Device and system for predicting service life of automobile universal joint

Abstract

The invention relates to the technical field of data processing, in particular to a device and a system for predicting the service life of an automobile universal joint, and aims to realize personalized prediction of the service life of the universal joint according to different driving habits of users. The device comprises a driving style analysis model which analyzes driving habits in combination with historical data and identifies the driving style of a user; the personalized service life prediction module is used for constructing a personalized service life prediction model and adaptively selecting the service life prediction model according to the driving style to perform service life prediction on the universal joint; wherein the personalized life prediction model comprises a first prediction model, a second prediction model and a third prediction model; and the abnormal mode detection module is used for monitoring the running state of the universal joint, optimizing the personalized life prediction model when an abnormal condition is detected, and dynamically correcting a life prediction result.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and specifically to a service life prediction device and system for an automotive universal joint. Background Art

[0002] A universal joint is a component commonly used in a transmission system, and its service life directly affects driving safety and performance. However, in most cases, users only notice a problem when the universal joint shows obvious damage. But in these cases of early damage, its service life has been greatly shortened.

[0003] With the rapid development of computer technology, technologies such as computer-aided design (CAD) and computer-aided engineering (CAE) have provided new ways for calculating the service life of universal joints, promoting the innovation of calculation methods. However, traditional methods for predicting the service life of universal joints cannot perform personalized prediction of the service life of universal joints according to different driving habits of users, resulting in insufficient prediction accuracy and affecting driving safety.

[0004] Therefore, a service life prediction device and system for an automotive universal joint are proposed. Summary of the Invention

[0005] The purpose of the present invention is to provide a service life prediction device and system for an automotive universal joint to achieve personalized prediction of the service life of the universal joint according to different driving habits of users. The device includes a driving style analysis model that analyzes driving habits in combination with historical data to identify the driving style of the user; a personalized life prediction module for constructing a personalized life prediction model and adaptively selecting a life prediction model according to the driving style to predict the life of the universal joint; an abnormal mode detection module for monitoring the operating state of the universal joint and adjusting the parameters of the personalized life prediction model when an abnormal situation is detected to dynamically correct the result of the life prediction; the system includes a driving style analysis unit, a personalized life prediction unit, and an abnormal mode detection unit.

[0006] To achieve the above object, the present invention provides the following technical solutions: A service life prediction device for an automotive universal joint, comprising: A driving style analysis module for analyzing driving habits in combination with historical data to identify the driving style of the user, including: smooth type, sporty type, and heavy-load type; The personalized life prediction module is used to build a personalized life prediction model and adaptively select a life prediction model according to the driving style to predict the life of the universal joint; the personalized life prediction model includes: a first prediction model, a second prediction model, and a third prediction model; if the driving style is the stable type, the first prediction model is used to perform the first life prediction based on the operating temperature, torque load, and rotational speed; if the driving style is the sporty type, the second prediction model is used to perform the second life prediction based on torque shock, instantaneous high temperature, and high-frequency vibration; if the driving style is the heavy-load type, the third prediction model is used to perform the third life prediction by monitoring load changes, axial angle changes, and lubrication loss; The abnormal mode detection module is used to monitor the operating state of the universal joint and optimize the parameters of the personalized life prediction model and dynamically correct the results of the life prediction when an abnormal situation is detected.

[0007] Preferably, the historical data includes: vehicle operation data, driving behavior data, vehicle working condition data, and maintenance record data; The vehicle operation data includes: load, rotational speed, axial angle, and operating temperature; The driving behavior data includes: acceleration, braking force, steering angle, and driving time distribution; The vehicle working condition data includes: lubrication state, ambient humidity, temperature, and road condition type; The maintenance record data includes: lubricating oil replacement cycle, universal joint replacement record, and abnormal repair record; Combining the historical data to analyze the driving habits and identifying the driving style of the user, the specific process is as follows: Using the driving style analysis model to process the historical data, analyze the driving habits of the user, and classify the driving style of the user.

[0008] Preferably, the personalized life prediction model includes: the first prediction model, the second prediction model, and the third prediction model; The first prediction model includes: a temperature influence layer, a torque load calculation layer, a rotational speed analysis layer, and a first life prediction layer; The temperature influence layer evaluates the influence of temperature on the life based on the long-term operating temperature of the universal joint to obtain a temperature influence factor; The torque load calculation layer calculates the influence of the long-term load on the life to obtain a torque load influence factor; The rotational speed analysis layer evaluates the influence of long-term high-speed operation on the life to obtain a rotational speed influence factor; The first remaining life of the universal joint is calculated by the first remaining life prediction layer in combination with the temperature influence factor, the torque load influence factor, and the rotational speed influence factor; the calculation formula for the first remaining life is: ; wherein, is the first remaining life; is the theoretical maximum life; is the temperature influence weight; is the temperature influence factor; is the temperature; is the torque load influence weight; is the torque load influence factor; is the torque load; is the rotational speed influence weight; is the rotational speed influence factor; is the rotational speed; is the interaction effect function of temperature, torque load, and rotational speed.

