Vehicle diagnostic methods, devices, electronic equipment and storage media

Abstract

This application relates to the field of vehicle diagnostic technology, disclosing a vehicle diagnostic method, device, electronic device, and storage medium. The method includes: weighting and summing multiple historical diagnostic records according to time decay weights, such that historical anomalies closer to the present contribute more, thereby accurately reflecting the recent evolution trend of vehicle faults; then, converting the current diagnostic information and historical comprehensive information into numerical sets of anomaly dimensions respectively, and then fusing and analyzing the two sets to obtain a comprehensive diagnostic result. The beneficial effects of this invention are: improving the fit of the comprehensive diagnostic result to the actual vehicle fault state, reducing misdiagnosis or missed diagnosis caused by ignoring historical evolution, providing a structured data foundation for subsequent automated maintenance suggestion generation, and improving the accuracy and efficiency of vehicle diagnosis.

Description

Technical Field

[0001] This invention relates to the field of vehicle diagnostic technology, and in particular to a vehicle diagnostic method, apparatus, electronic device, and storage medium. Background Technology

[0002] In the field of vehicle diagnostic technology, traditional diagnostic equipment generates a diagnostic report and provides repair suggestions based solely on the fault codes and data streams detected each time a vehicle is scanned for faults. This single-diagnosis mode ignores the evolution of faults during long-term vehicle use, resulting in a lack of continuity in diagnostic results. Summary of the Invention

[0003] Therefore, it is necessary to propose a vehicle diagnostic method, device, electronic equipment, and storage medium to address existing vehicle diagnostic problems.

[0004] A vehicle diagnostic method, the method comprising: Establish a bus connection with the vehicle to be inspected in order to obtain the vehicle identification number of the vehicle to be inspected; Perform current vehicle diagnostics based on the vehicle identification code and generate current diagnostic information; Based on the vehicle identification code, retrieve multiple historical diagnostic records associated with the vehicle identification code from a preset server; Identify the first abnormal information in the current diagnostic information and convert it into the first dimension value of the corresponding abnormal dimension to obtain the first abnormal set; The system identifies the second abnormal information in each historical diagnostic record and converts it into the corresponding second dimension value of the abnormal dimension. The values ​​are then weighted and summed according to preset weights to obtain the comprehensive dimension value of each abnormal dimension, thus obtaining the second abnormal set. The weight of each second dimension value is related to the historical time; the closer to the current time, the higher the weight value. The first anomaly set and the second anomaly set are fused according to a preset fusion method to obtain a comprehensive anomaly set, and the comprehensive anomaly set is analyzed to obtain a comprehensive diagnostic result.

[0005] Further, the step of identifying the first abnormal information in the current diagnostic information and converting it into a first dimension value corresponding to the abnormal dimension to obtain the first abnormal set includes: Extract at least one fault code or abnormal data stream parameter from the current diagnostic information; Each extracted fault code or abnormal data stream parameter is mapped to one dimension of a pre-defined multi-dimensional anomaly space; where each dimension corresponds to a fault type or component. Based on the severity level of the fault code or the degree to which the abnormal data stream parameters deviate from the standard threshold, an initial value is assigned to the mapped dimension to obtain the values ​​of each first dimension in the first abnormal set.

[0006] Further, the step of identifying the second abnormal information in each historical diagnostic record, converting it into the second dimension value of the corresponding abnormal dimension, and performing a weighted summation according to a preset weight to obtain the comprehensive dimension value of each abnormal dimension, so as to obtain the second abnormal set, includes: For each historical diagnostic record, a time weighting coefficient is assigned to it; wherein, the time weighting coefficient is calculated using a time decay function; For each abnormal dimension, the second dimension values ​​belonging to that abnormal dimension in all historical diagnostic records are multiplied by the time weight coefficient corresponding to each historical diagnostic record, summed, and then divided by the sum of all time weight coefficients involved in the summation to obtain the comprehensive dimension value of that abnormal dimension, thus obtaining the second abnormal set.

[0007] Further, the step of fusing the first anomaly set and the second anomaly set according to a preset fusion method to obtain a comprehensive anomaly set includes: Iterate through the first exception set and the second exception set; If an abnormal dimension exists in both the first abnormal set and the second abnormal set, it is recorded as the same abnormal dimension. For any identical abnormal dimension, the dimension values ​​in the first abnormal set and the combined dimension values ​​in the second abnormal set are weighted and summed to obtain the fused dimension value; where the weight corresponding to the current diagnostic information is the first weight, and the weight corresponding to the historical diagnostic information is the second weight. The fused dimension values ​​are added to the comprehensive anomaly set as part of the comprehensive anomaly set.

[0008] Further, the step of fusing the first anomaly set and the second anomaly set according to a preset fusion method to obtain a comprehensive anomaly set includes: Iterate through the first exception set and the second exception set; If an abnormal dimension exists only in the first abnormal set but not in the second abnormal set, it is denoted as the current abnormal dimension. For any current abnormal dimension, directly add the current abnormal dimension and its corresponding dimension value in the first abnormal set as a new abnormal item to the comprehensive abnormal set; If an abnormal dimension exists only in the second abnormal set and not in the first abnormal set, it is denoted as a historical abnormal dimension. The comprehensive dimension value of the historical anomaly dimension in the second anomaly set is calculated and compared with a preset historical legacy anomaly importance threshold; If the historical anomaly importance threshold is greater than or equal to the historical anomaly importance threshold, it is added to the comprehensive anomaly set as a historical continuation anomaly and marked. If the historical anomaly importance threshold is less than the historical anomaly importance threshold, the historical anomaly dimension is discarded.

[0009] Further, after the step of fusing the first anomaly set and the second anomaly set according to a preset fusion method to obtain a comprehensive anomaly set, and analyzing the comprehensive anomaly set to obtain a comprehensive diagnostic result, the following steps are included: Obtain the target maintenance recommendations taken by the user based on the comprehensive diagnostic results, as well as the corresponding anomaly dimensions; Using the target repair suggestion action as a positive sample and other repair suggestion actions rejected or skipped by the user as negative samples, the preset weights and / or the preset fusion method are adjusted in reverse.

[0010] Further, the step of retrieving multiple historical diagnostic records associated with the vehicle identification code from a preset server based on the vehicle identification code includes: Based on the vehicle identification code, retrieve all historical diagnostic records associated with the vehicle identification code and their timestamps from the preset server; Get the preset maximum number of records or the preset time backtracking range; Based on the preset maximum number of records or the preset time backtracking range, select the historical diagnostic records from all historical diagnostic records that are within the maximum number of records or the preset time backtracking range, and use them as the multiple historical diagnostic records.

[0011] A vehicle diagnostic device, the device comprising: The module is used to establish a bus connection with the vehicle to be inspected in order to obtain the vehicle identification number of the vehicle to be inspected. The generation module is used to perform current vehicle diagnosis based on the vehicle identification code and generate current diagnostic information; The retrieval module is used to retrieve multiple historical diagnostic records associated with the vehicle identification code from a preset server based on the vehicle identification code. The identification module is used to identify the first abnormal information in the current diagnostic information and convert it into the first dimension value of the corresponding abnormal dimension to obtain the first abnormal set; The transformation module is used to identify the second abnormal information in each historical diagnostic record and transform it into the second dimension value of the corresponding abnormal dimension. The value is then weighted and summed according to preset weights to obtain the comprehensive dimension value of each abnormal dimension, thus obtaining the second abnormal set. The weight of each second dimension value is related to the historical time; the closer to the current time, the higher the weight value. The fusion module is used to fuse the first anomaly set and the second anomaly set according to a preset fusion method to obtain a comprehensive anomaly set, and to analyze the comprehensive anomaly set to obtain a comprehensive diagnostic result.

