Intelligent testing method for fatigue performance of wet brake

Abstract

The present application belongs to the field of mechanical performance testing, and relates to an intelligent test method for fatigue performance of a wet brake, which collects real-time operation data of the wet brake under a load applied by a loading device during a test process through deployment of a multi-source monitoring array, generates a fatigue stress distribution map representing a stress state of a preset test region of the wet brake based on working condition fusion analysis, and correspondingly identifies a failure warning signal representing a current fatigue damage stage to dynamically adjust parameters of the loading device, analyzes fatigue performance response effects of the wet brake under a current loading strategy during a loading intermittent period of a single test cycle, determines a correction direction of a loading strategy of a secondary test cycle, iteratively optimizes to achieve test convergence, improves consistency of test results and real fatigue damage rules, and improves test efficiency and evaluation accuracy through a closed-loop control and self-adaptive iteration framework to provide a reliable basis for fatigue performance evaluation of the wet brake.

Description

Technical Field

[0001] This invention belongs to the field of mechanical performance testing and relates to an intelligent testing method for the fatigue performance of wet brakes. Background Technology

[0002] In modern industry, wet brakes are widely used in heavy-duty equipment such as construction machinery, mining machinery, and heavy vehicles due to their excellent heat dissipation performance, stable braking effect, and long service life. These devices typically operate in complex and harsh working environments, such as heavy-load uphill climbing and frequent braking during mining operations, and continuous operation of construction machinery in muddy terrain.

[0003] As a core safety component, the fatigue performance of wet brakes directly affects the operational safety and reliability of equipment. Fatigue failure leading to brake malfunction can cause anything from equipment downtime and economic losses to potentially serious safety accidents. Therefore, accurate testing of the fatigue performance of wet brakes is of paramount importance.

[0004] Currently, the testing methods for the fatigue performance of wet brakes mainly rely on fixed loading strategies. By simulating load conditions under typical working conditions, the brakes are cyclically loaded to evaluate their fatigue characteristics. One common testing method is to use a preset constant load mode, apply a fixed braking force or torque to the brakes, and observe the performance changes after a certain number of cycles. Another method is based on a limited combination of working conditions, applying multiple typical working conditions to the brakes one by one and recording their fatigue damage characteristics.

[0005] Although the above methods can reflect the fatigue performance of wet brakes to a certain extent, they still have significant technical limitations in practical applications. Specifically, the fixed loading strategy in the existing technology is difficult to fully simulate the dynamic interaction between load and brake state in actual working conditions. Especially under complex working conditions, the changes in load intensity, frequency and environmental factors often exhibit nonlinear characteristics. The fixed loading mode may lead to a deviation between the test results and the actual fatigue damage law, thus affecting the accuracy of the test.

[0006] 2. Existing technologies lack the ability to provide real-time data feedback and dynamic adjustment during the testing process. They cannot flexibly optimize loading parameters based on the actual state of the brake during testing, which limits the further improvement of testing efficiency and evaluation accuracy. It is also difficult to accurately control the test termination point. This may result in incomplete data collection of key working conditions due to premature termination of testing, or waste of resources due to excessively extended testing cycles, further restricting the effectiveness of test results and the economy of the testing process. Summary of the Invention

[0007] In view of this, in order to solve the problems mentioned in the background technology, an intelligent testing method for the fatigue performance of wet brakes is proposed.

[0008] The objective of this invention can be achieved through the following technical solution: This invention provides an intelligent testing method for the fatigue performance of wet brakes, comprising: collecting real-time operating data of the wet brake under load applied by a loading device during the testing process.

[0009] The real-time operating data is subjected to working condition fusion analysis to generate a fatigue stress distribution map characterizing the stress state of the preset test area of ​​the wet brake, and correspondingly identify the failure warning signal characterizing the current fatigue damage stage.

[0010] Based on the fatigue stress distribution diagram and the failure warning signal, the loading parameters of the loading device are coordinated and adjusted.

[0011] During the loading intervals of a single test cycle, process data is continuously collected, the fatigue performance response of the wet brake under the current loading strategy is analyzed, and the direction of correction for the loading strategy in the next test cycle is determined.

[0012] The test process is executed iteratively until the preset test convergence conditions are met, at which point a fatigue performance test completion notification is issued.

[0013] Compared with the prior art, the beneficial effects of the present invention are as follows: (1) The present invention collects the operating status information of the wet brake in real time through a multi-source monitoring array, and combines the working condition fusion analysis to convert the real-time operating data into a fatigue stress distribution map that characterizes the stress state of the preset test area of ​​the wet brake, which helps to grasp the overall stress state of the brake in real time and accurately locate different fatigue stress level areas.