[0009] Preferably, the second prediction model includes: a torque impact monitoring layer, a temperature mutation detection layer, a vibration analysis layer, and a second remaining life prediction layer; The torque impact monitoring layer detects sudden torque impacts and analyzes the impact of the sudden torque impacts on the life; The temperature mutation detection layer identifies instantaneous high-temperature situations caused by aggressive driving and calculates the accelerated impact of the instantaneous high-temperature situations on the life; The vibration analysis layer monitors the amplitude and frequency of high-frequency vibrations during high-frequency operation of the universal joint and analyzes the wear impact of the high-frequency vibrations on the universal joint; The second remaining life prediction layer calculates the second remaining life prediction result by integrating the torque impact monitoring layer, the temperature mutation detection layer, and the vibration analysis layer.

[0010] Preferably, the third prediction model includes: a load change analysis layer, an axial angle monitoring layer, a lubrication loss assessment layer, and a third remaining life prediction layer; The load change analysis layer analyzes the long-term high-load situations borne by the universal joint and evaluates the impact of the high-load state on the life of the universal joint; The axial angle monitoring layer monitors the angle change of the universal joint and calculates the impact of the angle change on the working state of the universal joint under high-load conditions; The lubrication loss assessment layer assesses the lubricating oil state and loss situation of the universal joint and estimates the impact on the life of the universal joint by evaluating the lubricating oil loss situation under high-load, frequent starting, and braking conditions; The third remaining life prediction layer combines the data of the load change analysis layer, the axial angle monitoring layer, and the lubrication loss evaluation layer to calculate the third remaining life of the universal joint under heavy-duty driving.

[0011] Preferably, the process of the abnormal mode detection module includes: an abnormal data acquisition layer, an abnormal mode recognition layer, and a dynamic correction prediction layer; The abnormal data acquisition layer monitors the temperature, load, rotational speed, vibration, and lubrication status in real time through sensors and compares them with the normal operating state; The abnormal mode recognition layer uses machine learning algorithms to detect abnormal events within a short period of time, marks them as abnormal modes, and triggers an alarm; the abnormal modes include: sudden impact, high temperature, and overload; When the dynamic correction prediction layer detects the abnormal mode, it adjusts the parameters of the personalized life prediction model and dynamically corrects the life prediction value.

[0012] A service life prediction system for an automotive universal joint, comprising: A driving style analysis unit, configured to analyze driving habits by combining historical data, and identify the driving style of the user, including: smooth type, sporty type, and heavy-load type; A personalized life prediction unit, configured to construct a personalized life prediction model, and adaptively select a life prediction model according to the driving style to predict the life of the universal joint; the personalized life prediction model includes: a first prediction model, a second prediction model, and a third prediction model; if the driving style is the smooth type, use the first prediction model to perform a first life prediction based on the operating temperature, torque load, and rotational speed; if the driving style is the sporty type, perform a second life prediction through the second prediction model based on torque impact, instantaneous high temperature, and high-frequency vibration; if the driving style is the heavy-load type, use the third prediction model to perform a third life prediction by monitoring load changes, axial angle changes, and lubrication losses; An abnormal mode detection unit, configured to monitor the operating state of the universal joint, and optimize the parameters of the personalized life prediction model and dynamically correct the result of the life prediction when an abnormal situation is detected.

[0013] Compared with the prior art, the beneficial effects of the present invention are: 1. The present invention proposes a driving style analysis model that accurately identifies the driving style of the user by combining historical data, and can effectively distinguish the influence of different driving behaviors on the life of the universal joint. By adopting dynamic data analysis technology and taking into account actual driving behaviors (such as acceleration, braking force, steering angle, etc.), it can more accurately analyze the actual wear of the universal joint under different driving styles, providing a personalized basis for subsequent personalized life prediction according to the driving style, thereby improving the accuracy of life prediction.

[0014] 2. The present invention proposes a personalized life prediction model. According to the driving type data provided by the driving style analysis model, a suitable life prediction model (such as different models for steady type, sporty type, and heavy - duty type) is selected to perform personalized life prediction. For example, the driving habits of sporty drivers usually result in higher impact loads, while heavy - duty driving leads to higher continuous loads. This model can personalizedly select the life prediction method according to the driving style, improving the accuracy of the universal joint service life prediction.

[0015] 3. The present invention dynamically corrects the parameters of the life prediction model by real - time monitoring the operating state of the universal joint and detecting abnormal conditions. Through continuous monitoring of the temperature, load, rotational speed, vibration, and lubrication state of the universal joint, combined with machine learning algorithms, this module can accurately identify sudden abnormal patterns (such as sudden impacts, high temperatures, and overloading), and timely adjust the prediction results. This real - time correction mechanism greatly improves the accuracy and reliability of the prediction. Especially when dealing with unforeseen driving environments or unexpected events, it can prevent incorrect predictions and premature / late replacement operations. This method has dynamic adaptability and automatic adjustment capabilities, can accurately predict the life of the universal joint under abnormal working conditions, thus reducing maintenance costs, improving safety, and reducing the occurrence of maintenance accidents. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 It is a schematic structural diagram of a device for predicting the service life of an automotive universal joint provided by an embodiment of the present invention; Figure 2 It is a working principle diagram of a system for predicting the service life of an automotive universal joint provided by an embodiment of the present invention; Figure 3 It is a schematic structural diagram of a personalized prediction model provided by an embodiment of the present invention; Figure 4 It is a working principle diagram of an abnormal pattern detection module provided by an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0017] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0018] An automotive universal joint is a component commonly used in the transmission system. It is mainly used to connect two shafts that are not on the same axis, allowing them to transmit power at different angles. Its lifespan directly affects the driving safety and performance of the vehicle. However, in most cases, it is only when the universal joint shows obvious damage (such as abnormal noise, vibration, etc.) that the vehicle owner will notice the problem. But in these cases of early damage, the lifespan has been greatly shortened.

