Remote vehicle testing methods

Abstract

This invention relates to the field of automotive remote testing technology, specifically to a method for automotive remote testing. The invention utilizes a power supply status monitoring unit within the vehicle's vehicle networking module to capture in real-time waveforms of onboard power bus voltage fluctuations, sensor power supply line ripple coefficients, and transient operating current response curves of key electronic components. Based on this data, a three-dimensional power supply quality evaluation model is established to generate a power supply stability index, which is simultaneously encapsulated when transmitting vehicle operating data. Upon receiving the data on the cloud platform, a power supply perception data fusion module aligns the data in time and space, an adaptive filtering module dynamically denoises the data, and a dual-mode fault discrimination module operates in parallel with a benchmark and an elastic fault tree, comparing their confidence levels. An early warning is triggered only when both fault trees report the same fault code, effectively solving the problem of data deviation and false fault misjudgment caused by dynamic power supply fluctuations in existing technologies.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of automobile remote detection, in particular to an automobile remote detection method. BACKGROUND

[0002] With the improvement of automobile electronicization and intelligence level, automobile remote detection technology has become an important means to ensure the safe operation of vehicles and reduce maintenance costs. This technology mainly collects the running data of key components such as engine, transmission, brake system, electronic stability program, etc. through the vehicle networking module on the vehicle, such as speed, oil pressure, temperature, fault code, etc. Then these data are transmitted to the cloud management platform through the mobile network. After the cloud platform preliminarily processes the data, the result will be fed back to the vehicle owner or automobile service provider, so that the relevant personnel can timely master the running state of the vehicle without connecting the vehicle on site, and arrange maintenance in advance to avoid traffic safety hazards or maintenance delays caused by sudden failures.

[0003] The existing automobile remote detection method does not fully consider the data accuracy problem of the vehicle in the "dynamic power supply fluctuation scene" when collecting the running data of key components. When the vehicle encounters frequent start-stop (such as urban congested road sections), air conditioner and light are turned on at the same time, and the instantaneous load of the generator changes, etc. during driving, the voltage of the vehicle power supply system will fluctuate for a short time, and the fluctuation range is usually ±0.5V deviation within the standard range of 12V-14.5V. This voltage fluctuation will directly affect the signal output accuracy of electronic component sensors, causing slight deviations in the collected speed, oil pressure, etc. The cloud data processing algorithm of the existing remote detection method does not set a real-time correction mechanism for the data deviation caused by such power supply fluctuation, and still analyzes the data according to the calibration parameters under the standard power supply state, which is easy to misjudge the "false data anomaly" caused by power supply fluctuation as a fault of the component itself, and then send an error fault warning to the vehicle owner or service provider, which not only increases the unnecessary maintenance workload, but also reduces the user's trust in the reliability of the remote detection technology. SUMMARY

[0004] In view of the shortcomings of the prior art, the present application provides an automobile remote detection method, which solves the problem of sensor data deviation and false fault misjudgment caused by dynamic power supply fluctuation when compared with the prior art.

[0005] To achieve the above purpose, the present application realizes the following technical scheme: an automobile remote detection method, comprising:

[0006] In the vehicle Internet of Things module of the vehicle, the voltage fluctuation waveform of the vehicle-mounted power supply bus, the ripple coefficient of each sensor power supply circuit and the transient operating current response curve of the key electronic components are captured in real time by a power supply state monitoring unit, and a three-dimensional power supply quality evaluation model is established based on the waveform, the ripple coefficient and the transient response curve, and a power supply stability index is dynamically generated, which maps the potential influence degree of the current vehicle power supply environment on sensor data acquisition in a digital twin manner;

[0007] The vehicle Internet of Things module synchronously encapsulates the power supply stability index when transmitting the vehicle operation data;

[0008] The cloud platform receives the vehicle operation data and the power supply stability index, and performs dynamic compensation processing on the vehicle operation data based on the power supply stability index, which includes:

[0009] The power supply stability index and the vehicle operation data are spatio-temporally aligned by the power supply perception data fusion module to identify the coupling relationship between power supply fluctuations and sensor data anomalies;

[0010] Through the adaptive filtering module, a variable parameter digital filter is used to dynamically adjust the filter cutoff frequency and bandwidth of the variable parameter digital filter according to the power supply stability index, and the fluctuation sensitive parameters are dynamically denoised;

[0011] Through the dual-mode fault diagnosis module, a reference fault tree under normal power supply state and an elastic fault tree under dynamic power supply state are established, and the two diagnostic logics are run in parallel, and by comparing the confidence difference of the diagnostic results under the two modes, the false fault signals caused by power supply fluctuations are eliminated;

[0012] When and only when the same fault code is reported in both modes, the pre-warning information is triggered to be pushed.

[0013] Further, the power supply state monitoring unit captures the voltage fluctuation waveform of the vehicle-mounted power supply bus, the ripple coefficient of each sensor power supply circuit and the transient operating current response curve of the key electronic components in real time, including:

[0014] Multi-channel synchronous sampling technology is used to obtain the voltage change data of the vehicle-mounted power supply bus within a preset time period in real time;

[0015] The periodic ripple voltage amplitude of each sensor power supply circuit is obtained in real time;

[0016] The current response transient value of the key electronic components under different working loads is obtained in real time.

