability load, which serves as the core input for subsequent FDS and SRS calculations and bench PSD compilation.

[0049] By combining core road sections with the optimal number of cycles, the test track can effectively reproduce fatigue damage in real-world user scenarios, solving the problems of damage mismatch and detachment from actual working conditions in traditional testing. This enables equal damage mapping between public roads and the test track. Synchronous recording ensures that load data is consistent with the testing process, and integrity checks and dimensional consistency confirmations prevent data defects, guaranteeing the authenticity and representativeness of the final durability load test. This provides high-quality data support for subsequent accelerated bench testing.

[0050] S7. Bench testing verification is performed based on the final test durability load. Bench testing verification includes simulating a multiphysics coupling environment to obtain a verification durability load for verifying the final test durability load. Specifically, this includes: preprocessing the final test durability load and performing frequency domain transformation to calculate the fatigue damage spectrum and impact response spectrum, and compiling the triaxial vibration power spectral density; constructing a multiphysics coupling environment on the bench including triaxial random vibration, temperature and humidity control, and a hydrogen atmosphere, and applying the load according to the power spectral density; simultaneously acquiring acceleration, current, voltage, and hydrogen flow data of the test piece to obtain the verification durability load; and completing the verification by comparing the actual fatigue damage spectrum calculated based on the verification durability load with the target fatigue damage spectrum calculated based on the final test durability load.

[0051] Multiphysics coupled environment refers to a composite operating condition constructed on a test bench to reproduce the real operating conditions of a fuel cell vehicle. Load data synchronously acquired during bench testing, reflecting the actual response of the test specimen, including triaxial acceleration signals, stack current, voltage, and hydrogen flow rate, is the core input for calculating the actual fatigue damage spectrum. Frequency domain transformation, through Fast Fourier Transform, converts the preprocessed time-domain load data into frequency-domain data, a prerequisite for calculating FDS and SRS. The fatigue damage spectrum is a spectrum calculated based on frequency-domain data and Waller curve parameters, quantifying the cumulative fatigue damage of the test specimen under cyclic loading. It includes the target FDS calculated based on the final test durability load and the actual FDS calculated based on the verification durability load. The impact response spectrum is a spectrum calculated based on time-domain load data, reflecting the dynamic response of the test specimen under impact loading. Its maximum value envelope is used to set the upper limit boundary of the PSD calculation to avoid overload damage. The target fatigue damage spectrum is the FDS curve obtained by using the final test durability load as input and the FDS calculation formula; it is the damage benchmark that needs to be reproduced in the bench test. The actual fatigue damage spectrum is the FDS curve obtained by using the verification durability load as input and the FDS calculation formula. This curve is then compared and verified with the target FDS. The FDS calculation formula is as follows: ; Among them, f n Here, T represents the effective sampling time of the component load, in seconds; K represents the spring stiffness of the single-degree-of-freedom system; and b and c are the fatigue parameters of the Waller curve, defined as N = c. S -b N is the number of cyclic stress amplitudes, S is the number of fatigue failure cycles, Q is the dynamic amplification factor, and G is the dynamic amplification factor. Z (f n ) represents the power spectral density of the component load, in units of (m / s²)² / Hz, Γ( ) is the Gamma function.

[0052] Based on effective time-domain data obtained from typical road surfaces in the test field, the shock response spectrum (SRS) for each road segment is calculated. Then, the SRS of all road segments are superimposed, and the envelope of the maximum SRS value for each road surface is taken as the total SRS curve. Taking the vibration data of a Belgian road for a fuel cell stack as an example, the corresponding SRS distribution curve can be obtained through time-domain load data. The total SRS curve is used to set the upper limit boundary conditions for the PSD (Power Suppression Device) to avoid excessively severe loads during testing that could lead to abnormal failures. The formula for calculating SRS is as follows: ; ; Wherein, SRS(fn) is the impulse response spectrum value at frequency fn, in m / s². The measured acceleration load of the component is expressed in m / s², t is the time variable in seconds, T is the total sampling duration of the component load data in seconds, fn is the natural frequency in Hz, covering the key modal frequencies of the component from 10 to 2000 Hz, and ζ is the damping ratio, dimensionless, set according to the material properties of the component, generally taken as 0.05, and 0.02 to 0.03 for metal structural parts. Let f be the relative acceleration response of a single-degree-of-freedom system at frequency fn, in m / s², and τ be the integral variable, in seconds.