[0019] The present invention proposes a device and system for predicting the service life of an automotive universal joint, which realizes personalized prediction of the service life of the universal joint according to different driving habits of users, thereby improving the prediction accuracy of the service life of the universal joint. In order to illustrate that the method of the present invention can play a role in personalized prediction of the service life of the universal joint according to different driving habits of users, thereby improving the prediction accuracy of the service life of the universal joint, the effectiveness of the present invention will be described from two embodiments below.

[0020] Embodiment 1 In the embodiment of the present application, the process of personalized prediction of the service life of the universal joint according to different driving habits of users by using the method proposed by the present invention is described in detail. The following is based on Figure 1 the content to describe the process of predicting the service life of the automotive universal joint in detail; among them, Figure 1 is a schematic structural diagram of the service life prediction device proposed by the present invention, including: a driving style analysis module, which is used to analyze driving habits in combination with historical data to identify the driving style of the user; a personalized life prediction module, which is used to construct a personalized life prediction model and adaptively select a life prediction model according to the driving style to predict the life of the universal joint; an abnormal mode detection module, which is used to monitor the operating state of the universal joint and adjust the parameters of the personalized life prediction model when an abnormal situation is detected, and dynamically correct the result of the life prediction. Figure 2 is a specific structural diagram of the system proposed by the present invention, including: a driving style analysis unit, a personalized life prediction unit, and an abnormal mode detection unit. Combining Figure 1 and Figure 2 the content in the following description is as follows: A device for predicting the service life of an automotive universal joint, as Figure 1 shown, includes: A driving style analysis module, which is used to analyze driving habits in combination with historical data to identify the driving style of the user, including: smooth type, sports type, and load-carrying type; The historical data includes: vehicle operation data, driving behavior data, vehicle working condition data, and maintenance record data; The vehicle operation data includes: load, rotational speed, axial angle, and operating temperature; The driving behavior data includes: acceleration, braking force, steering angle, and driving time distribution; The vehicle operating condition data includes: lubrication state, ambient humidity, temperature, and road condition type; The maintenance record data includes: lubricating oil replacement cycle, universal joint replacement record, and abnormal repair record; Combining the historical data to analyze driving habits and identify the driving style of the user. The specific process is as follows: Use the driving style analysis model to process the historical data, analyze the driving habits of the user, and classify the driving style of the user.

[0021] Specifically, the driving style analysis model includes: a feature extraction layer and a driving style classification layer; The feature extraction layer extracts driving style features according to the historical data, including: average acceleration, acceleration variance, braking force distribution, axial angle change frequency, and long-term load condition; The driving style classification layer classifies the driving style through machine learning algorithms (such as K-means clustering, decision tree, or neural network).

[0022] Table 1 gives the driving style classification result table (g is the acceleration due to gravity).

[0023] Table 1 Driving Style Classification Results

[0024] It can be seen from Table 1 that vehicles A and D have lower acceleration, braking force, and rotational speed, so they are classified as steady driving; vehicles B and E have higher acceleration and rotational speed, and greater braking force, so they are classified as sporty driving; vehicle C has a higher load, lower rotational speed, but a larger axial angle change, so it is classified as load-carrying driving.

[0025] In the embodiment of the present application, through the driving style analysis model, combined with vehicle operation data, driving behavior data, vehicle operating condition data, and maintenance record data, the key factors affecting the life of the universal joint are comprehensively extracted, and the features such as average acceleration, acceleration variance, braking force distribution, axial angle change frequency, and long-term load condition are calculated by the feature extraction layer, so as to accurately depict the driving habits of the driver. Through the driving style classification layer, machine learning algorithms such as K-means clustering, decision tree, or neural network are used to intelligently classify the driving style, which can accurately distinguish steady, sporty, and load-carrying driving styles. This driving style analysis module can more comprehensively and accurately adapt to different driving environments, enable the personalized life prediction model to adaptively adjust according to different driving styles, improve the accuracy of universal joint life prediction, reduce the risk of abnormal wear, and thus enhance the safety and reliability of vehicle operation.