[0017] Further, the three-dimensional power supply quality evaluation model is established based on the waveform, ripple coefficient and transient response curve, and a power supply stability index is dynamically generated, which maps the potential influence degree of the current vehicle power supply environment on sensor data acquisition in a digital twin manner, including:

[0018] The voltage change data, ripple voltage amplitude and current response transient value are input into the pre-trained three-dimensional power supply quality evaluation model;

[0019] The three-dimensional power supply quality evaluation model outputs a dynamic power supply stability index based on the input data, and the size of the index indicates the influence degree of the power supply environment on the sensor data acquisition accuracy.

[0020] Further, the power supply stability index is spatiotemporally aligned with the vehicle operation data by the power supply perception data fusion module to identify the coupling relationship between power supply fluctuations and sensor data anomalies, including:

[0021] The timestamps of vehicle operation data acquisition and the timestamps of power supply stability index generation are obtained;

[0022] The power supply stability index and the vehicle operation data are aligned according to the timestamps;

[0023] The aligned data is analyzed to determine whether the change of the power supply stability index is related to the abnormal fluctuations in the vehicle operation data.

[0024] Further, the adaptive filtering module is used to dynamically adjust the filter cutoff frequency and bandwidth of the variable parameter digital filter according to the power supply stability index, and to perform dynamic noise reduction processing on the fluctuation-sensitive parameters, including:

[0025] The filter cutoff frequency and bandwidth of the variable parameter digital filter are calculated and updated in real time according to the power supply stability index;

[0026] The fluctuation-sensitive parameters are input into the variable parameter digital filter for processing to achieve intelligent trap wave within the power supply noise frequency band and eliminate the noise components introduced by power supply fluctuations.

[0027] Further, the dual-mode fault discrimination module is used to establish a reference fault tree under normal power supply state and an elastic fault tree under dynamic power supply state, and the two diagnostic logics are run in parallel, and the false fault signals caused by power supply fluctuations are eliminated by comparing the confidence difference of the diagnostic results in the two modes, including:

[0028] Based on the vehicle historical operation data and the standard fault mode, a reference fault tree under normal power supply state is constructed for diagnosing faults under stable power supply conditions;

[0029] Based on the power fluctuation simulation data and the dynamic calibration parameters, a dynamic power supply state elastic fault tree is constructed for diagnosing faults under fluctuating power supply conditions.

[0030] The denoised vehicle operation data is input into the reference fault tree and the elastic fault tree for diagnosis, and the respective diagnostic results and confidence levels are obtained.

[0031] Further, the confidence level difference between the diagnostic results in the two modes is compared to eliminate false fault signals caused by power fluctuations; only when the same fault code is reported in both modes, an early warning information push is triggered, including:

[0032] If the diagnostic results of the reference fault tree and the elastic fault tree are inconsistent, or the diagnostic results of the two modes are consistent but the confidence level difference exceeds a preset threshold, the diagnostic result is identified as a false fault signal and is eliminated;

[0033] If the reference fault tree and the elastic fault tree both diagnose the same fault code, and the confidence level difference of the diagnostic results of the two modes is within a preset threshold range, it is identified as a real fault, and an early warning information push is triggered.

[0034] Further, the vehicle operation data includes engine speed, oil pressure, transmission temperature and brake system state parameters; the fluctuation sensitive parameters include engine speed data and oil pressure data.

[0035] Further, the vehicle networking module synchronously encapsulates the power supply stability index when transmitting the vehicle operation data, including:

[0036] The vehicle operation data and the power supply stability index are packaged into a unified data frame format;

[0037] The data frame is transmitted to the cloud platform through a mobile network.

[0038] Further, the cloud platform receives the vehicle operation data and the power supply stability index, including:

[0039] The cloud platform receives the data frame from the vehicle networking module through a mobile network interface;

[0040] The cloud platform decapsulates the data frame to obtain the vehicle operation data and the power supply stability index.

[0041] Compared with the prior art, the present application has the following advantages:

[0042] The application establishes a three-dimensional power supply quality evaluation model based on the data to generate a power supply stability index, and synchronously encapsulates the index when transmitting vehicle operation data; after being received by the cloud platform, the power supply sensing data fusion module realizes data space-time alignment, the adaptive filtering module dynamically denoises, the double-mode fault discrimination module runs the parallel operation benchmark and the elastic fault tree and compares the confidence, and only when the two fault trees report the same fault code, the early warning is triggered, effectively solving the problem that the prior art does not consider dynamic power fluctuation to cause data deviation and false fault misjudgment, effectively improving the accuracy of automobile remote detection, reducing unnecessary maintenance and troubleshooting workload, enhancing the trust of users on the remote detection technology, and guaranteeing the reliability and effectiveness of vehicle fault early warning. BRIEF DESCRIPTION OF DRAWINGS

[0043] Figure 1 The power supply monitoring and index generation flowchart of the application is shown in the figure;

[0044] Figure 2 The data encapsulation and transmission flowchart of the application is shown in the figure;

[0045] Figure 3 The cloud data processing flowchart of the application is shown in the figure;

[0046] Figure 4 The double-mode fault discrimination flowchart of the application is shown in the figure. DETAILED DESCRIPTION

[0047] The technical solutions in the embodiments of the application will be described clearly and completely below with reference to the drawings in the embodiments of the application. Obviously, the described embodiments are only part of the embodiments of the application, rather than all the embodiments. Based on the embodiments in the application, all other embodiments obtained by those skilled in the art without creative labor fall within the protection scope of the application.

