Method and device for predicting lifetime of oxygen sensor, vehicle and storage medium

Abstract

The application relates to a life prediction method and device of an oxygen sensor, a vehicle and a storage medium, wherein the method comprises the following steps: obtaining historical running data and historical meteorological data of a target vehicle, wherein the historical running data comprises vehicle geographical position information, a time stamp and a voltage response time parameter of the oxygen sensor, and the historical meteorological data comprises an ambient temperature and an ambient humidity; based on the ambient temperature and the ambient humidity, the cumulative wet heat exposure of the target vehicle within a preset period is calculated; a correlation model between the cumulative wet heat exposure and the voltage response time decay rate of the oxygen sensor is constructed; based on the correlation model, the remaining service life of the oxygen sensor is calculated through the current voltage response time and the current cumulative wet heat exposure of the target vehicle, and early warning information is generated when the remaining service life is lower than a preset threshold. The application realizes early identification and early warning of the performance degradation trend of the oxygen sensor, and improves the life prediction accuracy of the oxygen sensor.

Description

Technical Field

[0001] This application relates to the field of automotive electronics technology, and in particular to a method, apparatus, vehicle, and storage medium for predicting the lifespan of an oxygen sensor. Background Technology

[0002] As a core component of automotive electronic control systems, the oxygen sensor plays a crucial role in real-time monitoring of exhaust oxygen content. Its operational status directly determines the accuracy of air-fuel ratio control, thereby affecting the vehicle's emissions compliance and fuel efficiency. During vehicle operation, this sensor is continuously exposed to a complex and ever-changing external environment, and its performance degradation often exhibits a gradual characteristic rather than sudden failure.

[0003] The current industry-standard diagnostic mechanisms heavily rely on on-board diagnostic systems for real-time monitoring of voltage response time or signal fluctuation frequency. This passive response mode has significant drawbacks: the system only issues an alarm when sensor performance deteriorates to the point of triggering a fault code, failing to capture early signs of slow performance degradation. This results in vehicle owners and repair shops losing a window for proactive intervention, making it difficult to develop preventative maintenance plans. A deeper problem lies in the serious neglect of the driving effect of environmental heat and humidity on the sensor aging process. In high-temperature and high-humidity areas, ambient temperature and humidity together constitute key external stress sources. High temperatures can cause irreversible changes in the conductivity of the electrolyte inside the oxygen sensor, accelerating the degradation of the platinum electrode's pore structure and promoting microcracks in the ceramic matrix. High humidity environments easily condense moisture inside the sensor cavity; the corrosive salts remaining after evaporation continue to erode the heating element and sensing film. Simultaneously, diurnal or seasonal temperature and humidity fluctuations can induce thermal fatigue, further exacerbating material performance degradation. This type of aging process caused by humid and hot environments is highly cumulative and insidious. Even if the vehicle is parked for a long time, extreme humid and hot conditions will continue to erode the internal structure of the sensor, causing imperceptible performance loss.

[0004] However, the existing diagnostic system does not take into account the dynamic damp heat exposure experienced by vehicles in reality. Although there are accelerated aging test methods based on the superposition of photothermal and damp stress in the laboratory environment, the simulation conditions cannot reproduce the damp heat stress fluctuation characteristics caused by geographical location differences, seasonal changes and microenvironmental changes in real road scenarios, resulting in a lack of specificity and accuracy in the prediction of the remaining life of single vehicle sensors. Summary of the Invention

[0005] The purpose of this application is to provide a method, device, vehicle, and storage medium for predicting the lifespan of an oxygen sensor, enabling early identification and warning of oxygen sensor performance degradation trends, and improving the accuracy of oxygen sensor lifespan prediction.

[0006] To address the aforementioned technical problems, embodiments of this application provide a method for predicting the lifetime of an oxygen sensor, comprising: Acquire historical operating data and historical meteorological data of the target vehicle. The historical operating data includes vehicle geographical location information, timestamps, and voltage response time parameters of the oxygen sensor. The historical meteorological data includes ambient temperature and ambient humidity. Based on the ambient temperature and the ambient humidity, calculate the cumulative heat and humidity exposure of the target vehicle within a preset time period; Construct a correlation model between the cumulative damp heat exposure and the voltage response time decay rate of the oxygen sensor; Based on the aforementioned correlation model, the remaining service life of the oxygen sensor is calculated using the target vehicle's current voltage response time and current cumulative damp heat exposure, and a warning message is generated when the remaining service life is lower than a preset threshold.

[0007] To address the aforementioned technical problems, embodiments of this application provide a lifespan prediction device for an oxygen sensor, comprising: The historical data acquisition module is used to acquire historical operating data and historical meteorological data of the target vehicle. The historical operating data includes vehicle geographical location information, timestamps and voltage response time parameters of the oxygen sensor, and the historical meteorological data includes ambient temperature and ambient humidity. The damp heat exposure calculation module is used to calculate the cumulative damp heat exposure of the target vehicle within a preset time period based on the ambient temperature and the ambient humidity. The correlation model building module is used to build a correlation model between the cumulative damp heat exposure and the voltage response time decay rate of the oxygen sensor. The remaining service life prediction module is used to calculate the remaining service life of the oxygen sensor based on the correlation model, the current voltage response time of the target vehicle, and the current cumulative damp heat exposure, and generate a warning message when the remaining service life is lower than a preset threshold.

[0008] To solve the above-mentioned technical problems, one technical solution adopted by the present invention is to provide a vehicle, including one or more processors; and a memory for storing one or more programs, such that the one or more processors implement the oxygen sensor lifetime prediction method described in any one of the above-mentioned methods.

[0009] To solve the above-mentioned technical problems, one technical solution adopted by the present invention is: a computer-readable storage medium storing a computer program, wherein the computer program, when executed by a processor, implements the oxygen sensor lifetime prediction method described in any one of the above-mentioned methods.

[0010] This invention provides a method, device, vehicle, and storage medium for predicting the lifespan of an oxygen sensor. It acquires historical operating data and historical meteorological data of the target vehicle. The historical operating data includes the vehicle's geographical location information, timestamps, and the voltage response time parameter of the oxygen sensor. The historical meteorological data includes ambient temperature and humidity. Based on the ambient temperature and humidity, the cumulative damp heat exposure of the target vehicle within a preset time period is calculated. A correlation model is constructed between the cumulative damp heat exposure and the voltage response time decay rate of the oxygen sensor. Based on the correlation model, the remaining lifespan of the oxygen sensor is calculated using the current voltage response time and current cumulative damp heat exposure of the target vehicle. An early warning message is generated when the remaining lifespan is lower than a preset threshold. By quantifying the cumulative impact of the damp heat environment on the aging process of the oxygen sensor, a dynamic correlation mechanism between environmental exposure and performance degradation is established, effectively capturing early signs of slow sensor performance decline. This method has the advantages of achieving early warning of oxygen sensor lifespan and improving prediction accuracy. Attached Figure Description

[0011] To more clearly illustrate the solutions in this application, the accompanying drawings used in the description of the embodiments of this application will be briefly introduced below. Obviously, the accompanying drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0012] Figure 1 This is a flowchart of an implementation of the oxygen sensor lifetime prediction method provided in this application embodiment; Figure 2 This is a flowchart illustrating the implementation of the first sub-process in the oxygen sensor lifetime prediction method provided in this application embodiment; Figure 3 This is a flowchart illustrating the implementation of the second sub-process in the oxygen sensor lifetime prediction method provided in this application embodiment; Figure 4 This is a flowchart illustrating the implementation of the third sub-process in the oxygen sensor lifetime prediction method provided in this application embodiment; Figure 5 This is a flowchart illustrating the implementation of the fourth sub-process in the oxygen sensor lifetime prediction method provided in this application embodiment; Figure 6 This is a schematic diagram of the oxygen sensor lifetime prediction device provided in the embodiments of this application; Figure 7 This is a schematic diagram of the vehicle provided in the embodiments of this application. Detailed Implementation

[0013] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains; the terminology used herein in the specification of the application is for the purpose of describing particular embodiments only and is not intended to be limiting of the application; the terms "comprising" and "having," and any variations thereof, in the specification, claims, and foregoing drawings of this application, are intended to cover non-exclusive inclusion. The terms "first," "second," etc., in the specification, claims, or foregoing drawings of this application are used to distinguish different objects, not to describe a particular order.