[0012] An electronic device includes a memory and a processor, the memory storing a computer program that, when executed by the processor, causes the processor to perform the following steps: Establish a bus connection with the vehicle to be inspected in order to obtain the vehicle identification number of the vehicle to be inspected; Perform current vehicle diagnostics based on the vehicle identification code and generate current diagnostic information; Based on the vehicle identification code, retrieve multiple historical diagnostic records associated with the vehicle identification code from a preset server; Identify the first abnormal information in the current diagnostic information and convert it into the first dimension value of the corresponding abnormal dimension to obtain the first abnormal set; The system identifies the second abnormal information in each historical diagnostic record and converts it into the corresponding second dimension value of the abnormal dimension. The values ​​are then weighted and summed according to preset weights to obtain the comprehensive dimension value of each abnormal dimension, thus obtaining the second abnormal set. The weight of each second dimension value is related to the historical time; the closer to the current time, the higher the weight value. The first anomaly set and the second anomaly set are fused according to a preset fusion method to obtain a comprehensive anomaly set, and the comprehensive anomaly set is analyzed to obtain a comprehensive diagnostic result.

[0013] A computer-readable storage medium storing a computer program, which, when executed by a processor, causes the processor to perform the following steps: Establish a bus connection with the vehicle to be inspected in order to obtain the vehicle identification number of the vehicle to be inspected; Perform current vehicle diagnostics based on the vehicle identification code and generate current diagnostic information; Based on the vehicle identification code, retrieve multiple historical diagnostic records associated with the vehicle identification code from a preset server; Identify the first abnormal information in the current diagnostic information and convert it into the first dimension value of the corresponding abnormal dimension to obtain the first abnormal set; The system identifies the second abnormal information in each historical diagnostic record and converts it into the corresponding second dimension value of the abnormal dimension. The values ​​are then weighted and summed according to preset weights to obtain the comprehensive dimension value of each abnormal dimension, thus obtaining the second abnormal set. The weight of each second dimension value is related to the historical time; the closer to the current time, the higher the weight value. The first anomaly set and the second anomaly set are fused according to a preset fusion method to obtain a comprehensive anomaly set, and the comprehensive anomaly set is analyzed to obtain a comprehensive diagnostic result.

[0014] The beneficial effects of this invention are as follows: By weighting and summing multiple historical diagnostic records according to time decay weights, the contribution of historical anomalies closer to the present is increased, thus accurately reflecting the recent evolution trend of vehicle faults. Then, by converting the current diagnostic information and historical comprehensive information into numerical sets of anomaly dimensions respectively, and then fusing and analyzing the two sets, a comprehensive diagnostic result is obtained. This improves the fit of the comprehensive diagnostic result to the actual fault state of the vehicle, reduces misdiagnosis or missed diagnosis caused by ignoring historical evolution, provides a structured data foundation for subsequent automated maintenance suggestion generation, and improves the accuracy and efficiency of vehicle diagnosis. Attached Figure Description

[0015] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0016] in: Figure 1 This is a diagram illustrating the application environment of a vehicle diagnostic method in one embodiment; Figure 2 Here is a flowchart of a vehicle diagnostic method in one embodiment; Figure 3 This is a structural block diagram of a vehicle diagnostic device in one embodiment; Figure 4 This is a structural block diagram of an electronic device in one embodiment. Detailed Implementation

[0017] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0018] Figure 1 This is a diagram illustrating a vehicle diagnostic application environment in one embodiment. (Refer to...) Figure 1 This vehicle diagnostic method is applied to a vehicle diagnostic system. The vehicle diagnostic system includes a terminal 110 and a server 120. The terminal 110 and server 120 are connected via a network. The terminal 110 can be a desktop terminal or a mobile terminal; a mobile terminal can be at least one of a mobile phone, tablet, or laptop. The server 120 can be a standalone server or a server cluster consisting of multiple servers. The terminal 110 is used to establish a bus connection with the vehicle under test, and the server 120 is used to generate comprehensive diagnostic results.

[0019] like Figure 2 As shown, in one embodiment, a vehicle diagnostic method is provided. This method can be applied to either a terminal or a server; this embodiment uses a terminal as an example. The vehicle diagnostic method specifically includes the following steps: S1: Establish a bus connection with the vehicle to be inspected in order to obtain the vehicle identification number of the vehicle to be inspected; S2: Perform current vehicle diagnostics based on the vehicle identification code and generate current diagnostic information; S3: Based on the vehicle identification code, retrieve multiple historical diagnostic records associated with the vehicle identification code from a preset server; S4: Identify the first abnormal information in the current diagnostic information and convert it into the first dimension value of the corresponding abnormal dimension to obtain the first abnormal set; S5: Identify the second abnormal information in each historical diagnostic record and convert it into the second dimension value of the corresponding abnormal dimension. Then, perform weighted summation according to preset weights to obtain the comprehensive dimension value of each abnormal dimension, so as to obtain the second abnormal set. Among them, the weight of each second dimension value is related to the historical time. The closer to the current time, the higher the weight value. S6: The first abnormal set and the second abnormal set are fused according to a preset fusion method to obtain a comprehensive abnormal set, and the comprehensive abnormal set is analyzed to obtain a comprehensive diagnostic result.

[0020] As described in step S1 above, a bus connection is established with the vehicle under test to obtain its vehicle identification number (VIN). "Bus connection" refers to a communication protocol interface conforming to automotive industry standards, such as CAN (Controller Area Network), LIN (Local Interconnect Network), or FlexRay (FlexRay communication protocol, FlexRay bus). The diagnostic terminal physically connects through the vehicle's OBD-II (On-Board Diagnostics II) interface and sends a data frame request conforming to ISO 15765-2 (International Organization for Standardization 15765-2 standard). Upon receiving the request, the vehicle's ECU (Electronic Control Unit) reads the VIN (Vehicle Identification Number) information from a specific storage address and responds. For example, a handheld diagnostic tool is plugged into the interface under the driver's cab of a vehicle via an OBD-II adapter cable. After the CAN controller driver chip inside the diagnostic tool is initialized, it sends a diagnostic request message "020902". The vehicle gateway replies with a response message containing a 17-digit alphanumeric VIN code (such as "LRW3E7FS9RC123456"). The diagnostic tool successfully parses and stores the VIN.

[0021] As described in step S2 above, current vehicle diagnostics are performed based on the vehicle identification number (VIN) to generate current diagnostic information. After successfully obtaining the VIN, this unique identifier is used to trigger a comprehensive detection of the current vehicle's real-time status and generate current diagnostic information describing its health status. The terminal or cloud server pre-stores diagnostic mapping tables corresponding to different VIN segments. When a specific VIN is input, information such as the vehicle's manufacturer, brand, model year, and check digit is parsed, and then the corresponding diagnostic function list is downloaded or activated. The terminal polls and sends service requests to each ECU of the vehicle, collects response data, and generates a structured report, which is the "current diagnostic information." This report includes at least the currently existing DTCs (Diagnostic Trouble Codes), real-time data streams from each sensor, controller software version number, and system status flags. For example, for the Tesla Model 3 with the aforementioned VIN "LRW3E7FS9RC123456", the diagnostic tool identifies the vehicle as a 2024 Long Range version. It then automatically sends a request to the BMS (Battery Management System) to read the individual cell voltage difference values, and simultaneously reads the current DTC (e.g., "P0AA6: Hybrid / Electric Vehicle Battery Voltage Isolation Fault") in the Engine Controller (ECM). This real-time data, along with the timestamp and mileage, is packaged into a JSON (JavaScript Object Notation) or XML (Extensible Markup Language) file to generate the current diagnostic information.