[0026] Preferably, asFigure 3 As shown, the personalized life prediction module is used to build a personalized life prediction model and adaptively select a life prediction model according to the driving style to predict the life of the universal joint. If the driving style is the stable type, the first prediction model is used to perform the first life prediction based on the operating temperature, torque load, and rotational speed. If the driving style is the sports type, the second prediction model is used to perform the second life prediction based on torque shock, instantaneous high temperature, and high-frequency vibration. If the driving style is the heavy-load type, the third prediction model is used to perform the third life prediction by monitoring load changes, axial angle changes, and lubrication loss. The personalized life prediction model includes: the first prediction model, the second prediction model, and the third prediction model. The first prediction model includes: a temperature influence layer, a torque load calculation layer, a rotational speed analysis layer, and a first life prediction layer. The temperature influence layer evaluates the influence of temperature on life based on the long-term operating temperature of the universal joint to obtain a temperature influence factor. The torque load calculation layer calculates the influence of long-term load on life to obtain a torque load influence factor. The rotational speed analysis layer evaluates the influence of long-term high-speed operation on life to obtain a rotational speed influence factor. The first life prediction layer combines the temperature influence factor, the torque load influence factor, and the rotational speed influence factor to calculate the first remaining life of the universal joint. The calculation formula for the first remaining life is: ; Where, is the first remaining life; is the theoretical maximum life; is the temperature influence weight; is the temperature influence factor; is the temperature; is the torque load influence weight; is the torque load influence factor; is the torque load; is the rotational speed influence weight; is the rotational speed influence factor; is the rotational speed; is the interaction effect function of temperature, torque load, and rotational speed, representing the interaction between temperature, torque load, and rotational speed; is the normal temperature; is the high temperature threshold; is the standard torque load; is the standard rotational speed; 、 、 、 、 and are empirical adjustment coefficients; is the interaction effect of temperature and torque load; is the interaction effect of temperature and rotational speed; is the interaction effect of torque load and rotational speed.

[0027] The embodiment of the present application adaptively selects a suitable life prediction model based on the user's driving style to improve the accuracy of the universal joint life prediction. Among them, the first prediction model is a life prediction model based on a steady driving style. Through the temperature influence layer, torque load calculation layer and rotational speed analysis layer, the influences of temperature, load and rotational speed on the life of the universal joint under steady driving conditions are calculated respectively, and corresponding weight factors are used for quantification, so that the prediction result is more in line with the actual working conditions. At the same time, this model introduces an interaction effect function between temperature, torque load and rotational speed to comprehensively evaluate the coupling effect between different factors, making up for the limitation of traditional prediction methods that only consider the influence of a single factor. By optimizing the calculation formula, this method can not only more accurately predict the first remaining life of the universal joint under steady driving conditions, but also provide more targeted maintenance suggestions for the vehicle, extend the service life of key components, and improve the operation reliability and safety of the whole vehicle.

[0028] Preferably, the second prediction model includes: a torque impact monitoring layer, a temperature mutation detection layer, a vibration analysis layer and a second life prediction layer; The torque impact monitoring layer detects sudden torque impacts and analyzes the influence of the sudden torque impacts on the life; The temperature mutation detection layer identifies instantaneous high-temperature situations caused by intense driving and calculates the accelerated influence of the instantaneous high-temperature situations on the life; The vibration analysis layer monitors the amplitude and frequency of high-frequency vibrations when the universal joint operates at a high frequency and analyzes the wear influence of the high-frequency vibrations on the universal joint; The second life prediction layer calculates the second remaining life prediction result by comprehensively considering the torque impact monitoring layer, the temperature mutation detection layer and the vibration analysis layer.

[0029] Specifically, the torque impact monitoring layer uses a high-precision torque sensor installed near the universal joint to detect sudden torque impacts caused by sudden acceleration, sudden braking or sharp turning during driving in real time; when the sensor detects a torque peak exceeding the preset threshold, the data recording unit stores the data and filters out noise through a signal processing algorithm to extract effective impact amplitude and duration information; The temperature mutation detection layer monitors the instantaneous temperature change on the surface of the universal joint through a fast-response temperature sensor; when the vehicle experiences a rapid temperature rise during intense driving, it immediately captures the temperature mutation data and calculates the accelerated influence of the instantaneous high temperature on the material fatigue of the universal joint and the performance of the lubricating oil by using the temperature change rate algorithm; The vibration analysis layer uses a high-sampling-rate accelerometer and spectrum analysis technology to monitor the high-frequency vibration signals generated by the universal joint during high-speed operation. By extracting the amplitude, frequency distribution, and duration of the vibration, it evaluates the local wear and fatigue damage caused by continuous high-frequency vibration, and performs smoothing and feature extraction processing on the vibration data; The second remaining life prediction layer comprehensively uses the torque impact monitoring layer, the temperature mutation detection layer, and the vibration data obtained by the vibration analysis layer, and uses a mathematical model based on the fatigue accumulation theory (such as Miner's rule) to calculate the second remaining life of the universal joint.

[0030] In the embodiment of the present application, the second prediction model realizes the accurate prediction of the life of the universal joint under severe driving conditions (sports driving style) through the collaborative work of the torque impact monitoring layer, the temperature mutation detection layer, the vibration analysis layer, and the second remaining life prediction layer. The torque impact monitoring layer can detect in real time the torque impacts caused by operations such as sudden acceleration, sudden braking, and sharp turning, extract the impact amplitude and duration, and accurately evaluate their impact on the fatigue life of the universal joint; the temperature mutation detection layer uses a fast-response temperature sensor to identify the instantaneous high temperature caused by severe driving, and calculates the accelerated impact of the high temperature on the material properties and lubrication state of the universal joint through a temperature change rate algorithm; the vibration analysis layer uses a high-sampling-rate accelerometer and spectrum analysis technology to accurately identify the vibration characteristics of the universal joint during high-frequency operation, evaluate the impact of high-frequency vibration on local wear, and perform data optimization processing; the second remaining life prediction layer is based on the fatigue accumulation theory, comprehensively analyzes the losses of the universal joint caused by torque impact, temperature mutation, and vibration, and calculates the second remaining life. This model can effectively improve the accuracy of predicting the life of the universal joint under severe driving conditions, reduce the risk of accidental damage caused by sudden working conditions, and improve the safety of vehicle operation and the scientific nature of maintenance.