[0014] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0015] Current oxygen sensor diagnostics primarily rely on on-board diagnostic systems to monitor voltage response time or signal fluctuation frequency in real time. However, this diagnostic approach is passive, often only triggering an alarm when sensor performance has severely deteriorated and fault codes have been activated. It fails to provide early warnings in the early stages of progressive sensor performance degradation, making it difficult to support proactive maintenance decisions for vehicle owners or repair shops. Furthermore, existing methods do not incorporate the actual external environmental humidity and heat stress experienced by the vehicle into the lifespan assessment system, making it difficult to accurately predict the remaining lifespan of individual vehicle sensors.

[0016] To address this issue, this application proposes a method for predicting the lifespan of an oxygen sensor, comprising: acquiring historical operating data and historical meteorological data of a target vehicle, wherein the historical operating data includes vehicle geographical location information, timestamps, and voltage response time parameters of the oxygen sensor, and the historical meteorological data includes ambient temperature and ambient humidity; calculating the cumulative damp heat exposure of the target vehicle within a preset time period based on the ambient temperature and ambient humidity; constructing a correlation model between the cumulative damp heat exposure and the voltage response time decay rate of the oxygen sensor; and calculating the remaining lifespan of the oxygen sensor based on the correlation model, using the current voltage response time and current cumulative damp heat exposure of the target vehicle, and generating a warning message when the remaining lifespan is lower than a preset threshold.

[0017] For ease of understanding, the following explains some key terms in this embodiment: Historical operational data refers to records related to the operating status of a target vehicle over a period of time, such as the vehicle's geographical location information, data acquisition timestamps, and the voltage response time parameters of the oxygen sensor under different operating conditions. This data is used to depict the vehicle's usage trajectory and the changes in sensor performance over time.

[0018] Historical meteorological data refers to records of environmental conditions experienced by a target vehicle at a specific geographical location and time point, mainly including ambient temperature and humidity. This data is used to quantify the external environmental stresses experienced by the vehicle.

[0019] The voltage response time parameter of an oxygen sensor refers to the time required for the oxygen sensor to output a stable voltage signal after receiving an exhaust signal. This parameter is a key indicator of oxygen sensor performance, and its extension usually indicates a degradation in sensor performance.

[0020] Cumulative damp heat exposure refers to the total amount of damp heat stress that has a cumulative effect on the aging of oxygen sensors, calculated using a specific algorithm based on the temperature and humidity conditions of the target vehicle's environment over a preset period. This quantitative indicator reflects the degree of external environmental corrosion experienced by the sensor.

[0021] Voltage response time decay rate refers to the rate at which the voltage response time of an oxygen sensor increases with time or cumulative exposure to damp heat. This decay rate is an important indicator for evaluating the rate of sensor performance degradation.

[0022] A correlation model is a predictive model used to describe the mathematical relationship between cumulative damp heat exposure and the voltage response time decay rate of an oxygen sensor. This model, constructed through statistical analysis or machine learning methods, can quantitatively map environmental stress to sensor performance degradation.

[0023] Remaining service life refers to the length of time an oxygen sensor can continue to function normally under its current performance condition and expected future environmental stresses until its performance reaches the failure standard or a preset threshold.

[0024] The preset threshold refers to a critical value set in the lifespan prediction or early warning mechanism. When the remaining lifespan of the oxygen sensor falls below this threshold, the system will trigger an early warning message to prompt the user to pay attention or perform maintenance.

[0025] This application integrates vehicle operation data and environmental meteorological data to quantify cumulative damp heat exposure and constructs a correlation model between this exposure and the oxygen sensor voltage response time decay rate. This enables accurate prediction of the remaining lifespan of the oxygen sensor, overcoming the limitations of traditional passive diagnostics. This method incorporates external environmental damp heat stress into the evaluation system, providing proactive warnings in the early stages of progressive sensor performance degradation. It offers forward-looking maintenance decision support for vehicle owners and repair shops, significantly improving the reliability and fuel economy of vehicle emission control systems.

[0026] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings.

[0027] Please see Figure 1 , Figure 1 A specific implementation of a method for predicting the lifetime of an oxygen sensor is shown.

[0028] It should be noted that if substantially the same result is obtained, the method of this invention is not based on... Figure 1 Limited to the order of the processes shown, this method includes the following steps: S1: Obtain historical operating data and historical meteorological data of the target vehicle. The historical operating data includes vehicle geographical location information, timestamps, and voltage response time parameters of the oxygen sensor. The historical meteorological data includes ambient temperature and ambient humidity.

[0029] Specifically, historical operating data can be collected in real time by the on-board diagnostic system, including the vehicle's geographical location information (e.g., obtained through a GPS module), data acquisition timestamps, and the voltage response time parameters of the oxygen sensor under different operating conditions. Historical meteorological data can be obtained through the vehicle's built-in meteorological sensors or periodically retrieved from an external meteorological service platform via the on-board communication module. As an optional implementation, during vehicle maintenance, historical operating data stored in the vehicle's electronic control unit can be read using diagnostic tools. Meteorological data can be retrieved retrospectively from a public meteorological database based on the vehicle's driving records (geographical location and time). For example, during vehicle operation, its on-board system can record the current location, time, and oxygen sensor voltage response time every minute. Simultaneously, the on-board meteorological module can obtain the ambient temperature and humidity of the current location via a network connection. This data is stored in the vehicle's memory or uploaded to a cloud server as historical data.

[0030] Please see Figure 2 , Figure 2 A specific implementation of step S1 when the target vehicle is in a static parking scenario is shown below: S11: Obtain the static geographic location information of the fixed location where the target vehicle is parked, and obtain the parking timestamp of the target vehicle. S12: Based on the static geographic location information and the parking timestamp, obtain the corresponding target ambient temperature and target ambient humidity. S13: Incorporate the target ambient temperature and target ambient humidity into the historical meteorological data.

[0031] Specifically, obtaining the static geographic location information of the target vehicle at its fixed parking location refers to acquiring its precise geographic coordinates when the vehicle is parked. This can be achieved in various ways. For example, the vehicle's built-in Global Positioning System (GPS) module can automatically record and upload its final location coordinates when the vehicle is turned off or enters parking mode; or, the vehicle can connect to external devices such as smartphones and use the phone's positioning function to obtain and synchronize location information. Obtaining the target vehicle's parking timestamp refers to recording the precise time when the vehicle begins and / or ends parking. For example, the vehicle's on-board computer or on-board diagnostic system (OBD) can automatically record the current time as the parking start time when the engine is off and as the parking end time when the engine is started; or, the vehicle's Telematics Control Unit (TCU) can periodically send vehicle status information to the cloud platform, including whether the vehicle is parked and the corresponding system time.