[0022] As described in step S3 above, based on the vehicle identification number (VIN), multiple historical diagnostic records associated with the VIN are retrieved from a preset server. The diagnostic perspective expands from a single current moment to a time dimension. Using the VIN as the database primary key, all or key past diagnostic records of the vehicle are retrieved from the cloud server, thereby introducing historical status information. After obtaining the VIN, the diagnostic terminal sends a request containing the VIN and the requested time range to the server API (Application Programming Interface) via HTTPS (Hypertext Transfer Protocol Secure). The server executes an SQL (Structured Query Language) query in its diagnostic record database to filter out all historical diagnostic events corresponding to the VIN. Each event may contain information such as historical diagnostic time, historical fault code list, historical data snapshot, and maintenance action record. The result set returned by the server is arranged in descending order of time, forming multiple historical diagnostic records.

[0023] As described in step S4 above, the first abnormal information in the current diagnostic information is identified and converted into the first dimension value of the corresponding abnormal dimension to obtain the first abnormal set. Abnormal information refers to all fields or fault codes in the current diagnostic information that exceed a preset normal threshold or do not match the standard state. An abnormal dimension refers to a predefined vehicle fault category or performance degradation category, such as "voltage abnormal dimension," "temperature abnormal dimension," "communication loss dimension," "emission exceeding standard dimension," etc. The conversion process uses a mapping function or scoring rule: for each identified abnormal information, a table is looked up or its corresponding quantitative score for a certain dimension is calculated. The final "first abnormal set" is a set of key-value pairs containing dimension identifiers and values. For example: assuming the current diagnostic information contains fault code "P0AA6" (low insulation resistance in the high-voltage system) and a data stream "the highest single cell voltage of the battery pack is 4.25V (threshold 4.20V)", these two abnormalities are identified respectively: P0AA6 is mapped to the "insulation abnormal dimension," and converted into a dimension value of 0.85 according to the severity level of the fault code; the voltage exceeding limit is mapped to the "voltage consistency abnormal dimension."

[0024] As described in step S5 above, the second abnormal information in each historical diagnostic record is identified and converted into the second dimension value of the corresponding abnormal dimension. A weighted sum is then performed according to a preset weight to obtain the comprehensive dimension value of each abnormal dimension, thus obtaining the second abnormal set. The weight of each second dimension value is related to historical time; the closer to the current time, the higher the weight value. From each historical diagnostic record, using the same or compatible rules as in step S4, the second abnormal information is identified and converted into the second dimension value of the corresponding abnormal dimension. For each abnormal dimension (e.g., "insulation abnormal dimension"), the values ​​of that dimension in all historical records are collected and multiplied by a weight coefficient determined by "historical time". After performing this process on all traversed dimensions, a "second abnormal set" containing multiple dimensions and their comprehensive values ​​is formed. For example, for the "insulation abnormal dimension," there are three historical records: 3 months ago (weight coefficient set to 0.5) a value of 0.7; 1 month ago (weight 0.8) a value of 0.6; and 1 week ago (weight 0.95) a value of 0.9. The weighted sum is then 0.5*0.7+0.8*0.6+0.95*0.9=0.35+0.48+0.855=1.685. If the vehicle also experienced an insulation anomaly two years ago (weight 0.05), its contribution is negligible, and the final second anomaly set is {"Insulation Anomaly": 1.685, "Voltage Consistency Anomaly": 0.2, ...}.

[0025] As described in step S6 above, the first anomaly set and the second anomaly set are fused according to a preset fusion method to obtain a comprehensive anomaly set. Analysis is then performed based on this comprehensive anomaly set to obtain a comprehensive diagnostic result. The preset fusion method can be various mathematical or logical operations, including but not limited to: weighted average, maximum value selection, Bayesian fusion, fuzzy logic synthesis, etc. For example, if it is believed that the current diagnosis has the highest authority in confirming immediate faults, while historical records are more valuable for predicting recurrence risk, the fusion formula can be set as comprehensive dimension value = max(first dimension value, second dimension value) or comprehensive dimension value = 0.7 * first dimension value + 0.3 * second dimension value. The fused "comprehensive anomaly set" contains the final values ​​of each anomaly dimension. Analysis is performed based on this set: a threshold can be set, triggering specific maintenance suggestions for dimensions exceeding a certain value; or the multi-dimensional values ​​can be input into a pre-trained classifier, outputting the corresponding fault category and confidence level. The final generated "comprehensive diagnostic result" should be an executable suggestion in natural language or structured code form. For example, the first set of anomalies is {insulation anomaly: 0.85}, and the second set of anomalies is {insulation anomaly: 1.685}. If a weighted average is used, the corresponding weight parameters can be iteratively optimized through empirical settings, statistical calibration based on historical data, or through the user feedback mechanism described in claim 6. That is, if the current weight is 0.6 and the historical weight is 0.4, then the comprehensive insulation anomaly value = 0.6 * 0.85 + 0.4 * 1.685 = 0.51 + 0.674 = 1.184. Analyzing this value, it is higher than the preset "attention threshold" (0.5) and "warning threshold" (1.0), so the comprehensive diagnostic result is output: "The insulation system has a continuous and recently aggravated deterioration trend. It is recommended to immediately conduct a special test on the insulation resistance of the high-voltage system, and check the battery pack sealing and high-voltage harness wear. Since this fault has repeatedly occurred in the historical record (the weighted value has reached 1.184), it is not recommended to simply clear the fault code; the insulation monitoring module should be replaced." In one embodiment, step S4, which involves identifying the first abnormal information in the current diagnostic information and converting it into a first dimension value corresponding to the abnormal dimension to obtain the first abnormal set, includes: S401: Extract at least one fault code or abnormal data stream parameter from the current diagnostic information; S402: Map each extracted fault code or abnormal data stream parameter to a dimension in a preset multi-dimensional anomaly space; where each dimension corresponds to a fault type or component; S403: Based on the severity level of the fault code or the degree to which the abnormal data stream parameters deviate from the standard threshold, assign an initial value to the mapped dimension to obtain the values ​​of each first dimension in the first abnormal set.

[0026] As described in step S401 above, at least one DTC (Diagnostic Trouble Code) or abnormal data stream parameter is extracted from the current diagnostic information. From this broad dataset of generated current diagnostic information, core indicators characterizing vehicle health issues are selectively identified: fault codes and abnormal data stream parameters. Fault codes are standardized fault identifiers generated by the vehicle's electronic control unit according to a self-diagnostic protocol; for example, "P0301" indicates a misfire in cylinder 1. Abnormal data stream parameters refer to continuous monitoring values ​​that, while not meeting the conditions for generating a fault code, deviate from the preset normal operating range. The diagnostic terminal or cloud processor parses the current diagnostic information: if it is a list of fault codes, it directly iterates through and extracts each five-character DTC; if it is a data stream snapshot, it compares each item with a standard threshold library for the vehicle model stored locally or in the cloud, marking parameters exceeding or falling below the threshold as "abnormal data stream parameters." For example: Suppose that the current diagnostic information generated in step S2 contains a standard fault code "P0AA6" (high voltage system insulation fault) and a real-time data stream "right front oxygen sensor voltage continuously outputs 0.08V, while the standard range is 0.1V-0.9V". P0AA6 is extracted by the DTC parser. At the same time, through the threshold comparison algorithm, it is found that 0.08V is lower than the minimum threshold of 0.1V. Therefore, "right front oxygen sensor voltage 0.08V" is extracted as an abnormal data stream parameter.