[0014] (2) The present invention dynamically adjusts the loading parameters of the loading device through a dual feedback mechanism of fatigue stress distribution diagram and failure warning signal, so that the loading conditions always match the actual failure risk scenario, avoids the test deviation caused by the solidification of loading parameters in the prior art, and further improves the test efficiency and evaluation accuracy.

[0015] (3) This invention analyzes the fatigue performance response effect of the wet brake under the current loading strategy during the loading interval period during a single test cycle, determines the correction direction of the loading strategy for the next test cycle, promotes the continuous optimization of the loading strategy during the test process until the test is converged and ends, so that the test results can more realistically reflect the overall performance level of the wet brake and improve the test accuracy. Attached Figure Description

[0016] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1 This is a flowchart illustrating the implementation steps of the method of the present invention.

[0018] Figure 2 This is a logic diagram for generating failure warning signals in this invention.

[0019] Figure 3 This is a schematic diagram of the test loop optimization iterative framework of the present invention. Detailed Implementation

[0020] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0021] Please see Figure 1 As shown, the present invention provides an intelligent testing method for the fatigue performance of a wet brake, including: S11. Collecting real-time operating data of the wet brake under load applied by a loading device during the testing process.

[0022] In a preferred embodiment of the present invention, the real-time operation data acquisition process includes: arranging a multi-source monitoring array in a preset test area of ​​the wet brake, and synchronously triggering various sensors in the array to acquire data using a unified clock. The various sensors include at least a temperature sensor, a vibration sensor, and a pressure sensor.

[0023] It should be noted that the pre-designed test area for wet brakes is not a single part, but a cluster encompassing multiple key components. This area is centered on the friction pair, including the working surface of the brake disc and the contact surface of the friction pads. Linked by the oil circulation path, it covers key nodes such as the oil chamber interior, oil inlet, oil outlet, and oil cooler inlet. It also includes the brake housing load-bearing area and the brake hydraulic actuator area as structural support. Therefore, the entire test area is an organic whole, designed to systematically evaluate the comprehensive performance of the brake.

[0024] Real-time acquisition of temperature data, vibration signal data, and braking pressure data of each spatial unit in the preset test area.

[0025] It should be noted that the above spatial unit division is based on the minimum monitoring coverage of multiple sensors, and is determined in combination with the structural characteristics of the brake and monitoring requirements. Specifically, based on the minimum area that each type of sensor can effectively sense, and taking into account the boundaries of key components, thermal and hydraulic distribution characteristics, the test area is divided into several spatial units.

[0026] The collected raw data is processed in a contextualized manner, and corresponding features such as temperature gradient, vibration spectrum, and pressure distribution are extracted.

[0027] It should be noted that the above-mentioned scenario-based processing includes: pressure distribution characteristics are obtained by spatial interpolation and distribution fitting of the braking pressure data collected in real time in each spatial unit, combined with the area of ​​the spatial unit to calculate the pressure distribution, and analyzing the time-series fluctuation variance of the pressure distribution.

[0028] The temperature gradient characteristics are based on the temperature data collected synchronously from each spatial unit. According to the principle of heat conduction and the temperature difference between adjacent units, the temperature field gradient distribution is analyzed by the finite difference element method to reflect the heat accumulation and diffusion characteristics during the braking process.

[0029] Vibration spectrum characteristics are obtained by performing a fast Fourier transform on vibration signal data to convert the time-domain waveform into a frequency-domain energy distribution, extracting the vibration energy characteristics of each spatial unit at different frequencies, thereby identifying the vibration modes and abnormal frequency components during the braking process.

[0030] The extracted features are integrated into vector form and used as real-time running data.

[0031] S12. Perform working condition fusion analysis on the real-time operating data to generate a fatigue stress distribution map characterizing the stress state of the preset test area of ​​the wet brake, and correspondingly identify the failure warning signal characterizing the current fatigue damage stage.

[0032] In a preferred embodiment of the present invention, the step of performing condition fusion analysis on real-time operating data includes: identifying a high-confidence set of operating conditions by verifying the consistency of pressure distribution characteristics and temperature gradient characteristics.