[0032] Based on this, obtaining the corresponding target ambient temperature and humidity according to the static geographic location information and the parking timestamp refers to querying and obtaining ambient temperature and humidity data within a given time period using known parking location and time information. This can be achieved by calling a third-party meteorological service interface (such as a weather API), using the static geographic location information and parking timestamp as query parameters to obtain historical meteorological data for that location within a specified time period; or by using a regional meteorological database established by the vehicle manufacturer or a third-party data service provider. This database integrates meteorological data from weather stations, satellite remote sensing, or densely deployed IoT sensor networks, and performs data retrieval through geospatial and temporal matching.

[0033] Subsequently, incorporating the target ambient temperature and humidity into the historical meteorological data refers to integrating the ambient temperature and humidity data acquired under static parking conditions into the overall historical meteorological dataset used for oxygen sensor lifetime prediction. This can be achieved by inserting these data into the existing historical meteorological data stream in time-series form, ensuring the temporal continuity and integrity of the data; or, at the data storage level, by merging these data with meteorological data collected during vehicle operation to form a unified database containing environmental exposure information throughout the vehicle's entire lifecycle.

[0034] S2: Based on the ambient temperature and the ambient humidity, calculate the cumulative heat and humidity exposure of the target vehicle within a preset time period.

[0035] Specifically, a simple weighted average method can be used to sum the ambient temperature and humidity at each time point within a preset time period to obtain a cumulative value. For example, temperature and humidity can be multiplied by an empirical coefficient and then summed. Alternatively, an integral method can be used, integrating temperature and humidity over time within the preset time period to reflect the continuous effect of damp-heat stress. For example, a damp-heat stress function can be defined as a composite function of temperature and humidity, and then this function can be integrated over the preset time period.

[0036] In one specific embodiment, when the target vehicle is parked, the cumulative damp heat exposure is calculated using the following formula:

[0037] in, For the first Ambient temperature (degrees Celsius) during the period. For the first Relative humidity (%) during the time period The baseline humidity is 50% (standard laboratory environment). Here, A, B, and Ea are time weighting coefficients, and A, B, and Ea are fitting constants. Cumulative damp heat exposure Please see Figure 3 , Figure 3 A specific implementation method following step S2 is shown below: S21: Acquire the target sensor signals of the target vehicle, wherein the target sensor signals include solar radiation sensor signals and rainfall / humidity sensor signals. S22: Determine an environmental exposure reduction factor based on the signals. S23: Correct the cumulative damp heat exposure using the environmental exposure reduction factor.

[0038] Specifically, acquiring the target sensor signals of the target vehicle aims to directly capture the micro-environmental characteristics of the vehicle. The target sensor signals may include solar radiation sensor signals and rain / humidity sensor signals. The solar radiation sensor signals can be acquired, for example, through photosensitive elements (such as photoresistors, photodiodes, or phototransistors) installed on the exterior of the vehicle (e.g., on the roof or below the windshield). These elements can monitor the light intensity around the vehicle in real time, thus reflecting whether the vehicle is under direct sunlight or shade. The rain / humidity sensor signals can be acquired, for example, through capacitive humidity sensors, resistive humidity sensors, or piezoelectric rain sensors installed on the exterior of the vehicle (e.g., near windows or rearview mirrors). These sensors can detect the relative humidity of the environment and whether rainfall has occurred in real time, thus reflecting whether the vehicle is wet or rained on. These sensor signals provide objective, real-time micro-environmental data support for subsequent corrections.

[0039] Based on this, determining the environmental exposure reduction factor based on the aforementioned signal is a key step in this application. Its purpose is to transform the abstract vehicle microenvironment state into a quantifiable correction parameter. Specifically, a mapping relationship can be pre-established, for example, by consulting a pre-defined lookup table or empirical formula. This lookup table or empirical formula can output a corresponding environmental exposure reduction factor based on the combination of the intensity of the sunlight radiation sensor signal (e.g., strong light, weak light, no light) and the state of the rain / humidity sensor signal (e.g., high humidity, low humidity, rain, no rain). For example, when the sunlight radiation intensity is high and the humidity is low, the reduction factor may be high; when the light intensity is weak and the humidity is high, the reduction factor may be low. Alternatively, machine learning methods can be used, such as training a decision tree model or support vector machine model, using historical sensor signal data as input and actual aging conditions as labels, to learn and output the corresponding environmental exposure reduction factor. This reduction factor can reflect the degree of damp-heat stress actually experienced by the oxygen sensor under different microenvironmental conditions.

[0040] Subsequently, the cumulative damp heat exposure is corrected using the environmental exposure reduction factor to adjust the cumulative damp heat exposure calculated from general meteorological data to an effective exposure amount that better reflects the actual microenvironment of the vehicle. One correction method is a multiplicative correction, which multiplies the originally calculated cumulative damp heat exposure by the determined environmental exposure reduction factor to obtain the corrected cumulative damp heat exposure. For example, if the vehicle is in a shaded state, the reduction factor may be less than 1, thus reducing the cumulative damp heat exposure; if the vehicle is in an extreme exposure state, the reduction factor may be greater than 1, thus increasing the cumulative damp heat exposure. Another correction method is to dynamically adjust the weights or coefficients in the cumulative damp heat exposure calculation formula based on the reduction factor. For example, when calculating damp heat exposure, the contribution ratio of temperature and humidity to the cumulative amount can be adjusted according to the reduction factor. Through this correction, calculation errors caused by differences in the microenvironment such as vehicle parking location and shading conditions can be effectively eliminated, allowing the cumulative damp heat exposure to more accurately reflect the physical stress experienced by the oxygen sensor during its actual service life.

[0041] The corrected cumulative damp heat exposure is calculated using the following formula: ,in, It is an environmental exposure reduction factor that is dynamically adjusted based on the light sensor signal. This is the corrected cumulative damp heat exposure. For the first Ambient temperature (degrees Celsius) during the period. For the first Relative humidity (%) during the time period.

[0042] Please see Figure 4 , Figure 4 A specific implementation of step S22 is shown below: S221: When the target vehicle is stationary and the sunlight radiation sensor detects that the light intensity is greater than a preset light threshold, the target vehicle is determined to be exposed to direct sunlight, and a first reduction factor value is assigned. S222: When the target vehicle is stationary and the light intensity is lower than the preset light threshold, and the rain sensor detects no precipitation, the target vehicle is determined to be parked under shade, and a second reduction factor value is assigned, wherein the second reduction factor value is less than the first reduction factor value. S223: When the rain / humidity sensor detects that the relative humidity is continuously higher than a second threshold for more than a preset duration, the target vehicle is determined to be in an extreme humidity scenario, and the humidity sensitivity coefficient in the calculation of cumulative damp heat exposure is increased.

[0043] Specifically, the determination that the target vehicle is stationary can be achieved in several ways. For example, it can be based on the vehicle's Global Positioning System (GPS) data, determining that the vehicle is stationary when the GPS speed is zero for a continuous period of time (e.g., 30 seconds to 2 minutes). Alternatively, it can be combined with the vehicle's wheel speed sensor signals, determining that the vehicle is stationary when the rotational speed signals of all wheels are below a certain extremely low threshold (e.g., close to zero).

[0044] A solar radiation sensor is typically a photosensitive device, such as a photodiode or photoresistor, that measures the intensity of solar radiation in the environment and outputs a corresponding electrical signal. This sensor is generally installed on the exterior of a vehicle, such as above the dashboard or on the roof, to detect whether the vehicle is directly exposed to sunlight. The preset light threshold is an empirical or calibrated value used to distinguish between strong and weak sunlight. Its setting can be determined through experiments or simulations based on typical solar radiation intensity data for different geographical regions and seasons; for example, it can be set to 500 W / m² or 1000 W / m². The first reduction factor value is a correction coefficient used to quantify the actual impact of environmental damp heat exposure caused by solar radiation under open-air exposure conditions. This factor value is typically greater than 1, indicating that the damp heat stress experienced by the oxygen sensor is amplified under open-air exposure conditions. For example, it can be set to 1.2 to 1.5, and the specific value can be calibrated through accelerated aging experiments and actual road test data.