[0053] For example, bench-accelerated power density spectrum (PSD) and verification include: PSD compilation The calculation formula is compiled using PSD, and the target fatigue damage value and test duration at each frequency point, as well as other fixed values ​​such as the safety factor k, are input. Then, the final vibration test PSDs in the X, Y, and Z directions can be obtained by processing them through the nCode software program.

[0054] ; Where Teq is the test duration, ΣFDS(fn) is the target FDS total curve, and k is the safety factor.

[0055] 1. PSD verification To verify the rationality and operability of the prepared PSD, the limiting response spectrum ERS generated by the PSD needs to be smaller than the total SRS curve. The purpose is to ensure that the test is a durability test within the normal load impact range; otherwise, it is a destructive test.

[0056] The core purpose of bench testing is not simply to verify damage data, but to achieve a closed loop of accelerated reproduction, performance verification, and quantitative evaluation based on equivalent damage data from the test field. It also includes damage calculation and comparison steps, and the specific logic is as follows: The core objectives of bench testing include: Accelerated reproduction of full life cycle damage: By using a programmed triaxial vibration PSD (power spectral density) to apply accelerated loads on a temperature-humidity-hydrogen coupling test bench, the user's full life cycle damage mapped in the test field is compressed into a short time, solving the pain point of long test cycles in real vehicles / test fields.

[0057] Verify the durability and reliability of components / systems: Simulate multi-physics coupled environments (vibration, hydrogen, and temperature and humidity) under real-world operating conditions to assess the fatigue life, structural integrity (such as whether cracks or detachment occur), and performance stability of key components such as fuel cell stacks and air compressors, rather than simply verifying damage data.

[0058] Quantifying the relationship between damage and performance: By calculating the damage from bench tests, a quantitative model of load-damage-performance degradation is established, providing direct data support for product design optimization and life prediction, and filling the gap in the test field's inability to accurately simulate multi-field coupling.

[0059] Damage data transmission and comparison logic The verification target is the target damage of the user throughout the entire life cycle after the test field is mapped, rather than the original damage data of the test field or the actual damage data of the user's road.

[0060] Transfer path: User road measured load - extrapolation of full life cycle load - calculation of user pseudo-damage - mapping to equivalent damage at the test site - summarization into target FDS - compilation of bench PSD - bench test to reproduce the damage corresponding to the target FDS.

[0061] The core of the comparison is that the bench test needs to calculate the actual damage it generates and compare it with the target FDS to ensure that the error between the two is within the allowable range (generally ≤5%). This proves that the bench acceleration load can accurately reproduce the target damage, rather than verifying the accuracy of the original test field data.

[0062] Damage calculation and comparison process of bench tests 2. Damage calculation stage: During bench testing, sensors were used to collect data such as triaxial acceleration of components, fuel cell current, voltage, and hydrogen flow rate. Based on the FDS calculation formula, substitute the load data from the bench test to calculate the actual FDS generated by the bench test, i.e., the actual damage value. Simultaneously, the limit response spectrum (ERS) of the bench test is calculated and compared with the total SRS to ensure that ERS ≤ total SRS and avoid overload failure.

[0063] Comparison and verification logic: First comparison: Compare the actual FDS on the test bench with the target FDS to ensure damage equivalence; Second comparison: Compare the bench ERS with the total SRS to ensure that the load is within the safety boundary; The third comparison: compare the performance data of the components tested on the bench with the target design requirements to verify whether the durability and reliability meet the standards.

[0064] Bench testing uses target damage as a benchmark, reproduces the damage by accelerating the load, confirms the reproduction accuracy by damage calculation and comparison, and finally verifies the product reliability.

[0065] 3. Multiphysics coupling mechanism of hydrogen-electricity-mechanical vibration To achieve a comprehensive evaluation of fuel cell vehicles under complex operating conditions, this invention proposes a multi-physics coupling testing mechanism involving hydrogen, electricity, and mechanical vibration. Under this mechanism, the relationship between hydrogen flow rate, vibration, and fuel cell stack power output is precisely quantified.

[0066] First, hydrogen flow rate directly affects the performance of the fuel cell stack. The current and voltage output of the fuel cell stack are closely related to the hydrogen supply. Insufficient hydrogen flow rate leads to unstable power output from the stack, affecting the vehicle's operating efficiency. The relationship between hydrogen flow rate and stack power can be expressed by the following formula: ; Where Q is the hydrogen flow rate, k' is a proportionality constant that depends on the fuel cell stack design, and I is the fuel cell stack output current. This formula shows the linear relationship between the hydrogen flow rate and the fuel cell stack current, reflecting how changes in hydrogen supply directly affect the fuel cell stack's output power.