[0027] As described in step S402 above: each extracted fault code or abnormal data stream parameter is mapped to a dimension in a preset multidimensional anomaly space; where each dimension corresponds to a fault type or component. A "semantic mapping" or "classification reduction" is performed to map individual, low-level fault codes or data stream parameters to a specific dimension in a high-level, abstract "multidimensional anomaly space." This multidimensional anomaly space is a pre-designed structured model, where each dimension represents an independent fault type (e.g., "ignition system anomaly," "fuel system anomaly," "emission aftertreatment anomaly") or vehicle component (e.g., "engine," "transmission," "battery management system," "braking system"). The establishment of this mapping relationship depends on the configuration of a domain knowledge base or fault code database: for example, all misfire fault codes starting with "P030" are mapped to the "combustion stability anomaly" dimension; while the "slow oxygen sensor response" data stream anomaly may be mapped to the "air-fuel ratio control anomaly" dimension. The multidimensional anomaly space can be predefined based on vehicle diagnostic fault code (DTC) classification standards (such as SAE J2012) or vehicle system architecture (such as powertrain, chassis, and body systems). A mapping table is established to associate standard DTCs (such as P0XXX, C0XXX) with preset anomaly dimensions (such as 'engine ignition anomaly', 'communication anomaly'). For abnormal data stream parameters, the corresponding component dimension is mapped by determining the sensor or controller module to which it belongs. For example, based on the fault code "P0AA6" extracted in step S401, the preset mapping table is consulted, and it is found that P0AA6 belongs to the "high voltage insulation system" fault type. Therefore, it is mapped to the "insulation resistance anomaly dimension" in the multidimensional anomaly space. The extracted abnormal data stream parameter "right front oxygen sensor voltage continuously 0.08V" is determined by the rule engine to be in a low voltage state, which usually indicates that the air-fuel mixture is too lean or the sensor itself is malfunctioning. Therefore, it is mapped to the "oxygen sensor signal anomaly dimension" or the broader "emission control system anomaly dimension". If the multidimensional anomaly space has defined 20 dimensions such as "insulation resistance anomaly," "oxygen sensor anomaly," and "engine misfire anomaly," then the two items mentioned above will be assigned to their respective dimensions. If multiple anomaly items are received under a dimension (for example, three different fault codes all pointing to "ignition system"), subsequent steps will comprehensively calculate the value of that dimension.

[0028] As described in step S403 above, based on the severity level of the fault code or the degree to which the abnormal data stream parameter deviates from the standard threshold, an initial value is assigned to the mapped dimension, resulting in the values ​​of each first dimension in the first anomaly set. This defines how to calculate a quantized "first dimension value" for each anomaly dimension, making the first anomaly set a structured data object capable of participating in mathematical operations. The assignment is based on two core sources: first, for fault codes, quantification is performed according to their preset "severity level"; second, for abnormal data stream parameters, quantification is performed according to their "degree" of deviation from the standard threshold. The severity level is typically set by combining the fault code category and its internal classification defined by the standard protocol with the priority defined by the manufacturer. It can be converted into a value from 0 to 1 or from 1 to 10, with higher values ​​indicating greater severity. Ultimately, for each anomaly dimension, if only one anomaly entry is mapped to this dimension, the value assigned to that entry is the first dimension value for that dimension; if multiple anomaly entries are mapped to the same dimension, the maximum value, average value, or other preset aggregation methods can be used to form a unique initial value for that dimension. For example, for fault code P0AA6 mapped to the "Insulation Resistance Anomaly Dimension", referring to its severity level table, this fault code is defined as "Severe - High-voltage insulation fault may lead to electric shock risk", with a level value of 0.9 (out of 1). An initial value of 0.9 is assigned to this dimension. For the abnormal data stream parameter "Right Front Oxygen Sensor Voltage 0.08V (Standard Lower Limit 0.1V)" mapped to the "Oxygen Sensor Signal Anomaly Dimension", the deviation is calculated as follows: the standard lower limit is 0.1V, the tolerance lower limit is 0.05V (exceeding 0.05V is considered complete failure), and the measured value is 0.08V. The deviation from the lower limit is calculated as (0.1-0.08) / (0.1-0.05) = 0.02 / 0.05 = 0.4. Considering that the low voltage duration exceeds the threshold, it can be further corrected to 0.45. An initial value of 0.45 can be assigned to this dimension, so that the first anomaly set contains two entries: {"Insulation Resistance Anomaly Dimension": 0.9, "Oxygen Sensor Signal Anomaly Dimension": 0.45}. It should be noted that if no abnormalities are extracted from the current diagnostic information, the set may be empty, but the steps can still be executed logically.

[0029] In one embodiment, step S5, which involves identifying the second abnormal information in each historical diagnostic record, converting it into a second-dimensional value corresponding to the abnormal dimension, and then performing a weighted summation according to a preset weight to obtain a comprehensive dimension value for each abnormal dimension, thereby obtaining the second abnormal set, includes: S501: For each historical diagnostic record, assign a time weighting coefficient; wherein, the time weighting coefficient is calculated using a time decay function; S502: For each abnormal dimension, multiply the second dimension values ​​belonging to that abnormal dimension in all historical diagnostic records by the time weight coefficient corresponding to each historical diagnostic record, sum them up, and then divide by the sum of all time weight coefficients involved in the summation to obtain the comprehensive dimension value of that abnormal dimension, so as to obtain the second abnormal set.

[0030] As described in step S501 above, a time weighting coefficient is assigned to each historical diagnostic record; wherein, the time weighting coefficient is calculated using a time decay function. Unlike simple averaging or counting, the time decay function can simulate the "forgetting" or "decaying" effect of fault information over time, avoiding excessive interference from past sporadic but repaired faults in current judgment. The time decay function is a mathematical function that takes the time between the historical record time and the current time as the independent variable and outputs a weighting coefficient between 0 and 1 to characterize the timeliness value of historical data. Generally, the closer to the current time, the larger the function output value. First, the timestamp of each historical diagnostic record is obtained, and the time difference Δt is calculated based on the current diagnostic time. Then, the preset time decay function is called. ,in, λ is the time weighting coefficient for the i-th historical diagnostic record; λ is the attenuation coefficient, a preset constant greater than 0, used to control the attenuation rate. The time interval (usually in days or months) between the i-th historical record and the current time; or a piecewise linear decay function can be used. Each historical record is assigned a unique weight coefficient, which is between 0 and 1 (usually not 0, but very old records can approach 0). For example: Suppose the current diagnosis time is May 24, 2026. Three historical diagnosis records are retrieved: Record A occurred on May 10, 2026 (14 days ago), Record B occurred on February 24, 2026 (approximately 90 days ago, 3 months ago), and Record C occurred on May 24, 2025 (1 year ago, 365 days ago). If an exponential decay function is used, the weight of the record from one year ago is only about 0.026, almost negligible; while the record from two weeks ago retains about 87% of the weight.