[0033] It should be added that the above consistency verification process is as follows: obtain the pressure fluctuation degree and temperature gradient intensity at the same time point of the same spatial unit, where the pressure fluctuation degree refers to the fluctuation variance value of the dynamic time series formed by the loading start time point to the current time point of the same spatial unit. Construct a coordinate system with the pressure fluctuation degree as the horizontal axis and the temperature gradient intensity as the vertical axis. Substitute the data pairs of pressure fluctuation degree and temperature gradient intensity at each unit time point during the loading period of the same spatial unit into the coordinate system, calculate the Pearson correlation coefficient of the corresponding fitted curve. If the coefficient is greater than or equal to the preset strong positive coefficient threshold, verify that the pressure distribution characteristics and temperature gradient characteristics of the spatial unit are consistent; otherwise, verify that they are not. The calibration of the preset reasonable coefficient threshold is based on the positive correlation characteristics of pressure distribution and temperature gradient. Its value range is at least greater than 0 and conforms to the principle of approaching 1. It can be set to 0.6 for example.

[0034] If it is verified that the pressure distribution characteristics and temperature gradient characteristics of a certain space unit are inconsistent, the local correlation coefficient in the fitted curve is quantified by sliding window. If the local correlation coefficient is less than 0, it is considered to be negatively correlated. If it is lower than the preset weak positive correlation coefficient threshold, it is considered to be weakly positively correlated. Data pairs that show negative correlation or weak positive correlation in the corresponding fitted curve are retrieved, and confidence weighting processing or neighbor data correction processing is performed on them. The preset weak positive correlation coefficient threshold is the difference between 1 and the preset strong positive coefficient threshold.

[0035] The basic principle of consistency verification is that the fatigue performance of wet brakes is affected by both mechanical stress distribution and heat dissipation, and the two are highly coupled physically. Pressure distribution characteristics reflect the uniformity of stress between the braking pairs, while temperature gradient characteristics reflect the dynamic balance between heat accumulation and diffusion. Theoretically, the two should have a significant positive correlation. Taking the pressure concentration area as an example, it is usually accompanied by significant temperature rise and pressure fluctuation changes.

[0036] Based on the fuzzy rule method, the relative performance of the pressure distribution characteristics and temperature gradient characteristics of each spatial unit in the feature set is mapped to a predefined fatigue stress level, wherein the fatigue stress level includes low, medium or high.

[0037] It should be noted that the above-mentioned fuzzy rule method is applied as follows: The fuzzy set and membership function of the defined pressure distribution features and temperature gradient features are used as benchmarks. The numerical values ​​of the pressure distribution features and temperature gradient features of each spatial unit in the feature set are converted into the membership degree corresponding to the fuzzy linguistic variables "low pressure / medium pressure / high pressure" and "small gradient / medium gradient / large gradient". This membership degree is a continuous value between 0 and 1, used to quantify the degree to which the feature value belongs to a certain fuzzy set.

[0038] According to the preset fuzzy rule library, all rules that match the current input membership degree are retrieved and activated. The truth degree of the premise part of each activated rule is obtained by the membership degree accumulation calculation. The calculation result will be used as the consequent part of this rule, that is, the output strength of the fatigue stress level.

[0039] Aggregate all fatigue stress fuzzy consequents with different output intensities derived from the activated rules, and merge the output fuzzy sets into a unified overall output fuzzy distribution that can comprehensively reflect the influence of all rules by selecting the maximum value.

[0040] The overall output fuzzy distribution obtained from the above aggregation is defuzzified. The centroid of the fuzzy distribution is calculated using the centroid method and converted into a scalar value representing the fatigue stress level.

[0041] The scalar value is compared with a predefined range of fatigue stress level thresholds, and the fatigue stress level of the spatial element is determined based on the range in which it falls.

[0042] The vibration spectrum characteristics of the spatial units mapped to low or medium fatigue stress levels are extracted, the energy overshoot in the preset frequency band is analyzed, and the fatigue stress level improvement requirements and corresponding improvement levels of the spatial units are determined.

[0043] It should be noted that the preset frequency band specifically refers to a specific frequency range that is significantly related to the initiation and propagation process of microcracks in materials, determined through prior knowledge of the fatigue damage mechanism of materials or spectral analysis of historical failure cases. The following is an example of obtaining the preset frequency band through experimental calibration: By conducting accelerated fatigue tests on samples or in-service equipment, vibration data from the healthy state to the failure process are collected. Through spectral comparison analysis, one or more characteristic frequency bands with significantly abnormally high energy during the crack propagation stage are identified. The significant abnormal increase is analyzed in the process of analyzing the energy overshoot of the preset frequency band, that is, the difference between the monitored energy value of the preset frequency band and the preset upper limit energy value is calculated. The ratio of the difference to the preset upper limit energy value is used as the energy overshoot of the preset frequency band. When the energy overshoot is greater than the preset significant overshoot standard, it is determined that there is a significant abnormal increase attribute, and it is simultaneously determined that there is a need to increase the fatigue stress level of the spatial unit. If the energy overshoot of the preset frequency band is less than 1, it is determined that it can be increased by one level, and vice versa.