[0031] Preferably, the third prediction model includes: a load change analysis layer, an axial angle monitoring layer, a lubrication loss evaluation layer, and a third remaining life prediction layer; The load change analysis layer analyzes the long-term high-load conditions borne by the universal joint and evaluates the impact on the life of the universal joint under high-load conditions; The axial angle monitoring layer monitors the angle change of the universal joint and calculates the impact of the angle change on the working state of the universal joint under high-load conditions; The lubrication loss evaluation layer evaluates the lubricating oil state and loss situation of the universal joint, and estimates the impact on the life of the universal joint by evaluating the lubricating oil loss situation under high load, frequent starting, and braking; The third remaining life prediction layer combines the data of the load change analysis layer, the axial angle monitoring layer, and the lubrication loss evaluation layer to calculate the third remaining life of the universal joint under heavy-load driving.

[0032] Specifically, the load change analysis layer uses high-precision strain sensors installed on the drive shaft to monitor the load conditions borne by the universal joint in real time; records the torque changes of the vehicle under different load states, and uses the long-term load trend analysis method to calculate the impact of high loads on the life of the universal joint; combines the vehicle driving data to analyze the fluctuation range and distribution characteristics of the load changes, so as to evaluate the impact of additional stresses caused by uneven load changes; The axial angle monitoring layer uses angle sensors and gyroscopes to measure the axial angle changes of the universal joint in real time, analyzes the axial angle change trend under high-load conditions based on statistical analysis methods, and combines the vehicle operation characteristics to evaluate the impact of axial angle changes on the lubrication effect and friction and wear; the lubrication loss evaluation layer detects the temperature, viscosity, impurity content and oxidation degree of the lubricating oil in real time, and combines the vehicle operation data to evaluate the deterioration trend of the lubricating oil and analyze the impact of the lubrication state on the service life of the universal joint; The third remaining life prediction layer synthesizes the data of the load change analysis layer, the axial angle monitoring layer and the lubrication loss evaluation layer, and uses big data statistical analysis and machine learning algorithms (such as regression analysis or deep neural networks) to predict the third remaining life. By performing pattern recognition and trend fitting on historical data, a non-linear prediction model of the universal joint life change under the heavy-duty driving style is established, the third remaining life is calculated, and maintenance suggestions are given.

[0033] The third prediction model proposed in the embodiment of the present application aims at the working condition characteristics of the universal joint under the heavy-duty driving style, and realizes accurate life prediction and maintenance suggestions. Through the load change analysis layer, the load conditions of the universal joint are monitored in real time, and the impact of long-term high loads on the life is evaluated, making the prediction results more targeted; the axial angle monitoring layer combines the data analysis of the angle sensor and the gyroscope to analyze the angle change of the universal joint and evaluate its impact on the lubrication state and friction and wear, improving the adaptability of the prediction model to different working states; the lubrication loss evaluation layer monitors various parameters of the lubricating oil to evaluate the deterioration trend of the lubricating oil, which helps to detect in advance the risk of shortened life caused by insufficient lubrication; the third remaining life prediction layer synthesizes the above multi-source data, uses big data statistical analysis and machine learning algorithms to achieve accurate life prediction under heavy-duty working conditions, and provides an optimized maintenance strategy to avoid unnecessary losses or safety hazards caused by premature or late maintenance, thereby improving the service life and reliability of the universal joint.

[0034] Table 2 shows the improvement effect of the personalized life prediction model on the prediction accuracy of the universal joint service life compared with the single prediction model.

[0035] Table 2 Improvement Table of Prediction Accuracy of Personalized Life Prediction Model

[0036] Preferably, asFigure 4 As shown, an abnormal mode detection module is used to monitor the operating state of the universal joint and adjust the parameters of the personalized life prediction model when an abnormal situation is detected, dynamically correcting the result of life prediction.

[0037] The process of the abnormal mode detection module includes: an abnormal data acquisition layer, an abnormal mode recognition layer, and a dynamic correction prediction layer; The abnormal data acquisition layer monitors the temperature, load, rotational speed, vibration, and lubrication state in real time through sensors and compares them with the normal operating state; The abnormal mode recognition layer uses machine learning algorithms to detect abnormal events within a short period of time, marks them as abnormal modes, and triggers an alarm; the abnormal modes include: sudden shock, high temperature, and overload; When the dynamic correction prediction layer detects the abnormal mode, it adjusts the parameters of the personalized life prediction model and dynamically corrects the life prediction value.