[0048] Please refer to Figures 1-4 The application provides an automobile remote detection method, which comprises:

[0049] In the vehicle networking module of the vehicle, the power supply state monitoring unit captures the voltage fluctuation waveform of the vehicle-mounted power supply bus, the ripple coefficient of each sensor power supply circuit, and the transient working current response curve of the key electronic components in real time, and establishes a three-dimensional power supply quality evaluation model based on the waveform, the ripple coefficient and the transient response curve, dynamically generates a power supply stability index, and maps the potential influence degree of the current vehicle power supply environment on sensor data acquisition in a digital twin manner.

[0050] The vehicle networking module synchronously encapsulates the power supply stability index when transmitting the vehicle operation data.

[0051] The cloud platform receives the vehicle operation data and the power supply stability index, and performs dynamic compensation processing on the vehicle operation data based on the power supply stability index. The dynamic compensation processing includes:

[0052] The power supply sensing data fusion module performs spatio-temporal alignment of the power supply stability index and the vehicle operation data, and identifies the coupling relationship between power supply fluctuations and sensor data anomalies.

[0053] The adaptive filtering module adopts a variable parameter digital filter, dynamically adjusts the filter cutoff frequency and bandwidth of the variable parameter digital filter according to the power supply stability index, and performs dynamic noise reduction processing on the fluctuation-sensitive parameters.

[0054] The dual-mode fault diagnosis module establishes a reference fault tree under normal power supply state and an elastic fault tree under dynamic power supply state, and runs the two diagnostic logics in parallel. By comparing the confidence difference of the diagnostic results under the two modes, false fault signals caused by power supply fluctuations are eliminated.

[0055] The warning information is triggered and pushed only when the same fault code is reported by both modes.

[0056] Specifically, in the vehicle networking module, a power supply state monitoring unit is installed. This unit operates continuously to capture three types of key data in real time. When the vehicle is in a dynamic power fluctuation scenario such as frequent start-stop in urban congestion sections, simultaneous opening of air conditioner and lights, etc., the vehicle power bus voltage will have a ±0.5V deviation within the 12V-14.5V standard range. The monitoring unit captures this voltage fluctuation waveform in real time through multi-channel synchronous sampling technology (existing mature technology widely used in multi-signal parallel sampling scenarios); at the same time, it obtains the ripple coefficient of each sensor power supply line in real time, such as the periodic ripple voltage amplitude of the engine speed sensor power supply line; it also tracks the transient operating current response curve of key electronic components such as ECU under different workloads.

[0057] Based on the above-mentioned captured waveforms, ripple coefficients and transient response curves, a three-dimensional power supply quality evaluation model is constructed. This model is trained by a large amount of historical power supply data and sensor acquisition accuracy data, and generates a power supply stability index dynamically through a formula. The specific formula is as follows:

[0058] ; (1)

[0059] wherein, is the power supply stability index, with a value range of 0-10. The smaller the value, the less the impact of the power supply environment on the sensor acquisition accuracy, and the larger the value, the greater the impact. , , respectively, the sum of the three is 1, set after multiple experimental calibration , , ; is the normalized value of the voltage fluctuation waveform, which is calculated by comparing the actual voltage fluctuation amplitude with the standard voltage interval deviation range, and the value is 0-10; is the normalized value of the ripple coefficient, which is converted from the ratio of the actual ripple voltage amplitude to the sensor allowed ripple threshold, and the value is 0-10; is the normalized value of the transient working current response curve, which is determined according to the proportion of the current response fluctuation amplitude and the rated current fluctuation range of the electronic component, and the value is 0-10.

[0060] This index intuitively maps the potential impact of the current power supply environment on sensor data collection in a digital twin manner. For example, when the index rises above 6, it means that power fluctuations may cause significant deviations in the speed and oil pressure data collected by the sensor.

[0061] When transmitting engine speed, oil pressure and other vehicle operation data, the vehicle networking module encapsulates the power supply stability index and operation data into a unified data frame, and synchronously sends it to the cloud platform through the mobile network. After receiving the data, the cloud platform starts dynamic compensation processing: first, the power supply awareness data fusion module aligns the power supply stability index and operation data in space and time according to the timestamp (timestamp matching is the existing data synchronization technology), accurately identifies the coupling relationship between power supply fluctuations and sensor data anomalies, such as whether the sudden drop in power supply index is synchronized with abnormal oil pressure data; Then, through the adaptive filtering module, the cutoff frequency and bandwidth of the variable parameter digital filter are dynamically adjusted according to the power supply stability index, and the fluctuation sensitive parameters such as engine speed are denoised to eliminate the noise introduced by power supply fluctuations; Finally, the dual-mode fault diagnosis module runs the baseline fault tree based on normal power supply data and the elastic fault tree based on power supply fluctuation data in parallel (the fault tree is constructed by using the existing tree logic modeling method in the field of fault diagnosis), and compares the confidence of the two diagnosis results.