[0031] As described in step S502 above, for each abnormal dimension, the second dimension values ​​belonging to that abnormal dimension in all historical diagnostic records are multiplied by the time weight coefficient corresponding to each historical diagnostic record, summed, and then divided by the sum of all time weight coefficients involved in the summation to obtain the comprehensive dimension value of that abnormal dimension, thus obtaining the second abnormal set. The time-weighted aggregation calculation of historical data within each abnormal dimension is essentially a "weighted arithmetic average" operation, but the denominator is not simply the number of records, but the sum of all weight coefficients involved in the summation. This normalization process can eliminate the bias caused by the uneven number of historical records, ensuring that the final comprehensive dimension value remains within a reasonable range of dimensions. First, the historical diagnostic records with assigned time weight coefficients in step S501 are traversed. For the currently processed abnormal dimension, the second dimension value of that dimension is extracted from each historical record. Then, the weighted sum is calculated, and the weighted sum is also calculated. Finally, the comprehensive dimension value = weighted sum / weighted sum. If a certain anomalous dimension has never appeared in any historical records (i.e., all second dimension values ​​are 0), the denominator may be 0. In this case, the comprehensive dimension value can be agreed to be 0 or the default minimum value can be used. Repeat the above calculation for all predefined anomalous dimensions to obtain the second anomaly set. For example: following the weight example in step S501 (record A weight 0.869, record B weight 0.407, record C weight 0.026). Now calculate the comprehensive dimension value of the "insulation anomaly dimension". Assume that the second dimension values ​​of this dimension in the three historical records are as follows: record A is 0.9 (recent insulation anomaly is severe), record B is 0.2 (slight anomaly three months ago), and record C is 0.8 (more severe one year ago but has been repaired). The weighted sum is 0.9 × 0.869 + 0.2 × 0.407 + 0.8 × 0.026 = 0.7821 + 0.0814 + 0.0208 = 0.8843, and the weighted sum is 0.869 + 0.407 + 0.026 = 1.302. Therefore, the overall dimension value is approximately 0.8843 / 1.302 ≈ 0.679. Next, calculate the "oxygen sensor anomaly dimension." Assuming that only record B has this dimension value of 0.6 in the history, while records A and C have a value of 0 (or do not exist), the weighted sum is 0.6 × 0.407 = 0.2442, the weighted sum is 0.407, and the overall value is 0.2442 / 0.407 = 0.6. For the "transmission anomaly dimension," which has never appeared, the weighted sum is 0, and the overall value can be set to 0. The final second anomaly set will look like this: {"Insulation Anomaly": 0.679, "Oxygen Sensor Anomaly": 0.6, "Transmission Anomaly": 0}. Note that because the denominator uses a weighted sum instead of a simple count of records, this result will differ from a direct weighted average (which uses the weighted sum of all records, but includes records with a value of 0 in the denominator).

[0032] In one embodiment, step S6, which involves fusing the first anomaly set and the second anomaly set according to a preset fusion method to obtain a comprehensive anomaly set, includes: S601: Traverse the first exception set and the second exception set; S602: If an abnormal dimension exists in both the first abnormal set and the second abnormal set, it is recorded as the same abnormal dimension; S603: For any identical abnormal dimension, the dimension value in the first abnormal set and the comprehensive dimension value in the second abnormal set are weighted and summed to obtain the fused dimension value; where the weight corresponding to the current diagnostic information is the first weight, and the weight corresponding to the historical diagnostic information is the second weight. S604: Add the fused dimension values ​​to the comprehensive anomaly set as part of the comprehensive anomaly set.

[0033] As described in step S601 above, the first anomaly set and the second anomaly set are traversed. The diagnostic system can first obtain a list of all dimensions in the first anomaly set, and simultaneously obtain a list of all dimensions in the second anomaly set, and then traverse them separately; or it can use a joint traversal method to merge the dimension names in the two sets into a deduplicated dimension candidate set, and then process them one by one. For example: Suppose the first anomaly set contains {"insulation anomaly": 0.9, "oxygen sensor anomaly": 0.45}, and the second anomaly set contains {"insulation anomaly": 0.679, "transmission anomaly": 0}, step S601 will start a traverser, first pointing to the first dimension "insulation anomaly" in the first anomaly set, and then pointing to the second dimension "oxygen sensor anomaly"; then it traverses the first dimension "insulation anomaly" in the second anomaly set, and then points to "transmission anomaly". Through this systematic scanning, it can be identified that "insulation anomaly" is a dimension that exists in both sets, while "oxygen sensor anomaly" exists only in the first set, and "transmission anomaly" exists only in the second set.

[0034] As described in step S602 above, if an abnormal dimension exists in both the first abnormal set and the second abnormal set, it is recorded as the same abnormal dimension. Abnormal dimensions that appear both in the current diagnostic information and in the weighted aggregation results of historical diagnostic records are identified and marked as "same abnormal dimensions". For example, continuing the example from step S601, the first abnormal set has "insulation abnormality" and "oxygen sensor abnormality"; the second abnormal set has "insulation abnormality" and "transmission abnormality". By comparing the dimension names, the system finds that "insulation abnormality" appears in both sets, so it is marked as "same abnormal dimension", while "oxygen sensor abnormality" appears only in the first set and "transmission abnormality" appears only in the second set; these two are not marked as "same abnormal dimensions". One noteworthy boundary case is that if a dimension exists in both sets, but the dimension value in one set is 0 or null, it should still be considered "existing" according to the designer's intent, because the dimension itself is defined, and a value of 0 does not mean that the dimension does not exist. The output of this step is a list of dimension names (e.g., ["Insulation Anomaly"]) to guide the specific calculations in step S603.

[0035] As described in step S603 above, for any identical abnormal dimension, the dimension values ​​in the first abnormal set and the comprehensive dimension values ​​in the second abnormal set are weighted and summed to obtain the fused dimension value; where the weight corresponding to the current diagnostic information is the first weight, and the weight corresponding to the historical diagnostic information is the second weight. The weighted summation formula can be expressed as: Fusion value = w_current × V_current + w_history × V_history, where w_current is the first weight, w_history is the second weight, V_current is the dimension value in the first abnormal set, and V_history is the comprehensive dimension value in the second abnormal set. These two weights are preset hyperparameters and can be adjusted according to the actual diagnostic scenario: for example, if the immediacy of the current fault is more important, w_current can be set to 0.7, w_history to 0.3; if the historical recurrence trend is more important, they can be set to 0.4 and 0.6. The sum of the weights is 1 (it can also be other than 1, but it is usually 1 to maintain consistent numerical dimensions). Specifically, the first dimension value of this dimension is read from the first anomaly set, and the comprehensive dimension value of this dimension is read from the second anomaly set. Then, a weighted summation function is called to calculate the result. For example, assuming the same anomaly dimension is "insulation anomaly", its value in the first anomaly set is 0.9, and its comprehensive dimension value in the second anomaly set is 0.679 (from step S502). The preset first weight (current diagnostic information weight) is 0.6, and the second weight (historical diagnostic information weight) is 0.4. Then, the fused dimension value = 0.6 × 0.9 + 0.4 × 0.679 = 0.54 + 0.2716 = 0.8116. If the weight configuration is different, for example, the current weight is 0.3 and the historical weight is 0.7, then the fused value = 0.3 × 0.9 + 0.7 × 0.679 = 0.27 + 0.4753 = 0.7453. It can be seen that by adjusting the weights, the value can be biased towards immediate faults or historical trends.

[0036] As described in step S604 above, the fused dimension values ​​are added to the comprehensive anomaly set as part of it. The fused dimension values, obtained through weighted summation, are stored in the comprehensive anomaly set as key-value pairs, forming a complete structured data object available for subsequent analysis. The comprehensive anomaly set is a newly created set that may contain three types of entries: first, "identical anomaly dimensions" processed in this step and their fused values; second, dimensions existing only in the first anomaly set, whose first dimension values ​​can be directly used; and third, dimensions existing only in the second anomaly set, whose comprehensive dimension values ​​can be directly used. An empty dictionary object can be created (e.g., comprehensive_set={}), and then for each identical anomaly dimension, comprehensive_set[dimension]=fused_value is executed. For example, if the fused value calculated in step S603 is 0.8116, the operation is: comprehensive_set["insulation anomaly"]=0.8116. For dimensions other than those of the same anomaly dimension, they can also be added to the comprehensive anomaly set to maintain integrity. For example, "oxygen sensor anomaly" (0.45), which exists only in the first set, can be directly added to the comprehensive anomaly set; "transmission anomaly" (0), which exists only in the second set, can also be directly added. The final comprehensive anomaly set may look like this: {"insulation anomaly": 0.8116, "oxygen sensor anomaly": 0.45, "transmission anomaly": 0}. This set integrates historical and current comprehensive assessments of insulation anomalies while retaining anomaly information from other single sources, providing a complete diagnostic feature vector with time-weighted significance for subsequent analysis steps.