[0045] Rain sensors are used to detect precipitation, employing either optical principles (detecting changes in the refractive index of light to identify raindrops) or piezoelectric principles (detecting vibrations caused by raindrop impacts). These sensors are typically mounted near the windshield of vehicles to provide precipitation information. The second reduction factor is a correction coefficient used to quantify the actual impact of environmental heat and humidity exposure under shaded parking conditions. This factor value is usually less than the first reduction factor value and may be less than or equal to 1, indicating that the oxygen sensor experiences relatively low heat and humidity stress under shaded conditions. For example, it can be set to 0.8 to 1.0, but the specific value also needs to be calibrated through experiments and data.

[0046] The rain / humidity sensor can be an integrated temperature and humidity sensor with raindrop detection capabilities, or it can work in conjunction with a separate rain sensor and humidity sensor. This sensor can simultaneously detect precipitation and relative humidity in the environment, providing more comprehensive environmental humidity information. The second threshold is a relative humidity percentage used to define "high humidity," such as 90% RH. The preset duration is used to define the length of time of "continuous high humidity," such as 2 hours or 4 hours. The settings of these two parameters are intended to identify extreme humidity scenarios where prolonged exposure to high humidity can lead to condensation and accelerated corrosion. Increasing the humidity sensitivity coefficient in the calculation of cumulative damp heat exposure means that in extreme humidity scenarios, the model will more significantly amplify the impact of humidity, thus more accurately reflecting the accelerated aging trend of the sensor due to high humidity. For example, the original humidity sensitivity coefficient can be multiplied by a correction factor greater than 1, or it can be directly replaced with a higher preset coefficient.

[0047] S3: Construct a correlation model between the cumulative damp heat exposure and the voltage response time decay rate of the oxygen sensor.

[0048] Specifically, a linear regression model can be used, with cumulative damp heat exposure as the independent variable and voltage response time decay rate as the dependent variable, fitting the linear relationship between the two using the least squares method. As a preferred implementation, a multinomial regression model can be employed to capture the non-linear relationship between cumulative damp heat exposure and decay rate. Alternatively, machine learning models such as support vector machines or neural networks can be used to learn the complex mapping relationship between cumulative damp heat exposure and decay rate by training on historical data. For example, a large amount of cumulative damp heat exposure data and corresponding oxygen sensor voltage response time decay rate data can be collected from vehicles. Using this data, a simple neural network model can be trained to learn and predict the oxygen sensor voltage response time decay rate under a given cumulative damp heat exposure.

[0049] Please see Figure 5 , Figure 5A specific implementation of step S3 is shown below: S31: Based on the voltage response time parameter of the oxygen sensor, the historical operating data is divided into failure samples and truncated samples, and a status label variable is constructed. The failure sample refers to the sample whose voltage response time exceeds a preset threshold or has triggered a fault code, and the truncated sample refers to the sample whose sensor is still working.

[0050] Specifically, dividing the data into different samples aims to fully utilize all available historical data, including data from sensors that have failed and those that are still operating but have not yet failed. For example, a voltage response time threshold (e.g., 200 milliseconds) can be set, and a sensor whose voltage response time consistently exceeds this threshold can be marked as a failed sample; alternatively, when the on-board diagnostic system (OBD) detects specific fault codes related to the oxygen sensor (e.g., P0133, P0134), it can also be marked as a failed sample. A truncated sample refers to a sensor that is still functioning normally at the time of data acquisition cutoff, where its voltage response time has not met the failure criteria. By constructing a state label variable for each sample—for example, labeling failed samples as 1 and truncated samples as 0—the state of each data point can be clearly identified, providing a foundation for subsequent reliability analysis.

[0051] S32: Construct a joint likelihood function based on the Weibull distribution model, which includes the probability density function of the failed samples and the survival function of the truncated samples.

[0052] Specifically, the Weibull distribution model is widely used in reliability engineering to describe the lifespan distribution of products, especially suitable for describing components with different failure modes and aging processes. For failure samples, its contribution is reflected in the probability density function of the Weibull distribution, which describes the probability of failure occurring at a specific time point. For truncated samples, its contribution is reflected in the survival function of the Weibull distribution, which describes the probability that the sensor is still alive after a specific time point. By multiplying the probability density function values ​​of all failure samples and the survival function values ​​of all truncated samples, a joint likelihood function can be constructed. This joint likelihood function comprehensively reflects the probability of all observed data (including failure and truncated data) under given Weibull distribution parameters.

[0053] In one specific embodiment, the joint likelihood function is constructed using the following formula: ; Here, f(t) is the probability density function (describing failed samples), and S(t) is the survival function (describing the probability of survival of non-failed samples). By maximizing this likelihood function, the model can simultaneously use "failed data" to determine the end of the lifetime and "non-failed data" to determine the shape parameter of the lifetime distribution, thereby avoiding prediction bias caused by ignoring truncated data.

[0054] S33: By maximizing the joint likelihood function, the shape and scale parameters of the Weibull distribution model are estimated to construct a correlation model between the cumulative damp heat exposure and the voltage response time decay rate of the oxygen sensor.

[0055] Specifically, maximum likelihood estimation (MLE) is a commonly used statistical method for estimating model parameters from data. Through iterative optimization algorithms, such as the Newton-Raphson method or gradient descent, the shape parameter (β) and scale parameter (η) that maximize the joint likelihood function can be found. The shape parameter β reflects the trend of the failure rate over time, while the scale parameter η is closely related to the sensor's characteristic lifetime or average lifetime. In this way, cumulative damp heat exposure can be introduced as a covariate into the parameters of the Weibull distribution (e.g., by modeling the scale parameter η as a function of cumulative damp heat exposure), thereby establishing a quantitative correlation model between cumulative damp heat exposure and the decay rate of the oxygen sensor's voltage response time.

[0056] This application effectively addresses the problem of incomplete data in sensor lifespan prediction by fully utilizing historical data containing a large number of truncated samples of unfailed sensors, thus avoiding sample bias caused by relying solely on failed data. Furthermore, by introducing a Weibull distribution model, it accurately characterizes the stochastic degradation pattern of oxygen sensors under different environmental stresses, transforming the degradation process of sensors under different levels of damp heat exposure into a statistical parameter estimation problem. This method not only handles asymmetric lifespan data but also establishes a mathematical mapping between environmental exposure and sensor performance degradation through parameterization, thereby enabling scientific prediction of the remaining lifespan of oxygen sensors. This significantly improves the accuracy and reliability of the prediction, providing more precise proactive maintenance decision support for vehicle owners or repair shops.

[0057] Furthermore, this application also provides a specific embodiment, including: introducing an Arrhenius-humidity-heat coupling acceleration factor into the Weibull distribution model, wherein the acceleration factor is based on ambient temperature and relative humidity, and the scale parameter is modified to reflect the accelerating effect of high humidity and heat environment on the lifespan of the oxygen sensor; when estimating the shape parameter and scale parameter of the Weibull distribution model, elastic network regression is used for regularization processing, and at least one of the following physical constraints is applied: the temperature effect satisfies the monotonically increasing characteristic of Arrhenius, and the aging acceleration increases nonlinearly when the relative humidity exceeds a first threshold.