[0062] Only when both fault trees report the same fault code and the confidence difference is within the preset range, an early warning information push is triggered. This embodiment effectively solves the problem of data deviation and false fault misdiagnosis caused by dynamic power fluctuations in the prior art, significantly improves the accuracy of remote detection, reduces unnecessary maintenance and troubleshooting work, and enhances user trust in detection technology.

[0063] In this embodiment, the power supply state monitoring unit captures the voltage fluctuation waveform of the vehicle-mounted power bus, the ripple coefficient of each sensor power supply line, and the transient working current response curve of the key electronic components in real time, including:

[0064] The multi-channel synchronous sampling technology is used to obtain the voltage change data of the vehicle-mounted power bus in a preset time period in real time.

[0065] The periodic ripple voltage amplitude of each sensor power supply circuit is obtained in real time.

[0066] The current response transient value of the key electronic components under different working loads is obtained in real time.

[0067] Specifically, the power supply state monitoring unit uses the multi-channel synchronous sampling technology (existing mature technology, widely used in industrial data acquisition field) to carry out data capture work. For the voltage fluctuation waveform of the vehicle-mounted power bus, 100 milliseconds is set as the preset time period, and the voltage change data of the power bus in this time period is collected simultaneously through the multi-channel synchronous sampling module, ensuring that the instantaneous fluctuation of the voltage under frequent start-stop and other scenarios can be captured completely, and the loss of waveform information caused by asynchronous sampling is avoided.

[0068] For the ripple coefficient of each sensor power supply circuit, the monitoring unit monitors the periodic ripple voltage amplitude of each sensor power supply circuit in real time through the built-in ripple detection circuit. Whether it is an engine speed sensor, a brake system sensor or a transmission temperature sensor, the ripple condition of the line can be continuously obtained, and the power supply ripple data of any sensor is not missed, providing a comprehensive basis for subsequent judgment of whether the sensor data is affected by the ripple.

[0069] In the capture of the transient working current response curve of the key electronic components, the monitoring unit cooperates with the current sampling resistor and the high-speed ADC module to track the state of the electronic components under different working loads, such as the increase of ECU load when the vehicle accelerates, the decrease of load when decelerating, the change of generator load when starting high-power electrical equipment, etc. The current response transient value of the electronic components under corresponding working conditions is collected in real time, thereby forming a complete transient working current response curve. Through the above-mentioned manner, the power supply related data can be obtained comprehensively and in real time, providing accurate and comprehensive basic data for subsequent construction of power supply quality evaluation model by substituting formula (1), and ensuring the reliability of subsequent index generation.

[0070] In this embodiment, a three-dimensional power supply quality evaluation model is established based on the waveform, ripple coefficient and transient response curve, and a power supply stability index is dynamically generated. The power supply stability index maps the potential influence degree of the current vehicle power supply environment on sensor data acquisition in a digital twin way, including:

[0071] The voltage change data, ripple voltage amplitude and current response transient value are input into the pre-trained three-dimensional power supply quality evaluation model;

[0072] The three-dimensional power supply quality evaluation model is based on input data and outputs a dynamic power supply stability index. The size of the index indicates the degree of influence of the power supply environment on the sensor data collection accuracy.

[0073] Specifically, the three-dimensional power supply quality evaluation model is based on a large amount of historical power supply data (covering voltage, ripple, and current data under different power supply fluctuation scenarios) and sensor collection accuracy data under the corresponding scenarios. Through machine learning algorithm iterative optimization of model parameters, the model has the ability to accurately analyze the influence of power supply state on sensor collection.

[0074] The voltage variation data, ripple voltage amplitude, and current response transient value captured by the power supply state monitoring unit are normalized respectively to obtain , , three parameters, and then calculated according to formula (1) to dynamically output the power supply stability index . For example, when the vehicle simultaneously starts the air conditioner and the light, the voltage fluctuation amplitude increases from 2 to 5, the increase in ripple voltage amplitude makes from 1 to 3, and the electronic component current response fluctuation intensifies from 1 to 2, and by substituting the formula, we can get , indicating that the power supply environment has a certain influence on sensor collection accuracy.

[0075] The index is presented in a numerical range of 0-10, and is visually displayed in the vehicle digital model constructed on the cloud platform in the form of digital twinning. Through the digital model, the staff can clearly know whether the current power supply state will interfere with the sensor data, providing clear basis for subsequent processing of vehicle operation data and improving the pertinence of data processing.