[0037] In one embodiment, step S6, which involves fusing the first anomaly set and the second anomaly set according to a preset fusion method to obtain a comprehensive anomaly set, includes: S611: Traverse the first exception set and the second exception set; S612: If an abnormal dimension exists only in the first abnormal set and not in the second abnormal set, it is recorded as the current abnormal dimension; S613: For any current abnormal dimension, directly add the current abnormal dimension and its corresponding dimension value in the first abnormal set as a new abnormal item to the comprehensive abnormal set. S614: If an abnormal dimension exists only in the second abnormal set and not in the first abnormal set, it is denoted as a historical abnormal dimension; S615: Calculate the comprehensive dimension value of the historical anomaly dimension in the second anomaly set and compare it with the preset historical legacy anomaly importance threshold; S616: If the historical anomaly importance threshold is greater than or equal to the historical anomaly importance threshold, then it is added to the comprehensive anomaly set as a historical continuation anomaly and marked; if it is less than the historical anomaly importance threshold, then the historical anomaly dimension is discarded.

[0038] As described in step S611 above, the first anomaly set and the second anomaly set are traversed. All entries in the first and second anomaly sets are systematically accessed to provide a complete list of dimensions for subsequent differentiation of "current anomaly dimensions," "historical anomaly dimensions," and "same anomaly dimensions." From a computer program implementation perspective, traversal can be accomplished using iterators or loop statements. For example: Suppose the first anomaly set contains {"insulation anomaly": 0.9, "oxygen sensor anomaly": 0.45}, and the second anomaly set contains {"insulation anomaly": 0.679, "transmission anomaly": 0}. Step S611 initiates the traversal process: First, the first set is traversed to obtain the two dimensions "insulation anomaly" and "oxygen sensor anomaly"; then, the second set is traversed to obtain the two dimensions "insulation anomaly" and "transmission anomaly," internally maintaining a mapping table of the frequency of each dimension (e.g., {"insulation anomaly": 2, "oxygen sensor anomaly": 1, "transmission anomaly": 1}). In this way, it is possible to clearly identify which dimension appears in both sets (count is 2), which dimension appears only in the first set (count is 1 and originates from the first set), and which dimension appears only in the second set (count is 1 and originates from the second set).

[0039] As described in step S612 above, if an abnormal dimension exists only in the first abnormal set and not in the second abnormal set, it is recorded as the current abnormal dimension. From all abnormal dimensions, those "abnormal dimensions reflected only in the current diagnostic information" are selected. These dimensions are characterized by: the vehicle exhibiting a certain abnormality at the current moment (e.g., oxygen sensor voltage abnormality), but this abnormality has never appeared in historical diagnostic records, or it appeared historically but its value is 0 after weighted aggregation. These dimensions are marked as "current abnormal dimensions," meaning they represent new, sudden, or recently occurring fault symptoms, and do not have historical continuity. For example: continuing the previous example, the traversed occurrence mapping is {"insulation abnormality": 2, "oxygen sensor abnormality": 1, "transmission abnormality": 1}. The judgment is as follows: for "oxygen sensor abnormality," its count is 1 and it originates from the first set, satisfying the condition of "existing only in the first abnormal set," so it is recorded as the current abnormal dimension. For "insulation abnormality," the count is 2, which does not satisfy the condition; "transmission abnormality," although its count is 1, originates from the second set, and therefore also does not satisfy the condition. Ultimately, the current list of abnormal dimensions is ["Oxygen sensor abnormal"].

[0040] As described in step S613 above, for any current abnormal dimension, the current abnormal dimension and its corresponding dimension value in the first abnormal set are directly added as new abnormal items to the comprehensive abnormal set. This "direct addition" strategy reflects the importance attached to the first appearance of anomalies in the current diagnostic information: even if the anomaly is not supported by historical records, its original severity is fully preserved without any weighting or attenuation, because the current fault may represent a real problem that just occurred in the vehicle and should not be ignored or weakened due to the lack of history. From a patent protection perspective, this step clarifies a branch of the fusion rule: for abnormal dimensions that only appear in the current situation and not in the history, the original value is preserved without any mathematical transformation or threshold filtering.

[0041] As described in step S614 above, if an abnormal dimension exists only in the second abnormal set and not in the first abnormal set, it is recorded as a historical abnormal dimension. Filter out those abnormal dimensions that "exist only in historical diagnostic records but do not appear in the current diagnostic information." A typical characteristic of this type of dimension is that the vehicle experienced a certain abnormality in the past, but this abnormality has temporarily disappeared during the current diagnosis.

[0042] As described in step S615 above, the comprehensive dimension value of the historical anomaly dimension in the second anomaly set is calculated and compared with a preset historical anomaly importance threshold. A quantitative threshold, the "historical anomaly importance threshold," is introduced to distinguish which historical anomaly dimensions are worth retaining in the comprehensive diagnostic results and which can be ignored. Since historical anomaly dimensions do not represent the current instantaneous state, if their comprehensive dimension value is too small, it indicates that the anomaly either occurred very infrequently in the past or was ancient and minor, thus having limited reference value for the current diagnosis. Adding it to the comprehensive anomaly set may introduce noise and interfere with the diagnostic personnel's judgment of the truly urgent issues. This threshold is a preset empirical parameter that can be set through experimental calibration, machine learning, or expert rules, for example, a threshold of 0.3, 0.5, or 0.6. The comparison operation is a simple numerical comparison: if the comprehensive dimension value of the historical anomaly dimension is greater than or equal to the threshold, it is determined to be important and should be retained; otherwise, it is determined to be unimportant and should be discarded.

[0043] As described in step S616 above, if the value is greater than or equal to the historical anomaly importance threshold, it is added to the comprehensive anomaly set as a historical continuation anomaly and marked. If the value is less than the historical anomaly importance threshold, the historical anomaly dimension is discarded. When the comprehensive dimension value of a historical anomaly dimension reaches or exceeds a preset threshold, it indicates that the historical anomaly is statistically significant enough and should be included in the comprehensive anomaly set to alert diagnostic personnel to this potential recurrence. If the value is lower than the threshold, it is directly discarded, i.e., not added to the comprehensive anomaly set, thereby avoiding interference from low-value noise information in the comprehensive diagnostic results. For example, the preset threshold is 0.4. For the historical anomaly dimension "transmission anomaly," the comprehensive dimension value is 0, which is less than 0.4, so this dimension is discarded and not added to the comprehensive anomaly set. For another historical anomaly dimension "brake pad wear anomaly," the value is 0.65, which meets the condition, so it is added to the comprehensive anomaly set.

[0044] In one embodiment, after step S6, which involves fusing the first anomaly set and the second anomaly set according to a preset fusion method to obtain a comprehensive anomaly set, and then analyzing the comprehensive anomaly set to obtain a comprehensive diagnostic result, the method further includes: S701: Obtain the target maintenance recommendation actions taken by the user based on the comprehensive diagnostic results and their corresponding abnormal dimensions; S702: Using the target maintenance suggestion action as a positive sample and other maintenance suggestion actions rejected or skipped by the user as negative samples, adjust the preset weights and / or the preset fusion method in reverse.

[0045] As described in step S701 above, the target repair suggestion action taken by the user based on the comprehensive diagnostic results and its corresponding anomaly dimension are obtained. Repair operations actually performed by the actual repair personnel or vehicle owner based on the comprehensive diagnostic results are collected and associated with their corresponding anomaly dimension. This association provides "real and valid" positive sample data for repairs, offering a supervisory signal for subsequent adjustments to weighting or fusion methods. The "target repair suggestion action" refers to the one the user actually selects and executes from among the multiple repair suggestions provided, such as "replacing the oxygen sensor," "cleaning the throttle body," or "updating the BMS firmware." The "corresponding anomaly dimension" refers to the anomaly dimension that the repair action aims to resolve (e.g., "oxygen sensor anomaly dimension," "throttle body carbon buildup dimension").