[0044] The fatigue stress level of each spatial unit after calibration is mapped to the preset test area design drawing of the wet brake to generate a fatigue stress distribution map.

[0045] This invention uses a multi-source monitoring array to collect real-time operating status information of wet brakes. Combined with working condition fusion analysis, the real-time operating data is converted into a fatigue stress distribution map that characterizes the stress state of the preset test area of ​​the wet brake. This helps to grasp the overall stress condition of the brake in real time and accurately locate different fatigue stress level areas.

[0046] Please see Figure 2 As shown in a preferred embodiment of the present invention, the failure warning signal identification process includes: continuously performing low-frequency energy monitoring on the vibration spectrum characteristics of each spatial unit, and performing thermal slope analysis on the time series of temperature data of each spatial unit to identify low-frequency energy mutation points and thermal slope anomalies.

[0047] It should be noted that the above-mentioned low-frequency energy mutation point identification process is as follows: the filtered time-domain signal is subjected to sliding window processing, and the energy integral within each time window is calculated in real time to form a time-varying energy integral sequence. The short-term time gradient of the energy integral sequence is calculated by first-order difference. When the short-term time gradient corresponding to a certain time window is greater than the first preset dynamic threshold, it is determined that a low-frequency energy mutation has occurred in that time window and is marked as a low-frequency energy mutation point. Since the sliding window processing is performed at the millisecond level, the time window can be regarded as a timestamp. The first preset dynamic threshold is dynamically generated based on the moving average and standard deviation of the recent historical energy gradient sequence, for example, the sum of the moving average and three times the standard deviation. The low-frequency band and the above-mentioned preset frequency band are the same frequency band, both corresponding to the dominant frequency of the stress wave excited by the propagation of microcracks in the material.

[0048] The process for identifying thermal slope anomalies is as follows: the second derivative of the time series of temperature data for each spatial unit is performed to obtain the thermal slope. If the absolute value of the thermal slope at a certain time point is greater than the second preset dynamic threshold, it is determined that the thermal slope has changed and is marked as a thermal slope anomaly. The principle of obtaining the second preset dynamic threshold is the same as that of obtaining the first preset dynamic threshold, and will not be elaborated here.

[0049] The low-frequency energy mutation point and the thermal slope anomaly point are correlated in time and space. When the time synchronization and spatial coincidence conditions are met at the same time, it is determined to be a failure precursor event.

[0050] Based on the fatigue stress level of the spatial unit where the failure precursor event is located, its fatigue damage stage is determined. The fatigue damage stage includes the critical failure stage, the intermediate decay stage, and the initial wear stage.

[0051] It should be noted that the criteria for determining the fatigue damage stage mentioned above are as follows: if the fatigue stress level of the spatial element where the precursor event of failure is located is high, it is determined to be the critical failure stage; if it is medium, it is determined to be the intermediate attenuation stage; and if it is low, it is determined to be the initial wear stage.

[0052] A structured failure warning signal that identifies the spatial unit where the failure precursor event is located, including a timestamp and a fatigue damage stage label.

[0053] S13. Based on the fatigue stress distribution diagram and the failure warning signal, coordinate and adjust the loading parameters of the loading device.

[0054] In a preferred embodiment of the present invention, the loading parameters of the coordinated control loading device include: for spatial units without failure warning signal indicators, if they belong to a high fatigue stress level, applying a low-frequency high-amplitude loading mode, reducing the loading load by a first preset proportion and extending the loading interval period to a preset duration.

[0055] If it belongs to a low fatigue stress level, a high-frequency, low-amplitude loading mode is applied and a random disturbance waveform is introduced to increase the loading load by a second preset ratio and simultaneously shorten the loading interval period.

[0056] It should be noted that the specific values ​​of the first and second preset ratios mentioned above can be calibrated through a limited number of tests based on the model, material properties, and testing standards of the wet brake being tested. As an illustrative rather than restrictive explanation, the first preset ratio is the load reduction ratio, which can typically be set in the range of 5% to 15%; the second preset ratio is the load increase ratio, which can typically be set in the range of 10% to 20%.

[0057] If it belongs to the fatigue stress level, then apply a medium-frequency medium-amplitude loading mode and introduce a periodic amplitude modulation stage.