[0038] Specifically, the abnormal data acquisition layer monitors the operating temperature on the surface and inside of the universal joint through a temperature sensor and compares it with the normal temperature range; Measures the torque load of the universal joint under different working conditions through a load sensor to determine whether there is abnormal overload; Monitors the rotational speed of the universal joint through a rotational speed sensor to determine whether there is abnormal rapid acceleration or deceleration; Detects high-frequency vibration through an acceleration sensor to identify the impact of sudden shock on the universal joint; Measures the viscosity, temperature, and impurity content of the lubricating oil through a lubrication state sensor to analyze abnormal changes in the lubrication state; The abnormal mode recognition layer uses machine learning algorithms (such as support vector machine SVM, random forest, or deep learning neural network) to analyze abnormal data within a short period of time, identify, and mark abnormal modes; When the acceleration sensor detects an instantaneous shock exceeding the set threshold (such as 50g, where g is the acceleration due to gravity), it is marked as the sudden shock abnormal mode; When the temperature sensor detects that the temperature exceeds the normal range (such as exceeding 120°C), it is marked as the high temperature abnormal mode; When the load sensor detects that the load continuously exceeds the rated load (such as more than 120% of the rated torque), it is marked as the overload abnormal mode; When the abnormal mode is recognized, an alarm is immediately generated, and the abnormal data is transmitted to the dynamic correction prediction layer; The dynamic correction prediction layer adjusts the parameters of the personalized life prediction model according to the abnormal type, such as: If an instantaneous torque shock is detected, the impact factor weight is increased in the life prediction model to enhance the impact of impact damage on the life. If the temperature exceeds the normal range, the temperature impact factor is adjusted to enhance the impact of temperature on the life and reduce the predicted remaining life value. If the universal joint continuously operates under overload, the predicted life value is dynamically reduced, and at the same time, the user is advised to reduce the overload operation.

[0039] Table 3 shows the life prediction correction result table of the abnormal mode detection module. Among them, the predicted life before correction is the theoretical life prediction value, and the predicted life after correction takes into account the influence of abnormal working conditions.

[0040] Table 3 Life Prediction Correction Result Table

[0041] In the embodiment of the present application, by real-time monitoring the temperature, load, rotation speed, vibration and lubrication state of the universal joint and comparing with the normal operation data, abnormal conditions such as sudden shocks, high temperatures or overloads caused by aggressive driving can be accurately identified. Once an abnormal mode is detected, the system uses machine learning algorithms such as support vector machines, random forests or deep learning to quickly analyze and label the abnormal data, generate an alarm, and at the same time transmit the abnormal information to the dynamic correction prediction layer. This layer automatically adjusts the key parameters in the personalized life prediction model according to the type of abnormality, such as increasing the impact factor weight, adjusting the temperature impact factor or reducing the load parameter, so as to correct the predicted remaining life value in real time. This technical solution greatly improves the accuracy and real-time performance of the universal joint life prediction, making the prediction result better reflect the actual working condition changes, providing a scientific and timely decision-making basis for vehicle maintenance and fault prevention, effectively reducing the damage risk caused by abnormal working conditions and enhancing the overall operation safety and economy of the vehicle.

[0042] The device for predicting the service life of an automotive universal joint provided by the embodiments of the present application can combine a driving style analysis module, a personalized life prediction module, and an abnormal mode detection module to achieve personalized prediction of the service life of the universal joint according to different user driving habits, thereby improving the accuracy of predicting the service life of the universal joint. Through the driving style analysis module, the system can identify the driving habits of the driver and select the corresponding life prediction model based on different driving styles, thereby improving the pertinence and accuracy of the prediction. The personalized life prediction module combines key factors such as operating temperature, torque load, torque impact, high-frequency vibration, and lubrication loss to construct personalized life prediction models suitable for smooth, sporty, and heavy-duty driving styles, making the life prediction results more in line with reality. Finally, the abnormal mode detection module monitors the operating state of the universal joint in real time. In the case of sudden shocks, high temperatures, or overloading abnormalities, it dynamically adjusts the parameters of the life prediction model to ensure that the personalized prediction results are accurately corrected as the actual working conditions change. This device effectively improves the accuracy and reliability of the universal joint life prediction, provides a scientific basis for vehicle maintenance and fault warning, thereby reducing the risk of accidental damage and improving the safety and economy of vehicle operation.

[0043] Embodiment 2 In Embodiment 1, the device proposed by the present invention successfully achieved personalized prediction of the service life of the universal joint according to different user driving habits, thereby improving the accuracy of predicting the service life of the universal joint. To further verify the effectiveness of the present invention, the service life of the automotive universal joint of a certain universal joint company was also predicted in the embodiments of the present application.

[0044] A service life prediction system for an automotive universal joint, as Figure 2 shown, includes: A driving style analysis unit for analyzing driving habits in combination with historical data to identify the driving style of the user, including: smooth type, sporty type, and heavy-duty type; The historical data includes: vehicle operation data, driving behavior data, vehicle working condition data, and maintenance record data; The vehicle operation data includes: load, rotational speed, axial angle, and operating temperature; The driving behavior data includes: acceleration, braking force, steering angle, and driving time distribution; The vehicle working condition data includes: lubrication state, ambient humidity, temperature, and road condition type; The maintenance record data includes: lubricating oil replacement cycle, universal joint replacement record, and abnormal repair record; Analyzing driving habits in combination with the historical data to identify the driving style of the user, the specific process is: Process the historical data using the driving style analysis model, analyze the driving habits of the user, and classify the driving style of the user.