[0067] Secondly, mechanical vibration affects the internal structure of the fuel cell stack, leading to instability in its power output. Vibration influences the electrochemical reactions by altering the physical stress within the stack. This effect can be mathematically modeled using stress and displacement formulas. The relationship between the vibration-induced displacement Δx and stress σ and the stack's power output can be expressed by the following formula: ; Where P is the fuel cell stack output power, I is the fuel cell stack output current, V is the fuel cell stack output voltage, Δx is the displacement caused by vibration, and σ is the stress caused by vibration. This formula can quantify the impact of vibration on the fuel cell stack power output and help assess the long-term stability of the fuel cell stack in a vibrating environment.

[0068] Finally, to achieve multi-physics coupling of hydrogen, vibration, and fuel cell power, a multi-physics coupling test platform was designed. This platform integrates hydrogen flow and pressure sensors and a vibration platform, enabling it to simulate the operating conditions of fuel cells under different environments and monitor and analyze the correlation between vibration and power output in real time. Through this platform, this embodiment can accurately simulate changes in hydrogen flow and pressure, as well as complex vibration conditions, ensuring the authenticity and validity of the test results.

[0069] 4. Correlation Modeling and Testing Methods between Vibration and Stack Power To accurately model the impact of vibration on the power output of the fuel cell stack, this embodiment employs the following methods to analyze the relationship between vibration and fuel cell stack power.

[0070] Vibration displacement and stress modeling First, the impact of vibration on the internal physical structure of the fuel cell stack can be modeled using vibration displacement and stress formulas. Changes in vibration displacement lead to changes in stress in the stack structure, which in turn affects the electrochemical reactions and power output of the stack. The vibration-induced displacement Δx can be expressed as: ; Where Δx is the vibration displacement, A is the vibration amplitude, ω=2πf is the angular frequency, f is the vibration frequency, and t is the time. The amplitude and frequency of vibration are crucial to the impact of the fuel cell stack, and these parameters directly affect the output power of the fuel cell stack.

[0071] The stress σ caused by vibration can be calculated using the following formula: ; Where σ is the stress and E is the elastic modulus of the material. Let Δx be the strain, Δx be the displacement caused by vibration, and L be the length of the fuel cell stack material. This formula reveals how vibrational displacement is transmitted to the fuel cell stack material through stress, thereby affecting the electrochemical reactions and power output of the fuel cell stack.

[0072] Testing System and Feedback Mechanism The multiphysics coupling test platform collects real-time data on hydrogen flow rate, vibration, current, and voltage, and features adaptive feedback adjustment: if vibration causes significant fluctuations in power output, the system automatically adjusts the hydrogen flow rate or vibration frequency to ensure the stability and accuracy of the test. This modeling method and test system accurately reveal the impact of vibration on the power output of the fuel cell stack.

[0073] The multiphysics coupling test platform of this embodiment can acquire hydrogen flow rate, vibration, current, and voltage data in real time and adjust test conditions using a feedback mechanism. For example, if vibration causes significant fluctuations in power output, the system will automatically adjust the hydrogen flow rate or vibration frequency to maintain test stability and accuracy. Through these modeling methods and test systems, this invention can accurately reveal the impact of vibration on the power output of fuel cell stacks and provide strong data support for the durability design, life prediction, and standardized testing of fuel cell vehicles.

[0074] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the technical solutions of the embodiments of the present invention.

Claims

1. A method for testing the mechanical durability of a vehicle and its components with fuel cell characteristics, characterized in that, include: Test channels are arranged at multiple key nodes of the target vehicle model, including the suspension nodes of the fuel cell stack, air compressor and on-board hydrogen system, as well as the vehicle structure nodes. Corresponding data acquisition devices are deployed at the test channel to obtain the measured durability load under the target working condition. Based on the measured durability load, the complete durability load for the entire life cycle of the target vehicle model is derived. The target spurious damage for each test channel is determined based on the complete durability load. Collect the test durability load of the test track section, and calculate the unit test spurious damage of each test channel based on the test durability load; Based on the target spurious damage and the unit test spurious damage, an equal damage model is established, and the number of cycles for each road condition in the test field is determined based on the equal damage model. Conduct test field tests according to the determined number of cycles to determine the final test durability load; Bench testing is performed based on the final test durability load. Bench testing includes simulating a multiphysics coupling environment to obtain a verification durability load, which is then used to verify the final test durability load.