[0076] In this embodiment, the power supply stability index and vehicle operation data are spatio-temporally aligned by the power supply perception data fusion module to identify the coupling relationship between power supply fluctuation and sensor data anomaly, including:

[0077] Obtain the time stamp of vehicle operation data collection and the time stamp of power supply stability index generation;

[0078] Align the power supply stability index and vehicle operation data according to the time stamp;

[0079] Analyze the aligned data to determine whether the change in the power supply stability index is related to the abnormal fluctuation in the vehicle operation data.

[0080] Specifically, the power supply awareness data fusion module first acquires the time stamp of the vehicle operation data collection and the time stamp of the power supply stability index generation, both of which are accurate to the millisecond level, ensuring the accuracy of the time record. This process is implemented using the existing time synchronization protocol.

[0081] Subsequently, the module aligns the power supply stability index with the vehicle operation data according to the time stamp, so that the power supply index at the same time point corresponds to the operation data one by one. For example, the power supply stability index generated at 10:05:23.120 is associated with the data collected at that time, such as engine speed 2500 rpm, oil pressure 3.2 bar, etc. After alignment, the module analyzes the data and determines whether the change of the power supply stability index is coupled with the abnormal operation data by calculating the correlation of their change trends. For example, when the power supply stability index suddenly rises from 3.0 to 7.5 at a certain time, and the engine speed data suddenly jumps from 2200 rpm to 2800 rpm, and the power supply index falls back, and the speed data also returns to normal, it can be determined that the power supply fluctuation is coupled with the abnormal speed data.

[0082] In this way, the influence of power supply fluctuation on sensor data abnormality can be accurately located, laying a foundation for subsequent targeted data processing and avoiding blind adjustment of operation data.

[0083] In this embodiment, the adaptive filtering module uses a variable parameter digital filter to dynamically adjust the filter cutoff frequency and bandwidth of the variable parameter digital filter according to the power supply stability index, and performs dynamic noise reduction processing on the fluctuation sensitive parameters, including:

[0084] According to the power supply stability index, the filter cutoff frequency and bandwidth of the variable parameter digital filter are calculated and updated in real time;

[0085] The fluctuation sensitive parameters are input to the variable parameter digital filter for processing to achieve intelligent trap wave in the power supply noise frequency band and eliminate the noise components introduced by power supply fluctuation.

[0086] Specifically, the adaptive filtering module carries a variable parameter digital filter, and the filter cutoff frequency and bandwidth of the filter can be adjusted in real time according to the power supply stability index. The cutoff frequency is adjusted by the formula, which is as follows:

[0087] ; (2)

[0088] In the formula, is the adjusted filter cutoff frequency; is the reference cutoff frequency, which is set to 1 kHz according to the effective signal frequency band of the vehicle sensor data; ​is the frequency adjustment coefficient, calibrated as 50 Hz / unit SI through experiments; is the current power supply stability index; is the reference power supply stability index, set as 3.0, representing the critical value of the basic stability of the power supply environment.

[0089] When the power supply stability index (higher than the reference value), it indicates that the power supply fluctuation has a greater impact on the sensor data, and substituting into formula (2) can obtain At the same time, the filter bandwidth is adjusted to be in a fixed proportion to the cutoff frequency, that is, is the filter bandwidth, and 0.2 is the proportion coefficient), at this time the bandwidth becomes 220 Hz, by increasing the cutoff frequency and narrowing the bandwidth, the filtering ability of high-frequency power supply noise is enhanced; when the power supply stability index (lower than the reference value), the power supply environment is relatively stable, and substituting into the formula obtains , the bandwidth becomes 190 Hz, avoiding filtering out the effective information in the running data.

[0090] In the filtering process, the engine speed data, oil pressure data and other fluctuation sensitive parameters are input into the variable parameter digital filter, and the filter realizes intelligent wave trapping in the power supply noise frequency band according to the adjusted cutoff frequency and bandwidth, and accurately eliminates the noise components introduced due to power supply fluctuation. After processing, the accuracy of the fluctuation sensitive parameters is greatly improved, providing more reliable data support for subsequent fault diagnosis.

[0091] In this embodiment, through the dual-mode fault judgment module, the reference fault tree under normal power supply state and the elastic fault tree under dynamic power supply state are established, and the two kinds of diagnosis logics are run in parallel, by comparing the confidence difference of the diagnosis results in the two modes, the pseudo-fault signals caused by power supply fluctuation are eliminated, including:

[0092] Based on the vehicle historical running data and the standard fault mode, the reference fault tree under normal power supply state is constructed, which is used for diagnosing faults under stable power supply conditions;

[0093] Based on the power supply fluctuation simulation data and the dynamic calibration parameters, the elastic fault tree under dynamic power supply state is constructed, which is used for diagnosing faults under fluctuating power supply conditions;

[0094] The denoised vehicle running data is input into the reference fault tree and the elastic fault tree for diagnosis at the same time, and the respective diagnosis results and confidence are obtained.