[0046] As described in step S702 above, the target repair suggestion action is used as a positive sample, and other repair suggestion actions rejected or skipped by the user are used as negative samples to adjust the preset weights and / or the preset fusion method in reverse. Using the repair actions actually selected by the user and the actions not selected as training signals, a reverse adjustment mechanism is used to correct two key configurable parameters involved in the diagnostic process: one is the "preset weight" used for historical data weighting; the other is the "preset fusion method" used to fuse the current anomaly set and the historical anomaly set. The goal of the adjustment is to enable the system to assign higher diagnostic confidence to the anomaly dimensions corresponding to the repair actions ultimately adopted by the user when facing similar vehicle conditions in the future, thereby improving the accuracy of diagnostic suggestions and user satisfaction. First, the anomaly dimension corresponding to the target repair suggestion action adopted by the user is regarded as the "real fault dimension," and it is assumed that an ideal diagnostic model should assign a higher comprehensive anomaly score to this dimension, thus ranking its repair suggestions higher. The fused dimension value of this dimension in the current diagnostic results is compared with the values ​​of other unadopted dimensions. If the value of the adopted dimension is lower than the values ​​of some unadopted dimensions, it indicates that there is a deviation in the current weights or fusion method, and adjustment is needed. Adjustments can be made incrementally: for example, increasing the first weight w_current (i.e., giving more weight to current diagnostic information) if the current fault code directly leads to oxygen sensor repair; or increasing the time decay rate of the "oxygen sensor anomaly dimension" in the second anomaly set in step S5, if the user selects it, indicating that the fault is time-sensitive. For negative samples, reduce their relevant weight or decrease their fused numerical contribution. For example, the current system configuration is w_current=0.4, w_history=0.6. For a certain diagnosis, the fused value of the "oxygen sensor anomaly" dimension in the comprehensive anomaly set is 0.55, and the fused value of the "insulation anomaly" dimension is 0.82. The system recommends prioritizing the repair of insulation problems (ranked 1st), and the oxygen sensor is ranked 2nd. However, users actually chose to repair the oxygen sensor (positive sample) and abandoned insulation repair (negative sample). This suggests that the importance of oxygen sensor anomalies may have been underestimated or insulation anomalies may have been overestimated. The following adjustments were made: `w_current` was increased by 0.05 (because the current oxygen sensor fault code is clear), and the historical weighting coefficient for the "insulation anomaly" dimension in the second anomaly set was multiplied by 0.9 (to reduce the impact of historical insulation faults). Simultaneously, the importance threshold for historical anomalies was reduced from 0.4 to 0.38 to allow more historical anomalies to be retained. After several iterations, the parameter configuration will gradually converge to better reflect the user's actual repair decisions.

[0047] In one embodiment, step S3, which retrieves multiple historical diagnostic records associated with the vehicle identification number from a preset server based on the vehicle identification number, includes: S301: Based on the vehicle identification code, obtain all historical diagnostic records associated with the vehicle identification code and their timestamps from the preset server; S302: Get the preset maximum number of records or the preset time backtracking range; S303: Based on the preset maximum number of records or the preset time backtracking range, select the historical diagnostic records from all historical diagnostic records that are within the maximum number of records or the preset time backtracking range, and use them as the multiple historical diagnostic records.

[0048] As described in step S301 above, based on the vehicle identification number (VIN), all historical diagnostic records associated with the VIN and their timestamps are retrieved from the preset server. Using the VIN as a unique primary key, all historical diagnostic records generated since the vehicle was put into use are completely retrieved from the cloud-based preset server, along with the timestamp information corresponding to each record. The diagnostic terminal (or server-side module) sends a request to the preset server via HTTPS, including the VIN parameter. The server executes an SQL query in its database, returning all records corresponding to the VIN. Each record contains at least structured fields such as a diagnostic timestamp, a fault code list, and a data stream snapshot.

[0049] As described in step S302 above, obtain the preset maximum number of records or the preset time backtracking range. The "maximum number of records" and "time backtracking range" are preset configurable hyperparameters that can be flexibly set according to the actual application scenario. These two parameters are read from a configuration file, environment variables, or user settings.

[0050] As described in step S303 above, based on the preset maximum number of records or the preset time backtracking range, historical diagnostic records within the maximum number of records or the preset time backtracking range are selected from all historical diagnostic records to serve as the multiple historical diagnostic records. Based on the filtering parameters obtained in step S302, all historical diagnostic records obtained in step S301 are cropped, retaining a subset of records that meet the quantity limit or time window limit, and this subset is used as the "multiple historical diagnostic records" in subsequent steps S5 (weighted summation) and S6 (fusion). First, it is determined which filtering mode to use: if a maximum number of records is preset and the value is valid, all historical records are sorted in descending order by timestamp, and the first N records (N is the maximum number of records) are taken; if no maximum number of records is preset but a time backtracking range is preset, the current time is calculated by subtracting the backtracking range to obtain the cutoff time point, and all records with timestamps greater than or equal to the cutoff point are selected. If neither parameter is preset, all records can be retained by default or a default value can be used (such as retaining the most recent 50 records).

[0051] Reference Figure 3The present invention provides a vehicle diagnostic device, the device comprising: Module 902 is used to establish a bus connection with the vehicle to be inspected in order to obtain the vehicle identification code of the vehicle to be inspected. The generation module 904 is used to perform current vehicle diagnosis based on the vehicle identification code and generate current diagnostic information; The retrieval module 906 is used to retrieve multiple historical diagnostic records associated with the vehicle identification code from a preset server based on the vehicle identification code. The identification module 908 is used to identify the first abnormal information in the current diagnostic information and convert it into the first dimension value of the corresponding abnormal dimension to obtain the first abnormal set. The conversion module 910 is used to identify the second abnormal information in each historical diagnostic record and convert it into the second dimension value of the corresponding abnormal dimension. The values ​​are then weighted and summed according to preset weights to obtain the comprehensive dimension value of each abnormal dimension, thus obtaining the second abnormal set. The weight of each second dimension value is related to the historical time; the closer it is to the current time, the higher the weight value. The fusion module 912 is used to fuse the first anomaly set and the second anomaly set according to a preset fusion method to obtain a comprehensive anomaly set, and to analyze the comprehensive anomaly set to obtain a comprehensive diagnostic result.

[0052] In one embodiment, the identification module 908 includes: An extraction submodule is used to extract at least one fault code or abnormal data stream parameter from the current diagnostic information; The mapping submodule is used to map each extracted fault code or abnormal data stream parameter to a dimension in a preset multi-dimensional anomaly space; where each dimension corresponds to a fault type or component. The assignment submodule is used to assign an initial value to the mapped dimension based on the severity level of the fault code or the degree to which the abnormal data stream parameter deviates from the standard threshold, so as to obtain the value of each first dimension in the first abnormal set.

[0053] In one embodiment, the conversion module 910 includes: The allocation submodule is used to assign a time weight coefficient to each historical diagnostic record; wherein the time weight coefficient is calculated using a time decay function. The summation submodule is used to, for each anomaly dimension, multiply the second dimension values ​​belonging to that anomaly dimension in all historical diagnostic records by the time weight coefficient corresponding to each historical diagnostic record, sum them up, and then divide by the sum of all time weight coefficients involved in the summation to obtain the comprehensive dimension value of that anomaly dimension, so as to obtain the second anomaly set.