[0045] Preferably, a personalized life prediction unit is configured to build a personalized life prediction model and adaptively select a life prediction model according to the driving style to predict the life of the universal joint; if the driving style is the stable type, use the first prediction model to perform a first life prediction based on the operating temperature, torque load, and rotational speed; if the driving style is the sporty type, perform a second life prediction based on torque shock, instantaneous high temperature, and high-frequency vibration through the second prediction model; if the driving style is the heavy-load type, use the third prediction model to perform a third life prediction by monitoring load changes, axial angle changes, and lubrication loss; The personalized life prediction model includes: the first prediction model, the second prediction model, and the third prediction model; The first prediction model includes: a temperature influence layer, a torque load calculation layer, a rotational speed analysis layer, and a first life prediction layer; The temperature influence layer evaluates the influence of temperature on life based on the long-term operating temperature of the universal joint to obtain a temperature influence factor; The torque load calculation layer calculates the influence of long-term load on life to obtain a torque load influence factor; The rotational speed analysis layer evaluates the influence of long-term high-speed operation on life to obtain a rotational speed influence factor; The first life prediction layer combines the temperature influence factor, the torque load influence factor, and the rotational speed influence factor to calculate the first remaining life of the universal joint; the calculation formula for the first remaining life is: ; Wherein, is the first remaining life; is the theoretical maximum life; is the temperature influence weight; is the temperature influence factor; is the temperature; is the torque load influence weight; is the torque load influence factor; is the torque load; is the rotational speed influence weight; is the rotational speed influence factor; is the rotational speed; is the interaction effect function of temperature, torque load, and rotational speed, representing the interaction between temperature, torque load, and rotational speed.

[0046] Preferably, the second prediction model includes: a torque shock monitoring layer, a temperature mutation detection layer, a vibration analysis layer, and a second life prediction layer; The torque impact monitoring layer detects sudden torque impacts and analyzes the impact of the sudden torque impacts on the lifespan. The temperature mutation detection layer identifies instantaneous high-temperature situations caused by aggressive driving and calculates the accelerating impact of the instantaneous high-temperature situations on the lifespan. The vibration analysis layer monitors the amplitude and frequency of high-frequency vibrations during high-frequency operation of the universal joint and analyzes the wear impact of the high-frequency vibrations on the universal joint. The second lifespan prediction layer calculates a second remaining lifespan prediction result by integrating the torque impact monitoring layer, the temperature mutation detection layer, and the vibration analysis layer.

[0047] Preferably, the third prediction model includes: a load change analysis layer, an axial angle monitoring layer, a lubrication loss assessment layer, and a third lifespan prediction layer. The load change analysis layer analyzes the long-term high-load situations borne by the universal joint and evaluates the impact of the high-load state on the lifespan of the universal joint. The axial angle monitoring layer monitors the angle changes of the universal joint and calculates the impact of the angle changes on the working state of the universal joint under high-load conditions. The lubrication loss assessment layer assesses the lubricating oil state and loss situation of the universal joint and estimates the impact on the lifespan of the universal joint by evaluating the lubricating oil loss situation under high-load, frequent starting, and braking conditions. The third lifespan prediction layer combines the data of the load change analysis layer, the axial angle monitoring layer, and the lubrication loss assessment layer to calculate the third remaining lifespan of the universal joint under heavy-duty driving.

[0048] Preferably, an abnormal mode detection unit is used to monitor the operating state of the universal joint and adjust the parameters of the personalized lifespan prediction model when an abnormal situation is detected, dynamically correcting the result of the lifespan prediction.

[0049] The process of the abnormal mode detection unit includes: an abnormal data acquisition layer, an abnormal mode recognition layer, and a dynamic correction prediction layer. The abnormal data acquisition layer monitors the temperature, load, rotational speed, vibration, and lubrication state in real time through sensors and compares them with the normal operating state. The abnormal mode recognition layer uses machine learning algorithms to detect abnormal events within a short period of time, marks them as abnormal modes, and triggers an alarm; the abnormal modes include: sudden impact, high temperature, and overload. When the dynamic correction prediction layer detects the abnormal mode, it adjusts the parameters of the personalized lifespan prediction model and dynamically corrects the lifespan prediction value.

[0050] Table 4 gives the universal joint service life prediction result table proposed in this application.

[0051] Table 4 Prediction Results Table of Universal Joint Service Life

[0052] Although the embodiments of the present invention have been shown and described, those of ordinary skill in the art can understand that various changes, modifications, substitutions, and variations can be made to these embodiments without departing from the principles and spirit of the present invention. The scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A service life prediction device for a vehicle universal joint, characterized in that: include: Driving style analysis module, which is used to analyze driving habits based on historical data and identify the user's driving style, including: stable, sporty and heavy; A personalized life prediction module, used to construct a personalized life prediction model, and adaptively select a life prediction model to predict the life of the universal joint according to the driving style; the personalized life prediction model includes: a first prediction model, a second prediction model and a third prediction model; if the driving style is the stable type, the first prediction model is used to perform a first life prediction based on the operating temperature, torque load and speed; if the driving style is the sporty type, the second prediction model is used to perform a second life prediction based on torque impact, instantaneous high temperature and high-frequency vibration; if the driving style is the load-bearing type, the third prediction model is used to perform a third life prediction by monitoring load changes, axial angle changes and lubrication loss; The abnormal mode detection module is used to monitor the operating status of the universal joint, and when an abnormal situation is detected, optimize the personalized life prediction model and dynamically correct the result of the life prediction.