[0095] ​Specifically, the dual-mode fault discrimination module first constructs two types of fault trees. The construction of the benchmark fault tree is based on the historical running data accumulated during long-term vehicle operation under normal power supply conditions, combined with the standard fault modes commonly used in the automotive industry (such as determining that the engine oil pressure is lower than 2.0 bar as an abnormal pressure), and through existing fault tree analysis methods, the characteristics and judgment logic of various faults under stable power supply conditions are determined for fault diagnosis under stable power supply conditions.

[0096] The elastic fault tree is constructed based on the power supply fluctuation simulation data obtained through simulation experiments (covering different amplitude and frequency power supply fluctuation scenarios) and the dynamic calibration parameters (such as the engine oil pressure fault judgment threshold being lowered by 0.1 bar for every 1.0 increase in the power supply index) calibrated according to the actual power supply fluctuation conditions. It can adapt to the fault diagnosis requirements under power supply fluctuation scenarios, and its construction method is also based on the existing fault tree modeling logic, only optimizing the parameters for power supply fluctuation scenarios.

[0097] When the vehicle running data after adaptive filtering is obtained, it is input into the benchmark fault tree and the elastic fault tree. The two fault trees analyze the data according to their own diagnosis logic and output their respective diagnosis results, while giving the confidence of the corresponding diagnosis results (confidence is calculated by matching the fault characteristics with the data, with a value of 0-1, the closer to 1, the more reliable), which provides a key basis for subsequent elimination of false fault signals.

[0098] In this embodiment, by comparing the confidence differences of the diagnosis results under the two modes, false fault signals caused by power supply fluctuations are eliminated; only when both modes report the same fault code, the pre-warning information is triggered, including:

[0099] If the diagnosis results of the benchmark fault tree and the elastic fault tree are inconsistent, or the diagnosis results of the two modes are consistent but the confidence difference exceeds the preset threshold, the diagnosis result is identified as a false fault signal and is eliminated;

[0100] If the benchmark fault tree and the elastic fault tree both diagnose the same fault code, and the confidence difference of the diagnosis results of the two modes is within the preset threshold, it is identified as a real fault, and the pre-warning information is triggered.

[0101] Specifically, after obtaining the diagnosis results and confidence of the benchmark fault tree and the elastic fault tree, the dual-mode fault discrimination module compares and analyzes them. If the benchmark fault tree determines that there is a gearbox temperature too high (more than 120℃) fault with a confidence of 0.9, while the elastic fault tree does not detect this fault with a confidence of 0.1; or both determine that there is this fault, but the confidence of the benchmark fault tree is 0.9 and the confidence of the elastic fault tree is 0.6, the difference exceeds the preset threshold of 0.2, then the diagnosis result is determined as a false fault signal and is eliminated, avoiding misjudgment caused by power supply fluctuations.

[0102] Only when the reference fault tree and the elastic fault tree diagnose the same fault code, such as both reporting the engine oil pressure is too low (less than 2.0 bar) fault, and the reference fault tree confidence is 0.85, the elastic fault tree confidence is 0.78, the difference between the two is 0.07, which is within the preset threshold 0.2, it is determined as a real fault, and the early warning information is triggered to push the fault information to the vehicle owner and the automobile service provider.

[0103] In this way, the false fault signal caused by power supply fluctuation is effectively eliminated, ensuring that the early warning information pushed is accurate and reliable, reducing unnecessary maintenance scheduling, and improving service efficiency.

[0104] In this embodiment, the vehicle operating data includes engine speed, oil pressure, transmission temperature and brake system state parameters; the fluctuation sensitive parameters include engine speed data and oil pressure data.

[0105] Specifically, the vehicle operating data covers engine speed, oil pressure, transmission temperature and brake system state parameters (such as brake pad wear, brake fluid level) and other key data, which are collected in real time by sensors mounted on each part of the vehicle, and can fully reflect the operating status of the core components of the vehicle: engine speed reflects the working strength of the engine, abnormal speed may reflect the ignition system or fuel supply problem; transmission temperature reflects the transmission cooling and operating condition, and high temperature may mean poor lubrication; brake system state parameters are directly related to driving safety, and need to be handled immediately when abnormal.

[0106] Among these operating data, engine speed data and oil pressure data are fluctuation sensitive parameters. This is because the signal output of engine speed sensor and oil pressure sensor is analog signal, which is easily affected by power supply voltage fluctuation. When the power supply voltage deviates by ±0.5V, the data collected by these two types of sensors will first show a slight deviation, and if not handled specifically, it will easily lead to fault misjudgment. Identifying fluctuation sensitive parameters can make subsequent adaptive filtering based on formula (2) more targeted, avoid over-filtering of transmission temperature and other data less affected by power supply, and improve data processing efficiency and accuracy.

[0107] In this embodiment, the Internet of Vehicles module encapsulates the power supply stability index when transmitting the vehicle operating data, including:

[0108] Packaging vehicle operating data and power supply stability index into a unified data frame format;

[0109] Transmit the data frame to the cloud platform through the mobile network.