[0058] It should be noted that for space units for which no failure warning signal has been detected, the core objective of the loading strategy is to actively and efficiently detect their fatigue performance boundaries and accelerate the evolution of potential damage, while ensuring that the testing process is always under control. The strategy is formulated based on the potential risk level revealed by their current fatigue stress level.

[0059] If the stress level is high, it indicates that although there are no signs of instability in the region, the static stress level is already extremely high, and the material's microstructure has accumulated a considerable amount of plastic deformation, reaching a critical point of accelerated damage accumulation. If conventional or higher-frequency loading is continued at this point, uncontrollable instantaneous failure can easily occur due to cyclic hardening / softening effects or thermo-coupling. Applying a low-frequency, high-amplitude loading mode is recommended. The high amplitude is to provide sufficient driving force to initiate and stably propagate microcracks, a necessary condition for probing the fatigue limit of the material in the high-stress region. The core purpose of the low frequency is to avoid the significant thermal effects caused by high-frequency vibrations, preventing a decrease in the material's yield strength due to local overheating, thus obscuring the failure mechanism of pure mechanical fatigue.

[0060] If the stress level is low, it indicates that the material in this region is in a state of primarily elastic deformation with extremely slow damage accumulation. Traditional constant amplitude loading in this region requires an extremely long testing period to induce failure, resulting in low testing efficiency and failing to reflect the randomness of loads in real-world operating conditions.

[0061] Applying a high-frequency, low-amplitude loading mode allows for a significant increase in the number of load cycles per unit time, thereby accelerating the accumulation of fatigue damage and effectively shortening the test cycle. The low amplitude ensures that the acceleration process remains within the elastic range of the material and does not introduce atypical plastic damage mechanisms. The introduction of random disturbance waveforms is to simulate uncertain loads in real service environments, assess the fatigue performance of materials under variable amplitude loads, and prevent overly optimistic conclusions from being drawn due to a single frequency loading.

[0062] If the region belongs to the fatigue stress level, it indicates that this area is the core region for stable accumulation of fatigue damage and is key to studying the fatigue life and crack propagation rate of materials. The strategic goal is to accurately characterize the damage evolution law and examine the load sequence effect while ensuring test stability.

[0063] Applying a medium-frequency, medium-amplitude loading mode ensures that damage accumulates at a stable and observable rate, facilitating the establishment of the core data segment in the stress-life curve. Medium frequency strikes a balance between testing efficiency and avoiding overheating, making it a common choice for industry standard testing. The introduction of a periodic amplitude modulation stage aims to periodically switch between high, medium, and low amplitudes to reveal whether microcracks have initiated and their propagation resistance, enabling online diagnosis without shutting down the system.

[0064] In a preferred embodiment of the present invention, the loading parameters of the coordinated control loading device further include: for a space unit carrying a failure warning signal identifier, if it belongs to a high fatigue stress level, triggering an immediate load reduction command and switching to a low-frequency, low-amplitude loading mode.

[0065] If the fatigue stress level is medium, the current loading parameters will be maintained, and the monitoring standards for this area will be upgraded accordingly.

[0066] If the stress level is low, maintain the current loading parameters and start the progressive loading mode.

[0067] It should be noted that the fatigue damage stage determination process reveals the specific working condition background of the failure event. For space units carrying failure warning signals, a differentiated and precise control strategy is implemented based on their current fatigue stress level. The logic is as follows: A high fatigue stress level indicates the emergence of microcrack instability propagation precursors in a region where stress concentration is already very severe and the material yield limit is approaching. This strongly suggests that the existing loading parameters have seriously exceeded the load-bearing capacity limit of this local area, and continued loading will very likely lead to catastrophic failure. Therefore, an instantaneous load reduction command with a millisecond response is triggered. This command directly controls the servo motor via a high-speed bus to rapidly reduce the loading amplitude, aiming to achieve instantaneous mechanical unloading of this high-risk area and force the rapid propagation of the fracture. At the same time, the loading mode of this area is switched to low-frequency, low-amplitude loading. The purpose of this mode is not to terminate the test, but to upgrade the test process from a rough destructive test to a refined observational test, extend the key observation window, and capture complete failure evolution data to significantly improve the test results.