[0058] Specifically, the Arrhenius-humidity-heat coupling acceleration factor aims to quantify the combined impact of ambient temperature and relative humidity on the aging rate of oxygen sensors. Its core idea is to combine the traditional Arrhenius equation (describing the effect of temperature on reaction rate) with a humidity-related correction term, thus forming a mathematical expression that simultaneously reflects the acceleration effect of temperature and humidity on sensor lifetime. This acceleration factor, by correcting the scaling parameter in the Weibull distribution model, allows the model to dynamically adapt to the aging rate under different environmental conditions, thereby more accurately predicting the lifetime of the oxygen sensor. The acceleration factor can be a composite function, for example, combining the temperature dependence of the Arrhenius equation with a nonlinear function based on relative humidity through multiplication or addition. Another implementation method is to construct a multidimensional lookup table using pre-established experimental data or simulation results. This lookup table takes ambient temperature and relative humidity as input and directly outputs the corresponding acceleration factor value, and obtains the acceleration factor at arbitrary temperatures and humidity levels through interpolation.

[0059] Elastic network regression is used for regularization when estimating the shape and scale parameters of the Weibull distribution model. Elastic network regression is a linear regression method that combines the advantages of Lasso (L1 regularization) and Ridge (L2 regularization) to prevent overfitting and improve the model's generalization ability during training. When estimating the shape and scale parameters of the Weibull distribution model, direct maximum likelihood estimation may lead to unstable parameter estimates or overfitting due to potential noise in the data or collinearity among features. By introducing elastic network regression for regularization, the absolute values ​​and sum of squares of the model parameters can be penalized while minimizing the loss function, thereby encouraging the model to select more important features and reduce the coefficients of less important features, improving the model's robustness. Specifically, an additional regularization term, which is a weighted sum of the L1 and L2 norms, can be added during the process of maximizing the joint likelihood function (or minimizing the negative log-likelihood function). Another approach is to gradually adjust the parameter values ​​during the parameter optimization iteration process by combining the iterative shrinking threshold (Lasso) and weight decay (Ridge) operations until convergence to the optimal solution, thereby achieving regularized parameter estimation.

[0060] Furthermore, this application imposes physical constraints, one of which is that the temperature effect satisfies the Arrhenius monotonically increasing property. This physical constraint ensures that the model's description of the temperature effect conforms to basic physicochemical laws, namely, within a certain temperature range, increasing temperature accelerates chemical reactions and material aging processes. In oxygen sensor aging prediction, this means that the sensor's aging rate should monotonically increase with increasing ambient temperature. Imposing this constraint can prevent the model from producing predictions that violate physical common sense under conditions of sparse or anomalous data. Specifically, during model parameter estimation, positive constraints can be imposed on temperature-related parameters. Another implementation method is that, in the optimization algorithm, if an iteration leads to a non-monotonic increase in the temperature effect, the parameters are forced back to the region satisfying the monotonically increasing property by adjusting the step size or parameter projection.

[0061] Another physical constraint is that aging accelerates nonlinearly when relative humidity exceeds a first threshold. This constraint aims to capture the nonlinear aging mechanism unique to oxygen sensors under high humidity conditions. When relative humidity exceeds a certain critical value (the first threshold), phenomena such as condensation, dissolution of corrosive substances, and salt deposition may occur. These factors lead to a significant acceleration in the sensor aging rate, and this acceleration effect is often not linear but exhibits a steeper growth trend. Imposing this constraint allows the model to more accurately reflect the destructive effect of extreme hot and humid environments on sensor lifespan. Specifically, in the design of the humidity term of the acceleration factor, piecewise functions or nonlinear functions can be used to describe the humidity effect. For example, when the relative humidity is below the first threshold, a linear or low-order polynomial function can be used; when the relative humidity exceeds the first threshold, an exponential function, a high-order polynomial function, or a power function can be switched to simulate nonlinear acceleration. Another implementation is to introduce an indicator variable that activates an additional nonlinear acceleration term when the relative humidity exceeds the first threshold. The coefficients of this acceleration term are optimized during parameter estimation to ensure that it reflects the characteristics of nonlinear growth.

[0062] In a specific example, a hierarchical modeling strategy is adopted to establish an association model, which includes: training a general Weibull aging model using historical data of multiple vehicle models; and for a specific vehicle model, introducing an offset factor on the basis of the general model to obtain the Weibull distribution parameters of that vehicle model.

[0063] This structure allows the model to automatically degenerate into a general layer prediction when there is limited data for new car models, thus avoiding prediction failures caused by data sparsity.

[0064] The embodiments of this application effectively solve the problems that traditional statistical regression is unable to accurately characterize the nonlinear effects of complex environmental stress and the lack of physical mechanism constraints on model parameters when constructing a correlation model between cumulative damp heat exposure and oxygen sensor voltage response time decay rate.

[0065] S4: Based on the correlation model, calculate the remaining service life of the oxygen sensor by the current voltage response time and current cumulative damp heat exposure of the target vehicle, and generate a warning message when the remaining service life is lower than a preset threshold.

[0066] Specifically, the current cumulative damp heat exposure is substituted into the established correlation model to predict the future voltage response time decay rate. Then, based on the difference between the current voltage response time and the preset failure voltage response time, combined with the predicted decay rate, the additional time required to reach the failure criterion is calculated, which is the remaining service life. Alternatively, the correlation model can directly output the total cumulative damp heat exposure required for the sensor to reach failure at the current level. By subtracting the current cumulative damp heat exposure and considering the vehicle's expected future damp heat exposure rate, the remaining service life can be estimated. For example, suppose the correlation model predicts that the oxygen sensor will fail when the cumulative damp heat exposure reaches value X. If the current cumulative damp heat exposure is Y, and the expected daily damp heat exposure for the vehicle is Z, then the remaining service life can be calculated as (XY) / Z days.

[0067] Finally, a warning message is generated when the remaining service life falls below a preset threshold. Specifically, when the calculated remaining service life is lower than the preset threshold, the system can send a warning message to the owner or maintenance personnel via the in-vehicle display, mobile application push notification, email, or SMS. The warning message may include an estimate of the remaining service life, a recommended inspection or replacement time, and an analysis of the main environmental factors that may cause sensor aging. For example, if the preset threshold is 30 days, when the calculated remaining service life of the oxygen sensor is 25 days, the in-vehicle system will automatically display a "Oxygen Sensor Life Warning: Inspection Recommended within 30 Days" message on the dashboard and send a detailed warning notification via the owner's mobile application.

[0068] Furthermore, this application also provides a specific embodiment, including: identifying the time node in the historical operating data where the oxygen sensor voltage response time suddenly increases using the isolated forest algorithm; extracting historical meteorological data within a preset period before the time node, and determining the key environmental condition combination that leads to sensor performance degradation through cluster analysis; quantifying the contribution weights of temperature and humidity factors to voltage response time decay in the key environmental condition combination using a game theory-based attribution analysis algorithm, and generating a visual attribution report.

[0069] Specifically, the Isolation Forest algorithm is used to identify time points in historical operational data where the oxygen sensor voltage response time suddenly increases. The aim is to efficiently and accurately detect time points in massive amounts of historical operational data where oxygen sensor performance is abnormal and the voltage response time parameter suddenly and significantly increases. These sudden increases are often clear signals of nonlinear degradation or impending failure of sensor performance. The Isolation Forest algorithm is an anomaly detection method based on decision trees. Its core idea is to quickly "isolate" anomalies from normal data points by randomly selecting features and split points. Since anomalies are usually few in number and their distribution differs greatly from normal points, they often have short path lengths in the randomly divided tree structure. One implementation involves constructing multiple random decision trees. Each tree randomly extracts a subsample from the dataset and randomly selects a feature for splitting. The degree of anomaly is determined by calculating the average path length of each data point; the shorter the path length, the higher the degree of anomaly. Another approach is to use distance- or density-based anomaly detection methods, such as the Local Anomaly Factor (LOF) algorithm or K-Nearest Neighbors (KNN) based anomaly detection. These methods identify outliers by evaluating the distance or density between a data point and its neighbors. However, Isolation Forest is generally more efficient when dealing with high-dimensional data and large-scale datasets.