[0110] Specifically, the vehicle networking module processes the vehicle operation data before transmitting the vehicle operation data. The vehicle operation data such as engine speed and oil pressure is packaged into a unified data frame in a preset format together with the power supply stability index calculated by formula (1). The data frame includes data identification (used to distinguish different vehicle data), operation data field (stored according to sensor type), power supply stability index field (SI value is stored separately), and the like, which ensures clear data structure and facilitates cloud platform analysis. The data frame format follows the existing vehicle networking data transmission protocol, and only the power supply stability index field is added.

[0111] After the data frame is packaged, the vehicle networking module transmits it to the cloud platform through the mobile network (such as 4G, 5G network) carried by the vehicle. The network transmission adopts existing wireless communication technology to ensure the stability and real-time performance of data transmission. The synchronous packaging and transmission method avoids the problem of asynchronous transmission of operation data and power supply stability index, ensures that the cloud platform can simultaneously obtain both types of data, provides guarantee for subsequent dynamic compensation processing, and improves the coherence of data processing.

[0112] In this embodiment, the cloud platform receives vehicle operation data and power supply stability index, including:

[0113] The cloud platform receives the data frame from the vehicle networking module through the mobile network interface;

[0114] The cloud platform unpacks the data frame to obtain the vehicle operation data and the power supply stability index.

[0115] Specifically, the cloud platform is pre-configured with an interface that adapts to vehicle mobile networks. The interface supports 4G, 5G and other mainstream mobile network data reception. The interface communication protocol uses existing vehicle networking general standards, which can be compatible with different brands of vehicle networking module data transmission requirements, ensuring the stability and timeliness of data reception, and avoiding data loss caused by protocol incompatibility.

[0116] After receiving, the cloud platform starts the data unpacking program, extracts the vehicle operation data and the power supply stability index (SI value) from the data frame according to the preset data frame format, and stores them in the corresponding data processing unit: the operation data is stored in the real-time database for fault diagnosis analysis; the power supply stability index is stored in the association database and bound with the operation data of the corresponding time period. The unpacking program is developed based on existing data analysis algorithms, and only the custom data frame format is optimized for analysis logic, which prepares data for subsequent power-aware data fusion, adaptive filtering, dual-mode fault discrimination and other dynamic compensation processing steps, and ensures smooth progress of the entire remote detection process.

[0117] In summary, the application captures the voltage fluctuation waveform of the vehicle power bus, the ripple coefficient of the sensor power supply circuit and the transient operating current response curve of the key electronic components in real time by the power supply state monitoring unit in the vehicle vehicle networking module, establishes a three-dimensional power supply quality evaluation model based on these data to generate a power supply stability index, and synchronously encapsulates the index when transmitting vehicle operation data; after receiving by the cloud platform, the power supply sensing data fusion module realizes data space-time alignment, the adaptive filtering module dynamically denoises, the double-mode fault discrimination module runs the parallel operation benchmark and the elastic fault tree and compares the confidence, and only when the two fault trees report the same fault code, the early warning is triggered, effectively solving the problem that the prior art does not consider the dynamic power fluctuation to cause data deviation and false fault misjudgment, effectively improving the accuracy of automobile remote detection, reducing unnecessary maintenance and troubleshooting workload, enhancing the trust of users on remote detection technology, and guaranteeing the reliability and effectiveness of vehicle fault early warning.

[0118] It should be noted that, in this document, the terms such as first and second are used merely to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply there is any such actual relationship or order between these entities or operations. Moreover, the terms "include", "contain" or any other variants thereof are intended to cover non-exclusive inclusion, so that the process, method, article or equipment including a series of elements not only includes those elements, but also includes other elements not explicitly listed or inherent to such process, method, article or equipment.

[0119] Although the embodiments of the present application have been shown and described, it will be understood by those of ordinary skill in the art that various changes, modifications, substitutions and variations can be made thereto without departing from the principles and spirit of the present application, and the scope of the present application is defined by the appended claims and their equivalents.

Claims

1. A method for remote detection of a vehicle, characterized in that, The application relates to a vehicle power supply state monitoring method based on a vehicle internet module. In a vehicle internet module of a vehicle, a voltage fluctuation waveform of a vehicle-mounted power supply bus, a ripple coefficient of each sensor power supply circuit and a transient working current response curve of a key electronic component are captured in real time by a power supply state monitoring unit, a three-dimensional power supply quality evaluation model is established based on the waveform, the ripple coefficient and the transient response curve, a power supply stability index is dynamically generated, and the power supply stability index is mapped to the potential influence degree of the current vehicle power supply environment on sensor data acquisition in a digital twin manner. The vehicle internet module synchronously encapsulates the power supply stability index when transmitting vehicle operation data. A cloud platform receives the vehicle operation data and the power supply stability index, and performs dynamic compensation processing on the vehicle operation data based on the power supply stability index, which includes: Through a power supply awareness data fusion module, the power supply stability index and the vehicle operation data are spatio-temporally aligned to identify the coupling relationship between power supply fluctuations and sensor data anomalies. Through an adaptive filtering module, a variable parameter digital filter is adopted, the filter cutoff frequency and bandwidth of the variable parameter digital filter are dynamically adjusted according to the power supply stability index, and dynamic denoising processing is performed on the fluctuation-sensitive parameters. Through a dual-mode fault diagnosis module, a baseline fault tree under normal power supply state and an elastic fault tree under dynamic power supply state are established, and the two kinds of diagnosis logics are run in parallel, the confidence difference between the diagnosis results in the two modes is compared, and the pseudo-fault signals caused by power supply fluctuations are eliminated. The warning information is pushed only when the same fault code is reported in both modes.