[0068] A medium fatigue stress level indicates that the spatial unit is in a critical state. Its stress level is sufficient to trigger initial micro-damage, but has not yet reached the point of immediate unstable propagation. Pre-failure indicators at this stage suggest that fatigue damage is steadily accumulating in this area under current conditions, representing a potential critical failure point. In this state, blindly reducing the load would interrupt valuable failure evolution data acquisition, while blindly loading the load could accelerate its transition to a high-risk state. Therefore, the optimal strategy is to maintain the current loading parameters to preserve the continuity of the current operating conditions and continuously observe the evolution of damage under stable stress levels. This facilitates more accurate fatigue performance analysis. Simultaneously, the monitoring standards for this area should be improved, specifically by increasing the sampling frequency of vibration spectrum and temperature data, shortening the judgment time window for energy mutation and thermal slope analysis, and lowering the threshold for anomaly detection. The aim is to focus on monitoring this area with higher sensitivity and real-time performance, ensuring that signals of damage transitioning from stable accumulation to unstable propagation can be captured.

[0069] A low fatigue stress level signifies an isolated failure event signal within a generally stable region with a low background stress level. This typically does not indicate an imminent risk of overall structural failure, but rather suggests that existing loading parameters are effectively simulating complex conditions and may be revealing inherent microscopic defects or fatigue weaknesses in the material that are activated at low stress levels. To further efficiently explore the material's fatigue performance boundaries, a progressive loading mode is initiated. This mode does not increase the load indefinitely, but rather increases the preset loading amplitude at fixed intervals of a predetermined number of cycles or gradually increases the loading amplitude according to a step function. Throughout the loading process, vibration and temperature feedback in this region are monitored more closely to establish load-response curves. If any feedback indicator shows a non-linear increase or reappears as a precursor to failure, the loading process is immediately paused or reversed to ensure that all failure propagation experiments are conducted within a strictly controllable range, maximizing the data value of each test and accurately plotting the material's fatigue limit.

[0070] It should also be noted that the different loading modes in the above-mentioned coordinated loading device are all built-in fixed modes. The specific loading parameter combinations in the modes have been predefined and optimized in the method integration stage based on a large number of previous simulation calculations, material property calibration tests and industry standards, and have been solidified in the intelligent loading control strategy library.

[0071] The embodiments of the present invention dynamically adjust the loading parameters of the loading device through a dual feedback mechanism of fatigue stress distribution diagram and failure early warning signal, so that the loading conditions always match the actual failure risk scenario, avoiding the test deviation caused by the solidification of loading parameters in the prior art, and further improving the test efficiency and evaluation accuracy.

[0072] Please see Figure 3 As shown in S14, during the loading interval period of a single test cycle, process data is continuously collected, the fatigue performance response effect of the wet brake under the current loading strategy is analyzed, and the correction direction of the loading strategy for the next test cycle is determined.

[0073] In a preferred embodiment of the present invention, the process data includes: during the loading interval, activating a multi-source monitoring array to actively detect and scan the wet brake, acquiring the brake pressure distribution and temperature change status of a preset test area through pressure sensors and temperature sensors respectively, analyzing the pressure distribution reconstruction features and temperature gradient reconstruction features, and using them together as process data.

[0074] In a preferred embodiment of the present invention, the analysis of the fatigue performance response effect of the wet brake under the current loading strategy includes: marking the pressure distribution features and temperature gradient features extracted during the loading process as reference features.

[0075] By comparing the reconstructed features with the baseline features, the force distribution optimization rate and thermal gradient change rate of each spatial unit are quantified. The test load performance coefficient of each spatial unit is obtained based on linear weighted fusion calculation. The test load performance coefficient is used as a parameter to characterize the fatigue performance response effect of each spatial unit in the preset test area of ​​the wet brake under the current loading strategy.

[0076] It should be noted that the above-mentioned quantification processes for the force distribution optimization rate and the thermal gradient change rate are both: the difference between the reconstructed feature and the baseline feature is calculated as a ratio to the baseline feature.

[0077] The force distribution optimization rate directly reflects the effect of loading on the mechanical performance of the brake. The higher the force distribution optimization rate, the better the homogenization effect of the loading. The thermal gradient change rate reflects the degree of influence of loading on the thermal performance of the brake. The larger the absolute value of the thermal gradient change rate, the more significant the thermal effect of the loading.

[0078] The linear weighting of the force distribution optimization rate and the thermal gradient change rate can be set based on industry experience or obtained through a limited number of experimental data. For example, historical data on the force distribution optimization rate, thermal gradient change rate, and test load performance can be collected first. Then, the correlation coefficients between different loading effects and the influence of the force distribution optimization rate and thermal gradient change rate can be calculated. Regression analysis or logistic regression analysis can be used to determine their respective contributions. Finally, after normalization, the contributions can be converted into weights, and their sum is 1.