[0070] After identifying the time point of sudden increase in voltage response time, it is necessary to extract historical meteorological data within a preset period prior to the identified time point and determine the key environmental condition combinations leading to sensor performance degradation through cluster analysis. The purpose of this step is to trace back to a period before the sensor performance degradation occurred, collect meteorological data such as ambient temperature and humidity during that period, and group this data using cluster analysis to discover specific environmental condition patterns or combinations closely related to sensor performance degradation. One implementation method is to use the K-Means clustering algorithm, which iteratively optimizes the allocation of data points into a preset number of K clusters, minimizing the distance between each data point and the centroid of its cluster. By analyzing the characteristics of each cluster (e.g., average temperature range, average humidity range), key environmental condition combinations that frequently occurred before sensor degradation can be identified. Another implementation method is to use the DBSCAN (Density-Based Spatial Clustering with Noise) algorithm, which can automatically discover clusters based on the density distribution of data points without pre-specifying the number of clusters and can effectively handle noise points, thus more flexibly identifying environmental condition combinations under different density distributions.

[0071] Subsequently, a game-theory-based attribution analysis algorithm quantifies the contribution weights of temperature and humidity factors to voltage response time decay within the key environmental condition combinations, generating a visual attribution report. This step aims to address the problem of how to fairly and accurately assess the contribution of each factor to sensor performance degradation under the combined effects of multiple factors (such as temperature and humidity). Game-theory attribution analysis algorithms can assign a quantified contribution weight to each influencing factor, thereby revealing the main drivers leading to sensor degradation. One implementation approach is to use the Shapley Value algorithm, derived from cooperative game theory, which assigns a contribution value to each participant (in this case, temperature and humidity), representing the participant's average marginal contribution to the total gain (in this case, voltage response time decay) across all possible cooperative alliances. By calculating the Shapley values ​​for temperature and humidity, their respective contribution weights can be quantified. Another approach is to employ interpretable AI methods such as LIME (Local Interpretable Model-agnostic Explanations) or SHAP (SHapley Additive exPlanations). These methods can interpret the predictions of any machine learning model by locally perturbing the input features and observing the changes in the model's output to estimate the contribution of each feature to the prediction, thereby quantifying the contributions of temperature and humidity. Finally, a visual attribution report is generated, presenting these contribution weights in an intuitive and easy-to-understand way to help users understand the specific reasons for sensor performance degradation.

[0072] This application's embodiments realize a shift from passive fault alarm to proactive lifespan prediction. The historical data acquisition module integrates multi-dimensional data to provide fundamental support for environmental stress assessment; the damp heat exposure calculation module transforms abstract environmental impacts into calculable physical quantities; the correlation model construction module establishes a quantitative mapping between environmental stress and performance degradation; and the remaining lifespan prediction module uses this to achieve dynamic early warning. The overall technical concept significantly improves prediction accuracy, enabling vehicle owners or maintenance personnel to take proactive maintenance measures before the sensor completely fails, avoiding emission exceedances or fuel economy degradation caused by sudden sensor failures. This provides forward-looking technical assurance for the reliability and fuel economy of vehicle emission control systems.

[0073] This application provides a specific embodiment in a practical application scenario where the oxygen sensor of a vehicle (e.g., vehicle A) needs to undergo lifespan prediction. Vehicle A travels and is parked in different geographical areas, experiencing various environmental conditions.

[0074] First, the system continuously acquires historical operational data and historical meteorological data for vehicle A. Historical operational data includes vehicle A's geographical location information (e.g., GPS coordinates), timestamps, and the voltage response time parameters of the oxygen sensor. Simultaneously, the system acquires ambient temperature and humidity data corresponding to vehicle A's geographical location and timestamp through an external meteorological service interface or onboard environmental sensors, using this data as historical meteorological data. When vehicle A is in a static parking scenario, the system acquires its static geographical location information and parking timestamp, and based on this information, obtains the corresponding target ambient temperature and humidity, incorporating them into the historical meteorological data to ensure accurate recording of its environmental heat and humidity exposure even when the vehicle is not in operation.

[0075] Next, based on the acquired ambient temperature and humidity, the system calculates the cumulative damp heat exposure of vehicle A over a preset period. For example, the system can cumulatively calculate the combined damp heat stress on the oxygen sensor caused by the temperature and humidity experienced by vehicle A each day over the past year. To improve the accuracy of the calculation, the system also acquires target sensor signals from vehicle A, such as solar radiation sensor signals and rain / humidity sensor signals. When vehicle A is stationary and the solar radiation sensor detects light intensity greater than a preset light threshold, the system determines that vehicle A is exposed to direct sunlight and assigns a first reduction factor value to correct the cumulative damp heat exposure. When vehicle A is stationary and the light intensity is lower than the preset light threshold, and the rain sensor detects no precipitation, the system determines that vehicle A is parked in a shaded state and assigns a second reduction factor value smaller than the first reduction factor value for correction. Furthermore, when the rain / humidity sensor detects relative humidity continuously higher than a second threshold for more than a preset duration, the system determines that vehicle A is in an extreme humidity scenario. In this case, the system increases the humidity sensitivity coefficient in the damp heat exposure calculation to more accurately reflect the impact of extreme humidity on the sensor. With these corrections, the calculation of cumulative damp heat exposure can more accurately reflect the actual exposure of vehicle A in different microenvironments, which significantly improves the accuracy of prediction compared to existing methods that rely solely on laboratory simulations or coarse environmental data.

[0076] Subsequently, the system constructs a correlation model between cumulative damp heat exposure and the voltage response time decay rate of the oxygen sensor. Specifically, based on the voltage response time parameter of the oxygen sensor, the system divides historical operating data into failed samples and truncated samples. Failed samples refer to those whose voltage response time exceeds a preset threshold or has triggered a fault code, while truncated samples refer to those whose sensors are still operational. The system constructs a joint likelihood function that includes the probability density function of failed samples and the survival function of truncated samples, based on the Weibull distribution model. By maximizing this joint likelihood function, the system estimates the shape and scale parameters of the Weibull distribution model, thereby establishing a correlation model between cumulative damp heat exposure and the voltage response time decay rate of the oxygen sensor. In this process, the system introduces an Arrhenius-damp heat coupling acceleration factor into the Weibull distribution model. This acceleration factor is based on environmental temperature and relative humidity to modify the scale parameters, reflecting the accelerating effect of high damp heat environment on the oxygen sensor's lifespan. When estimating the parameters, the system uses elastic network regression for regularization and applies physical constraints, such as the temperature effect satisfying the Arrhenius monotonically increasing characteristic, and the aging acceleration exhibiting nonlinear growth when relative humidity exceeds a first threshold. The introduction of these physical constraints and acceleration factors enables the model to more accurately capture the aging mechanism of oxygen sensors in real humid and hot environments, overcoming the problem of insufficient model generalization ability in existing technologies.

[0077] Finally, based on the constructed correlation model, the system calculates the remaining lifespan of the oxygen sensor using the current voltage response time and current cumulative damp heat exposure of vehicle A. For example, the system can predict how much longer or how many miles the oxygen sensor can continue to operate normally. Unlike existing technologies that only trigger an alarm when sensor performance deteriorates significantly and a fault code is triggered, this solution can predict the sensor's lifespan in advance. When the calculated remaining lifespan falls below a preset threshold (e.g., less than three months or 5,000 kilometers), the system immediately generates a warning message and sends it to the user of vehicle A or a service center. This proactive warning mechanism allows users to schedule repairs or replacements in advance, avoiding vehicle malfunctions and unnecessary losses due to sudden sensor failure.