2. The service life prediction device of a vehicle universal joint according to claim 1, characterized in that: The historical data includes: vehicle operation data, driving behavior data, vehicle working condition data and maintenance record data; the vehicle operation data includes: load, speed, axial angle and operating temperature; the driving behavior data includes: acceleration, braking force, steering angle and driving time distribution; the vehicle working condition data includes: lubrication status, ambient humidity, temperature and road condition type; the maintenance record data includes: lubricant replacement cycle, universal joint replacement record and abnormal maintenance record; The driving habits are analyzed in combination with the historical data to identify the user's driving style. The specific process is: using a driving style analysis model to process the historical data, analyze the user's driving habits, and classify the user's driving style.

3. The service life prediction device of a vehicle universal joint according to claim 1, characterized in that: The first prediction model includes: a temperature influence layer, a torque load calculation layer, a speed analysis layer and a first life prediction layer; the temperature influence layer evaluates the influence of temperature on life based on the long-term operating temperature of the universal joint to obtain a temperature influence factor; the torque load calculation layer calculates the influence of long-term load on life to obtain a torque load influence factor; the speed analysis layer evaluates the influence of long-term high-speed operation on life to obtain a speed influence factor; the first life prediction layer combines the temperature influence factor, the torque load influence factor and the speed influence factor to calculate the first remaining life of the universal joint.

4. The service life prediction device of a vehicle universal joint according to claim 1, characterized in that: The second prediction model includes: a torque impact monitoring layer, a temperature mutation detection layer, a vibration analysis layer, and a second life prediction layer; The torque impact monitoring layer detects sudden torque impacts and analyzes the impact of the sudden torque impacts on life; the temperature mutation detection layer identifies instantaneous high temperature conditions caused by intense driving and calculates the accelerated impact of instantaneous high temperature on life; the vibration analysis layer monitors the amplitude and frequency of high-frequency vibrations of the universal joint when it runs at high frequency, and analyzes the impact of the high-frequency vibrations on the wear of the universal joint; the second life prediction layer calculates the second remaining life prediction result by integrating the results of the torque impact monitoring layer, the temperature mutation detection layer and the vibration analysis layer.

5. The service life prediction device of a vehicle universal joint according to claim 1, characterized in that: The third prediction model includes: a load change analysis layer, an axial angle monitoring layer, a lubrication loss assessment layer and a third life prediction layer; The load change analysis layer analyzes the long-term high load conditions borne by the universal joint, and evaluates the impact on the life of the universal joint under high load conditions; the axial angle monitoring layer monitors the angle changes of the universal joint, and calculates the impact of the angle changes on the working state of the universal joint under high load conditions; the lubrication loss assessment layer analyzes the lubricant state and loss of the universal joint, and estimates the impact on the life of the universal joint by evaluating the lubricant loss under high load, frequent starting and braking; the third life prediction layer combines the data of the load change analysis layer, the axial angle monitoring layer and the lubrication loss assessment layer to calculate the third remaining life of the universal joint under heavy load driving.

6. The service life prediction device of a vehicle universal joint according to claim 1, characterized in that: The process of the abnormal pattern detection module includes: abnormal data collection layer, abnormal pattern recognition layer and dynamic correction prediction layer; The abnormal data acquisition layer monitors the temperature, load, speed, vibration and lubrication status in real time through sensors, and compares them with the normal operating status; the abnormal pattern recognition layer uses a machine learning algorithm to detect abnormal events in a short period of time, mark them as abnormal patterns and trigger alarms; the abnormal patterns include: sudden impact, high temperature and overload; when the dynamic correction prediction layer detects the abnormal pattern, it adjusts the parameters of the personalized life prediction model and dynamically corrects the life prediction value.

7. A service life prediction system for automobile universal joints, characterized in that: include: Driving style analysis unit, used to analyze driving habits based on historical data and identify the user's driving style, including: stable, sporty and heavy; A personalized life prediction unit, used to construct a personalized life prediction model, and adaptively select a life prediction model to predict the life of the universal joint according to the driving style; the personalized life prediction model includes: a first prediction model, a second prediction model and a third prediction model; if the driving style is the stable type, the first prediction model is used to perform a first life prediction based on the operating temperature, torque load and speed; if the driving style is the sporty type, the second prediction model is used to perform a second life prediction based on torque impact, instantaneous high temperature and high-frequency vibration; if the driving style is the load-bearing type, the third prediction model is used to perform a third life prediction by monitoring load changes, axial angle changes and lubrication loss; The abnormal mode detection unit is used to monitor the operating state of the universal joint and optimize the parameters of the personalized life prediction model when an abnormal situation is detected, and dynamically correct the result of the life prediction.

Images