[0079] In a preferred embodiment of the present invention, the process of determining the correction direction of the loading strategy in the second test cycle includes: comparing the test load performance coefficient with a preset reasonable coefficient range, and identifying the weak space unit and overload space unit in the preset test area.

[0080] It should be noted that the aforementioned weakly effective space unit refers to a space unit whose test load performance coefficient is less than the lower limit of the preset reasonable coefficient range, while the overload space unit refers to a space unit whose test load performance coefficient is greater than the upper limit of the preset reasonable coefficient range.

[0081] A first correction instruction is generated and executed for the weak space cell. The first correction instruction includes increasing the loading amplitude of the corresponding loading device or shortening the loading interval period.

[0082] It should be noted that fatigue testing requires sufficient mechanical driving force to cause microscopic defects inside the material to evolve into observable microcracks. Weak spatial cells, due to insufficient loading energy input, fail to effectively stimulate the fatigue damage mechanism of the material, resulting in invalid or low-value data in this region. By increasing the loading amplitude or shortening the loading interval, manufacturing defects or material inhomogeneities that may exist in this region and are hidden under low loads can be actively exposed.

[0083] A second correction instruction is generated and executed for the overload space unit. The second correction instruction includes reducing the loading amplitude of the corresponding loading device or reducing the loading time.

[0084] It should be noted that continuing to apply high loading energy to overloaded space units is a waste of resources. The loading amplitude of the corresponding loading device should be reduced or the loading time should be shortened to reduce unnecessary power consumption, equipment wear and tear and maintenance costs.

[0085] It should also be noted that all execution parameters for overloaded and inefficient space units are derived from a quantitative assessment of the gap between real-time performance and target performance, and are dynamically generated through a pre-set, extensively calibrated and verified control rule base. This achieves a feedback-based, adaptive, and precise loading control, rather than a simple call to fixed parameters.

[0086] S15. Iteratively execute the test process until the preset test convergence conditions are met, and issue a fatigue performance test completion notification.

[0087] In a preferred embodiment of the present invention, the test convergence condition is that there are no weak space units or overload space units within the preset test area of ​​the wet brake.

[0088] This invention, through the analysis of the fatigue performance response of the wet brake under the current loading strategy during the loading interval period of a single test cycle, determines the correction direction of the loading strategy for the next test cycle, promotes the continuous optimization of the loading strategy during the test process until the test converges and ends, so that the test results can more accurately reflect the overall performance level of the wet brake and improve the accuracy of the test.

[0089] The above content is merely an example and illustration of the concept of the present invention. Those skilled in the art can make various modifications or additions to the specific embodiments described, or use similar methods to replace them, as long as they do not deviate from the concept of the invention or exceed the scope defined by the present invention, and all such modifications and additions should fall within the protection scope of the present invention.

Claims

1. A wet brake fatigue performance intelligent testing method, characterized in that, The method comprises the following steps: collecting real-time operation data of the wet brake under the load of the loading device during the test; the real-time operation data includes temperature gradient characteristics, vibration frequency spectrum characteristics and pressure distribution characteristics of the preset test area; performing working condition fusion analysis on the real-time operation data to generate a fatigue stress distribution map representing the stress state of the wet brake in the preset test area, and correspondingly identifying a failure warning signal representing the current fatigue damage stage; the working condition fusion analysis on the real-time operation data comprises: identifying a high-confidence working condition feature set by consistency verification of the pressure distribution characteristics and the temperature gradient characteristics; mapping the relative performance of the pressure distribution characteristics and the temperature gradient characteristics of each spatial unit in the feature set to a predefined fatigue stress level based on the fuzzy rule method, wherein the fatigue stress level includes low, medium or high; extracting the vibration frequency spectrum characteristics of the spatial unit mapped to the low or medium fatigue stress level, analyzing the energy amplitude of the preset frequency band, judging the fatigue stress level promotion requirement and the corresponding promotion level of the spatial unit; mapping the fatigue stress level of each spatial unit to the design drawing of the preset test area of the wet brake to generate a fatigue stress distribution map; the failure warning signal identification process comprises: identifying low-frequency energy mutation points and thermal slope abnormal points by continuously performing low-frequency band energy monitoring on the vibration frequency spectrum characteristics of each spatial unit and performing thermal slope analysis on the time sequence composed of the temperature data of each spatial unit; correlate the low-frequency energy mutation points and the thermal slope abnormal points in time and space, and determine it as a failure precursor event when the time synchronization and spatial coincidence conditions are met simultaneously; determine the fatigue damage stage of the spatial unit where the failure precursor event is located in combination with the fatigue stress level of the spatial unit, wherein the fatigue damage stage includes a critical failure stage, a medium-term decay stage and an initial wear stage; identifying a structured failure warning signal including a timestamp and a fatigue damage stage label for the spatial unit where the failure precursor event is located; adjusting and controlling the loading parameters of the loading device according to the fatigue stress distribution map and the failure warning signal; during the loading interval of a single test cycle, continuously collecting process data, analyzing the fatigue performance response effect of the wet brake under the current loading strategy, and determining the correction direction of the loading strategy of the next test cycle; iteratively execute the test process until the preset test convergence condition is met, and issue a fatigue performance test completion notification.