[0078] Furthermore, to provide deeper diagnostic information, the system can identify time points in historical operational data where the oxygen sensor voltage response time suddenly increases. These time points typically indicate a sudden degradation in sensor performance. The system extracts historical meteorological data from a preset period prior to these time points and uses cluster analysis to determine the key environmental condition combinations that lead to sensor performance degradation (e.g., prolonged exposure to high temperature and humidity). Further, the system uses a game theory-based attribution analysis algorithm to quantify the contribution weights of temperature and humidity factors to voltage response time decay within these key environmental condition combinations and generates a visualized attribution report. This report visually demonstrates which environmental factors have the greatest impact on oxygen sensor aging, providing targeted maintenance recommendations for users and maintenance personnel—something difficult to achieve with existing technologies.

[0079] Please refer to Figure 6 As a response to the above Figure 1 The implementation of the method shown in this application provides an embodiment of an oxygen sensor lifetime prediction device, which is similar to... Figure 1 Corresponding to the method embodiments shown, the device can be specifically applied to various vehicles.

[0080] like Figure 6 As shown, the oxygen sensor lifetime prediction device in this embodiment includes: a historical data acquisition module 51, a damp heat exposure calculation module 52, a correlation model construction module 53, and a remaining lifetime prediction module 54, wherein: The historical data acquisition module 51 is used to acquire historical operating data and historical meteorological data of the target vehicle. The historical operating data includes vehicle geographical location information, timestamps and voltage response time parameters of the oxygen sensor, and the historical meteorological data includes ambient temperature and ambient humidity. The damp heat exposure calculation module 52 is used to calculate the cumulative damp heat exposure of the target vehicle within a preset time period based on the ambient temperature and the ambient humidity. The correlation model building module 53 is used to build a correlation model between the cumulative damp heat exposure and the voltage response time decay rate of the oxygen sensor. The remaining service life prediction module 54 is used to calculate the remaining service life of the oxygen sensor based on the correlation model, the current voltage response time of the target vehicle, and the current cumulative damp heat exposure, and generate a warning message when the remaining service life is lower than a preset threshold.

[0081] Furthermore, when the target vehicle is in a static parking scenario, the historical data acquisition module 51 includes: A static geographic location information acquisition unit is used to acquire the static geographic location information of the fixed location where the target vehicle is parked, and to acquire the parking timestamp of the target vehicle; The target environment humidity acquisition unit is used to acquire the corresponding target environment temperature and target environment humidity based on the static geographical location information and the parking timestamp; The data incorporation unit is used to incorporate the target ambient temperature and the target ambient humidity into the historical meteorological data.

[0082] Furthermore, the association model construction module 53 includes: The data partitioning unit is used to divide the historical operating data into failure samples and truncated samples based on the voltage response time parameter of the oxygen sensor, and to construct a status label variable. The failure sample refers to the sample whose voltage response time exceeds a preset threshold or has triggered a fault code, and the truncated sample refers to the sample whose sensor is still working. The joint likelihood function construction unit is used to construct a joint likelihood function based on the Weibull distribution model, which includes the probability density function of the failed samples and the survival function of the truncated samples. The association model generation unit is used to estimate the shape and scale parameters of the Weibull distribution model by maximizing the joint likelihood function, so as to construct an association model between the cumulative wet heat exposure and the voltage response time decay rate of the oxygen sensor.

[0083] Furthermore, the oxygen sensor's lifespan prediction device also includes: An acceleration factor introduction unit is used to introduce an Arrhenius-humid-thermal coupling acceleration factor into the Weibull distribution model, wherein the acceleration factor is based on ambient temperature and relative humidity, and modifies the scale parameter to reflect the accelerating effect of high humidity and heat environment on the lifetime of the oxygen sensor. The regularization processing unit is used to perform regularization processing using elastic network regression when estimating the shape parameters and scale parameters of the Weibull distribution model, and to apply at least one of the following physical constraints: the temperature effect satisfies the Arrhenius monotonically increasing property, and the aging acceleration increases nonlinearly when the relative humidity exceeds a first threshold.

[0084] Furthermore, the damp heat exposure calculation module 52 also includes: The target sensor signal acquisition module is used to acquire the target sensor signals of the target vehicle, wherein the target sensor signals include solar radiation sensor signals and rain / humidity sensor signals; An environmental exposure reduction factor determination module is used to determine the environmental exposure reduction factor based on the signal; The cumulative damp heat exposure correction module is used to correct the cumulative damp heat exposure using the environmental exposure reduction factor.

[0085] Furthermore, the environmental exposure reduction factor determination module includes: The first reduction factor value generation unit is used to determine that the target vehicle is exposed to the sun when the target vehicle is stationary and the sunlight radiation sensor detects that the light intensity is greater than the preset light threshold, and to assign the first reduction factor value. The second reduction factor value generation unit is used to determine that the target vehicle is in a shaded parking state when the target vehicle is stationary and the light intensity is lower than the preset light threshold, and the rain sensor has no precipitation signal, and assign a second reduction factor value, wherein the second reduction factor value is less than the first reduction factor value. The humidity sensitivity coefficient enhancement unit is used to determine that the target vehicle is in an extreme humidity scenario when the rain / humidity sensor detects that the relative humidity is continuously higher than the second threshold for more than a preset time, and to enhance the humidity sensitivity coefficient in the calculation of the cumulative damp heat exposure.

[0086] Furthermore, the oxygen sensor's lifespan prediction device also includes: The time node identification module is used to identify time nodes in the historical operating data where the oxygen sensor voltage response time suddenly increases, using the isolated forest algorithm. The key environmental condition combination determination module is used to extract historical meteorological data within a preset period before the time node and determine the key environmental condition combination that leads to sensor performance degradation through cluster analysis. The visualization attribution report generation module is used to quantify the contribution weights of temperature and humidity factors to voltage response time decay in the key environmental condition combination using a game theory-based attribution analysis algorithm, and generate a visualization attribution report.

[0087] To address the aforementioned technical problems, embodiments of this application also provide a vehicle. Please refer to the following for details. Figure 7 , Figure 7 This is a basic structural block diagram of the vehicle in this embodiment.

[0088] Vehicle 6 includes a memory 61, a processor 62, and a network interface 63 that are interconnected via a system bus. It should be noted that... Figure 7 The vehicle 6 shown only has three components: memory 61, processor 62, and network interface 63. However, it should be understood that it is not required to implement all of the components shown, and more or fewer components may be implemented instead.

[0089] The memory 61 includes at least one type of readable storage medium, including flash memory, hard disk, multimedia card, card-type memory (e.g., SD or DX memory), random access memory (RAM), static random access memory (SRAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), programmable read-only memory (PROM), magnetic memory, disk, optical disk, etc. In some embodiments, the memory 61 may be an internal storage unit of the vehicle 6, such as the hard disk or memory of the vehicle 6. In other embodiments, the memory 61 may also be an external storage device of the vehicle 6, such as a plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, etc., equipped on the vehicle 6. Of course, the memory 61 may also include both internal storage units and external storage devices of the vehicle 6. In this embodiment, the memory 61 is typically used to store the operating system and various application software installed on the vehicle 6, such as the program code of the oxygen sensor life prediction method. In addition, the memory 61 may also be used to temporarily store various types of data that have been output or will be output.