2. The automobile remote detection method according to claim 1, wherein The power supply state monitoring unit captures the voltage fluctuation waveform of the vehicle-mounted power supply bus, the ripple coefficient of each sensor power supply circuit and the transient working current response curve of the key electronic component in real time, which includes: The multi-channel synchronous sampling technology is adopted to acquire the voltage change data of the vehicle-mounted power supply bus in a preset time period in real time. The periodic ripple voltage amplitude of each sensor power supply circuit is acquired in real time. The current response transient value of the key electronic component under different working loads is acquired in real time.

3. The automobile remote detection method according to claim 2, wherein The three-dimensional power supply quality evaluation model is established based on the waveform, the ripple coefficient and the transient response curve, and the power supply stability index is dynamically generated, and the power supply stability index is mapped to the potential influence degree of the current vehicle power supply environment on sensor data acquisition in a digital twin manner, which includes: The voltage change data, the ripple voltage amplitude and the current response transient value are input into the pre-trained three-dimensional power supply quality evaluation model. The three-dimensional power supply quality evaluation model outputs the dynamic power supply stability index based on the input data, and the index size indicates the influence degree of the power supply environment on the sensor data acquisition accuracy.

4. The automobile remote detection method according to claim 1, wherein The power supply awareness data fusion module spatio-temporally aligns the power supply stability index and the vehicle operation data to identify the coupling relationship between power supply fluctuations and sensor data anomalies, which includes: The time stamp of the vehicle operation data acquisition and the time stamp of the power supply stability index generation are acquired. The power supply stability index and the vehicle operation data are aligned according to the time stamp. The aligned data are analyzed to determine whether the change of the power supply stability index is related to the abnormal fluctuation in the vehicle operation data.

5. The automobile remote detection method according to claim 1, wherein The adaptive filtering module adopts a variable parameter digital filter to dynamically adjust the filter cutoff frequency and bandwidth of the variable parameter digital filter according to the power supply stability index, and performs dynamic denoising processing on the fluctuation sensitive parameters, including: According to the high and low of the power supply stability index, the filter cutoff frequency and the filter bandwidth of the variable parameter digital filter are calculated and updated in real time; The fluctuation sensitive parameters are input into the variable parameter digital filter for processing to realize intelligent wave trapping in the power supply noise frequency band and eliminate the noise components introduced by power supply fluctuation.

6. The automobile remote detection method according to claim 1, wherein The dual-mode fault discrimination module establishes a reference fault tree under normal power supply state and an elastic fault tree under dynamic power supply state, and runs the two diagnostic logics in parallel, compares the confidence difference of the diagnostic results under the two modes, and eliminates the false fault signals caused by power supply fluctuation, including: Based on the vehicle historical operation data and the standard fault mode, a reference fault tree under normal power supply state is constructed for diagnosing faults under stable power supply conditions; Based on the power supply fluctuation simulation data and the dynamic calibration parameters, an elastic fault tree under dynamic power supply state is constructed for diagnosing faults under fluctuating power supply conditions; The denoised vehicle operation data is input into the reference fault tree and the elastic fault tree for diagnosis to obtain their respective diagnostic results and confidence.

7. The automotive remote detection method of claim 6, wherein, The confidence difference of the diagnostic results under the two modes is compared to eliminate the false fault signals caused by power supply fluctuation; only when the same fault code is reported by the two modes, the early warning information is triggered to be pushed, including: If the diagnostic results of the reference fault tree and the elastic fault tree are inconsistent, or the diagnostic results of the two modes are consistent but the confidence difference exceeds the preset threshold, the diagnostic result is identified as a false fault signal and is eliminated; If the reference fault tree and the elastic fault tree both diagnose the same fault code, and the confidence difference of the diagnostic results of the two modes is within the preset threshold range, it is identified as a real fault, and the early warning information is triggered to be pushed.

8. The automotive remote detection method of claim 1, wherein, The vehicle operation data includes engine speed, oil pressure, transmission temperature and brake system state parameters; the fluctuation sensitive parameters include engine speed data and oil pressure data.

9. The automobile remote detection method according to claim 1, wherein The vehicle networking module synchronously encapsulates the power supply stability index when transmitting the vehicle operation data, including: Pack the vehicle operation data and the power supply stability index into a unified data frame format; Transmit the data frame to the cloud platform through the mobile network.

10. The automotive remote detection method of claim 1, wherein, The cloud platform receives the vehicle operation data and the power supply stability index, including: The cloud platform receives the data frame from the vehicle networking module through the mobile network interface; The cloud platform unpacks the data frame to obtain the vehicle operation data and the power supply stability index.

Images