2. The method for testing the mechanical durability of a fuel cell-equipped vehicle and its components according to claim 1, characterized in that, Corresponding data acquisition devices are deployed at the test channel to acquire the measured durability load under the target working condition, including: A comprehensive survey of the test site's route, paving conditions, and safety was conducted to determine the road combinations covering the target working conditions and to arrange the test sequence. Accelerometers, strain sensors, and temperature, humidity and pressure sensors were installed in each test channel. After calibration in sequence for each channel, fixed sampling frequency, range and triggering conditions were set. After conducting test runs and checking for abnormal data, the vehicle is started, and acceleration, strain, temperature, humidity and pressure signals and vehicle positioning data of each test channel are collected simultaneously. After the data is collected, the data is formatted, time-stamped, and metadata is generated.

3. The method for testing the mechanical durability of a fuel cell-equipped vehicle and its components according to claim 1, characterized in that, Based on the measured durability loads, the complete durability loads for the entire lifecycle of the target vehicle model are derived, including: The measured durability load was processed using the rainflow cycle counting method to obtain a two-dimensional matrix of load cycle amplitude and mean values. The two-dimensional matrix information is extended based on the two-dimensional diffusion kernel density estimation method, and the complete durability load of the target vehicle throughout its entire life cycle is derived.

4. The method for testing the mechanical durability of a fuel cell-equipped vehicle and its components according to claim 3, characterized in that, The target pseudo-damage for each test channel is determined based on the complete durability load, including: calculating the cumulative pseudo-damage generated by each test channel during the entire life cycle based on the full life cycle load characteristics covered by the complete durability load and the fatigue parameters of the component materials, so as to determine the target pseudo-damage.

5. The method for testing the mechanical durability of a fuel cell-equipped vehicle and its components according to claim 1, characterized in that, The test durability loads of the test track sections are collected, and the unit test spurious damage of each test channel is calculated based on the test durability loads, including: Select n basic road sections at the test site, control the vehicle to drive at a specified speed under a specific load, and collect the test durability load of each road section; The collected load data for each road segment are processed to calculate the unit pseudo-damage generated by each road segment to each test channel. The unit pseudo-damage is the fatigue damage generated by each test channel when the vehicle travels through the corresponding road segment once.

6. The method for testing the mechanical durability of a fuel cell-equipped vehicle and its components according to claim 5, characterized in that, When calculating the unit test pseudo-damage of each test channel, the rainflow cycle counting method is used to process the test durability load of each section. After extracting the load cycle characteristic parameters, the unit pseudo-damage is calculated by combining the fatigue characteristic parameters of the component materials.

7. The method for testing the mechanical durability of a fuel cell-equipped vehicle and its components according to claim 1, characterized in that, An equal damage model is established based on the target spurious damage and the unit test spurious damage, and the number of cycles for each road condition in the test field is determined based on the equal damage model, including: A target pseudo-damage vector is constructed using the target pseudo-damage of each test channel, and a pseudo-damage matrix is ​​constructed using the unit pseudo-damage of each road segment to each test channel. An overdetermined set of equations is then established based on this vector and matrix. The optimal solution is obtained by solving the overdetermined system of equations using the least squares method.

8. The method for testing the mechanical durability of a fuel cell-equipped vehicle and its components according to claim 1, characterized in that, Test the equipment at the test site according to the determined number of cycles, and determine the final test durability load, including: Based on the determined number of cycles for each road condition, the core basic road sections to be tested in the test field and their corresponding number of cycles are identified. Perform a corresponding number of cyclic tests on the core basic road sections; The final test durability loads of the fuel cell stack, air compressor, and on-board hydrogen system suspension nodes were recorded simultaneously.

9. A method for testing the mechanical durability of a fuel cell-equipped vehicle and its components according to claim 8, characterized in that, When determining the final test durability load, the integrity of the collected test data must be verified to ensure that the data collection dimensions are consistent with the data collection dimensions of the actual measured durability load.

10. A method for testing the mechanical durability of a vehicle and its components with fuel cell characteristics according to claim 1, characterized in that, Bench testing was conducted to verify the durability of the final test loads. Bench testing included simulating a multiphysics coupling environment to obtain the verification durability loads, including: After preprocessing the final test durability load, the fatigue damage spectrum and impact response spectrum were calculated by frequency domain transformation, and the triaxial vibration power spectral density was compiled. The test bench was constructed to create a multi-physics field coupling environment including triaxial random vibration, temperature and humidity control and hydrogen atmosphere, and loads were applied according to the power spectral density. The acceleration, current, voltage, and hydrogen flow rate data of the test specimen are collected simultaneously to obtain the verification durability load; Verification was completed by comparing the actual fatigue damage spectrum calculated based on the verification durability load with the target fatigue damage spectrum calculated based on the final test durability load.

Images