2. The intelligent test method for fatigue performance of a wet brake according to claim 1, characterized in that, The real-time operation data collection process comprises: arranging a multi-source monitoring array in the preset test area of the wet brake to synchronously trigger various sensors in the array to collect data with a unified clock, wherein the various sensors at least include temperature sensors, vibration sensors and pressure sensors; real-time collection of temperature data, vibration signal data and brake pressure data of each spatial unit in the preset test area; scene processing of the collected raw data to extract temperature gradient characteristics, vibration frequency spectrum characteristics and pressure distribution characteristics; integrate the extracted features into a vector form as real-time operation data.

3. The intelligent test method for fatigue performance of a wet brake according to claim 1, characterized in that, The adjustment and control of the loading parameters of the loading device comprises: For the spatial unit without failure warning signal identification, if it belongs to high fatigue stress level, a low-frequency high-amplitude loading mode is applied, the loading load is reduced by a first preset proportion, and the loading interval period is prolonged to a preset time length; If it belongs to low fatigue stress level, a high-frequency low-amplitude loading mode is applied and a random disturbance waveform is introduced, the loading load is increased by a second preset proportion, and the loading interval period is shortened synchronously; If it belongs to medium fatigue stress level, a medium-frequency medium-amplitude loading mode is applied and a periodic amplitude modulation link is introduced.

4. The intelligent test method for fatigue performance of a wet brake according to claim 3, characterized in that, The coordinated regulation of the loading parameters of the loading device also includes: For the spatial unit carrying the failure warning signal identification, if it belongs to high fatigue stress level, an immediate load reduction instruction is triggered and a low-frequency low-amplitude loading mode is switched; If it belongs to medium fatigue stress level, the current loading parameters are maintained, and the regional monitoring standard is simultaneously improved; If it belongs to low fatigue stress level, the current loading parameters are maintained and a gradual loading mode is started.

5. The intelligent test method for fatigue performance of a wet brake according to claim 1, characterized in that, The process data includes: During the loading intermittent period, a multi-source monitoring array is started to actively detect and scan the wet brake, the brake pressure distribution condition and temperature change condition of the preset test area are obtained by pressure sensors and temperature sensors respectively, the pressure distribution reconstruction feature and temperature gradient reconstruction feature are analyzed, and they are used as process data.

6. The intelligent test method for fatigue performance of a wet brake according to claim 5, characterized in that, The analysis of the fatigue performance response effect of the wet brake under the current loading strategy includes: The pressure distribution features and temperature gradient features extracted during the loading process are all marked as reference features; By comparing the reconstruction features with the reference features, the force distribution optimization rate and the thermal gradient change rate of each spatial unit are quantified, the test load performance coefficient of each spatial unit is calculated based on linear weighted fusion, and the test load performance coefficient is used as a parameter representing the fatigue performance response effect of each spatial unit in the preset test area of the wet brake under the current loading strategy.

7. The intelligent test method for fatigue performance of a wet brake according to claim 1, characterized in that, The correction direction determination process of the loading strategy of the secondary wheel test cycle includes: The test load performance coefficient is compared with a preset reasonable coefficient interval to identify weak-efficiency spatial units and overloaded spatial units in the preset test area; A first correction instruction is generated and executed for the weak-efficiency spatial units, the first correction instruction includes increasing the loading amplitude of the corresponding loading device or shortening the loading interval period; A second correction instruction is generated and executed for the overloaded spatial units, the second correction instruction includes reducing the loading amplitude of the corresponding loading device or reducing the loading time length.

8. The intelligent test method for fatigue performance of a wet brake according to claim 7, characterized in that, The test convergence condition is that there is no weak-efficiency spatial unit and overloaded spatial unit in the preset test area of the wet brake.

Images