[0090] In some embodiments, processor 62 may be a central processing unit (CPU), controller, microcontroller, microprocessor, or other data processing chip. This processor 62 is typically used to control the overall operation of vehicle 6. In this embodiment, processor 62 is used to run program code stored in memory 61 or process data, for example, to run the program code of the oxygen sensor lifetime prediction method described above, to implement various embodiments of the oxygen sensor lifetime prediction method.

[0091] Network interface 63 may include a wireless network interface or a wired network interface, which is typically used to establish communication connections between vehicle 6 and other electronic devices.

[0092] This application also provides another embodiment, namely, a computer-readable storage medium storing a computer program that can be executed by at least one processor to cause the at least one processor to perform the steps of the oxygen sensor lifetime prediction method described above.

[0093] Obviously, the embodiments described above are merely some embodiments of this application, not all embodiments. The accompanying drawings show preferred embodiments of this application, but do not limit the scope of this application. This application can be implemented in many different forms; rather, these embodiments are provided to provide a more thorough and comprehensive understanding of the disclosure of this application. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing specific embodiments, or make equivalent substitutions for some of the technical features. Any equivalent structures made using the content of this application's specification and drawings, directly or indirectly applied to other related technical fields, are similarly within the scope of protection of this application.

Claims

1. A method for predicting the lifetime of an oxygen sensor, characterized in that, include: Acquire historical operating data and historical meteorological data of the target vehicle. The historical operating data includes vehicle geographical location information, timestamps, and voltage response time parameters of the oxygen sensor. The historical meteorological data includes ambient temperature and ambient humidity. Based on the ambient temperature and the ambient humidity, calculate the cumulative heat and humidity exposure of the target vehicle within a preset time period; Construct a correlation model between the cumulative damp heat exposure and the voltage response time decay rate of the oxygen sensor; Based on the aforementioned correlation model, the remaining service life of the oxygen sensor is calculated using the current voltage response time and current cumulative damp heat exposure of the target vehicle, and a warning message is generated when the remaining service life is lower than a preset threshold. The construction of the correlation model between the cumulative damp heat exposure and the voltage response time decay rate of the oxygen sensor includes: Based on the voltage response time parameter of the oxygen sensor, the historical operating data is divided into failure samples and truncated samples, and a status label variable is constructed. The failure sample refers to the sample whose voltage response time exceeds a preset threshold or has triggered a fault code, and the truncated sample refers to the sample whose sensor is still working. A joint likelihood function is constructed based on the Weibull distribution model, which includes the probability density function of the failed samples and the survival function of the truncated samples. By maximizing the joint likelihood function, the shape and scale parameters of the Weibull distribution model are estimated to construct a correlation model between the cumulative damp heat exposure and the voltage response time decay rate of the oxygen sensor.

2. The method for predicting the lifetime of an oxygen sensor according to claim 1, characterized in that, When the target vehicle is in a static parking scenario, the acquisition of the target vehicle's historical operating data and historical meteorological data includes: Obtain the static geographic location information of the fixed location where the target vehicle is parked, and obtain the parking timestamp of the target vehicle; Based on the static geographic location information and the parking timestamp, obtain the corresponding target ambient temperature and target ambient humidity; The target ambient temperature and the target ambient humidity are incorporated into the historical meteorological data.

3. The method for predicting the lifetime of an oxygen sensor according to claim 1, characterized in that, The method further includes: An Arrhenius-humid-thermal coupling acceleration factor is introduced into the Weibull distribution model, wherein the acceleration factor is based on ambient temperature and relative humidity, and the scale parameter is modified to reflect the accelerating effect of high humidity and heat environment on the lifetime of the oxygen sensor. When estimating the shape and scale parameters of the Weibull distribution model, regularization is performed using elastic network regression, and at least one of the following physical constraints is applied: the temperature effect satisfies the Arrhenius monotonically increasing property, and the aging acceleration exhibits nonlinear growth when the relative humidity exceeds a first threshold.

4. The method for predicting the lifetime of an oxygen sensor according to claim 1, characterized in that, After calculating the cumulative heat and humidity exposure of the target vehicle within a preset time period based on the ambient temperature and humidity, the method further includes: Acquire target sensor signals from the target vehicle, wherein the target sensor signals include solar radiation sensor signals and rain / humidity sensor signals; The environmental exposure reduction factor is determined based on the signal; The cumulative damp heat exposure is corrected using the environmental exposure reduction factor.

5. The method for predicting the lifetime of an oxygen sensor according to claim 4, characterized in that, The determination of the environmental exposure reduction factor based on the signal includes: When the target vehicle is stationary and the sunlight radiation sensor detects that the light intensity is greater than the preset light threshold, it is determined that the target vehicle is exposed to the sun and a first reduction factor value is assigned. When the target vehicle is stationary and the light intensity is lower than the preset light threshold, and the rain sensor has no precipitation signal, it is determined that the target vehicle is in a shaded parking state and a second reduction factor value is assigned, wherein the second reduction factor value is less than the first reduction factor value. When the rainfall / humidity sensor detects that the relative humidity is continuously higher than the second threshold for more than a preset time, it determines that the target vehicle is in an extreme humidity scenario and increases the humidity sensitivity coefficient in the calculation of the cumulative damp heat exposure.

6. The method for predicting the lifetime of an oxygen sensor according to any one of claims 1 to 5, characterized in that, After calculating the remaining lifespan of the oxygen sensor based on the correlation model using the target vehicle's current voltage response time and current cumulative damp heat exposure, and generating a warning message when the remaining lifespan falls below a preset threshold, the method further includes: The isolated forest algorithm was used to identify the time points in the historical operating data where the oxygen sensor voltage response time suddenly increased. Historical meteorological data within a preset time period prior to the stated time point are extracted, and cluster analysis is used to determine the key environmental condition combinations that lead to sensor performance degradation. The contribution weights of temperature and humidity factors to voltage response time decay in the key environmental condition combinations are quantified using a game theory-based attribution analysis algorithm, and a visual attribution report is generated.

7. A lifespan prediction device for an oxygen sensor, characterized in that, include: The historical data acquisition module is used to acquire historical operating data and historical meteorological data of the target vehicle. The historical operating data includes vehicle geographical location information, timestamps and voltage response time parameters of the oxygen sensor, and the historical meteorological data includes ambient temperature and ambient humidity. The damp heat exposure calculation module is used to calculate the cumulative damp heat exposure of the target vehicle within a preset time period based on the ambient temperature and the ambient humidity. The correlation model building module is used to build a correlation model between the cumulative damp heat exposure and the voltage response time decay rate of the oxygen sensor. The remaining service life prediction module is used to calculate the remaining service life of the oxygen sensor based on the correlation model, the current voltage response time of the target vehicle, and the current cumulative damp heat exposure, and generate a warning message when the remaining service life is lower than a preset threshold. The association model construction module includes: The data partitioning unit is used to divide the historical operating data into failure samples and truncated samples based on the voltage response time parameter of the oxygen sensor, and to construct a status label variable. The failure sample refers to the sample whose voltage response time exceeds a preset threshold or has triggered a fault code, and the truncated sample refers to the sample whose sensor is still working. The joint likelihood function construction unit is used to construct a joint likelihood function based on the Weibull distribution model, which includes the probability density function of the failed samples and the survival function of the truncated samples. The association model generation unit is used to estimate the shape and scale parameters of the Weibull distribution model by maximizing the joint likelihood function, so as to construct an association model between the cumulative wet heat exposure and the voltage response time decay rate of the oxygen sensor.

8. A vehicle, characterized in that, It includes a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the oxygen sensor lifetime prediction method as described in any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the lifetime prediction method for the oxygen sensor as described in any one of claims 1 to 6.

Images