[0054] In one embodiment, the fusion module 912 includes: The first traversal submodule is used to traverse the first exception set and the second exception set; The same anomaly dimension marking submodule is used to mark an anomaly dimension as the same if it exists in both the first anomaly set and the second anomaly set. The dimension value acquisition submodule is used to perform a weighted summation of the dimension values ​​in the first set of anomalies and the comprehensive dimension values ​​in the second set of anomalies for any identical anomaly dimension, to obtain the fused dimension value; wherein, the weight corresponding to the current diagnostic information is the first weight, and the weight corresponding to the historical diagnostic information is the second weight. The first addition submodule is used to add the fused dimension values ​​to the comprehensive anomaly set as part of the comprehensive anomaly set.

[0055] In one embodiment, the fusion module 912 includes: The second traversal submodule is used to traverse the first exception set and the second exception set; The current anomaly dimension marking submodule is used to mark an anomaly dimension as the current anomaly dimension if it exists only in the first anomaly set and not in the second anomaly set. The second addition submodule is used to directly add the current abnormal dimension and its corresponding dimension value in the first abnormal set as a new abnormal item to the comprehensive abnormal set for any current abnormal dimension. The historical anomaly dimension marking submodule is used to mark an anomaly dimension as a historical anomaly dimension if it exists only in the second anomaly set and not in the first anomaly set. The calculation submodule is used to calculate the comprehensive dimension value of the historical anomaly dimension in the second anomaly set and compare it with the preset historical legacy anomaly importance threshold. The third addition submodule is used to add a historical anomaly as a historical continuation anomaly to the comprehensive anomaly set and mark it if the anomaly importance threshold is greater than or equal to the historical anomaly importance threshold; otherwise, the historical anomaly dimension is discarded.

[0056] In one embodiment, a vehicle diagnostic device includes: The anomaly dimension acquisition module is used to obtain the target maintenance suggestion actions taken by the user based on the comprehensive diagnostic results and their corresponding anomaly dimensions; The adjustment module is used to adjust the preset weights and / or the preset fusion method in reverse, using the target maintenance suggestion action as a positive sample and other maintenance suggestion actions rejected or skipped by the user as negative samples.

[0057] In one embodiment, the retrieval module 906 includes: The timestamp acquisition submodule is used to obtain all historical diagnostic records associated with the vehicle identification code and their timestamps from the preset server based on the vehicle identification code. The Time Backtracking Range Acquisition Submodule is used to obtain the preset maximum number of records or the preset time backtracking range; The filtering submodule is used to filter historical diagnostic records from all historical diagnostic records according to a preset maximum number of records or a preset time backtracking range, and to use them as the multiple historical diagnostic records.

[0058] Figure 4 An internal structural diagram of an electronic device in one embodiment is shown. This electronic device can specifically be a terminal or a server, and more specifically, a computer device. Figure 4 As shown, the electronic device includes a processor, a memory, and a network interface connected via a system bus. The memory includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores an operating system and may also store a computer program. When executed by the processor, this computer program enables the processor to implement vehicle diagnostic methods. The internal memory may also store a computer program, which, when executed by the processor, enables the processor to perform vehicle diagnostic methods. Those skilled in the art will understand that… Figure 4 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the electronic device to which the present application is applied. The specific electronic device may include more or fewer components than shown in the figure, or combine certain components, or have different component arrangements.

[0059] In one embodiment, an electronic device is provided, including a memory and a processor, the memory storing a computer program that, when executed by the processor, causes the processor to perform the following steps: Establish a bus connection with the vehicle to be inspected in order to obtain the vehicle identification number of the vehicle to be inspected; Perform current vehicle diagnostics based on the vehicle identification code and generate current diagnostic information; Based on the vehicle identification code, retrieve multiple historical diagnostic records associated with the vehicle identification code from a preset server; Identify the first abnormal information in the current diagnostic information and convert it into the first dimension value of the corresponding abnormal dimension to obtain the first abnormal set; The system identifies the second abnormal information in each historical diagnostic record and converts it into the corresponding second dimension value of the abnormal dimension. The values ​​are then weighted and summed according to preset weights to obtain the comprehensive dimension value of each abnormal dimension, thus obtaining the second abnormal set. The weight of each second dimension value is related to the historical time; the closer to the current time, the higher the weight value. The first anomaly set and the second anomaly set are fused according to a preset fusion method to obtain a comprehensive anomaly set, and the comprehensive anomaly set is analyzed to obtain a comprehensive diagnostic result.

[0060] By weighting and summing multiple historical diagnostic records according to time decay weights, the contribution of historical anomalies closer to the present is increased, thus accurately reflecting the recent evolution trend of vehicle faults. Then, by converting the current diagnostic information and historical comprehensive information into numerical sets of anomaly dimensions respectively, and then fusing and analyzing the two sets, a comprehensive diagnostic result is obtained. This improves the fit of the comprehensive diagnostic result to the actual vehicle fault state, reduces misdiagnosis or missed diagnosis caused by ignoring historical evolution, provides a structured data foundation for subsequent automated repair suggestion generation, and improves the accuracy and efficiency of vehicle diagnosis.

[0061] In one embodiment, a computer-readable storage medium is provided storing a computer program that, when executed by a processor, causes the processor to perform the following steps: Establish a bus connection with the vehicle to be inspected in order to obtain the vehicle identification number of the vehicle to be inspected; Perform current vehicle diagnostics based on the vehicle identification code and generate current diagnostic information; Based on the vehicle identification code, retrieve multiple historical diagnostic records associated with the vehicle identification code from a preset server; Identify the first abnormal information in the current diagnostic information and convert it into the first dimension value of the corresponding abnormal dimension to obtain the first abnormal set; The system identifies the second abnormal information in each historical diagnostic record and converts it into the corresponding second dimension value of the abnormal dimension. The values ​​are then weighted and summed according to preset weights to obtain the comprehensive dimension value of each abnormal dimension, thus obtaining the second abnormal set. The weight of each second dimension value is related to the historical time; the closer to the current time, the higher the weight value. The first anomaly set and the second anomaly set are fused according to a preset fusion method to obtain a comprehensive anomaly set, and the comprehensive anomaly set is analyzed to obtain a comprehensive diagnostic result.

[0062] By weighting and summing multiple historical diagnostic records according to time decay weights, the contribution of historical anomalies closer to the present is increased, thus accurately reflecting the recent evolution trend of vehicle faults. Then, by converting the current diagnostic information and historical comprehensive information into numerical sets of anomaly dimensions respectively, and then fusing and analyzing the two sets, a comprehensive diagnostic result is obtained. This improves the fit of the comprehensive diagnostic result to the actual vehicle fault state, reduces misdiagnosis or missed diagnosis caused by ignoring historical evolution, provides a structured data foundation for subsequent automated repair suggestion generation, and improves the accuracy and efficiency of vehicle diagnosis.

[0063] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments described above. Any references to memory, storage, databases, or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), RAMbus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and RAMbus dynamic RAM (RDRAM), etc.

[0064] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0065] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this patent application should be determined by the appended claims.

Claims

1. A vehicle diagnostic method, characterized in that, The method includes: Establish a bus connection with the vehicle to be inspected in order to obtain the vehicle identification number of the vehicle to be inspected; Perform current vehicle diagnostics based on the vehicle identification code and generate current diagnostic information; Based on the vehicle identification code, retrieve multiple historical diagnostic records associated with the vehicle identification code from a preset server; Identify the first abnormal information in the current diagnostic information and convert it into the first dimension value of the corresponding abnormal dimension to obtain the first abnormal set; The system identifies the second abnormal information in each historical diagnostic record and converts it into the corresponding second dimension value of the abnormal dimension. The values ​​are then weighted and summed according to preset weights to obtain the comprehensive dimension value of each abnormal dimension, thus obtaining the second abnormal set. The weight of each second dimension value is related to the historical time; the closer to the current time, the higher the weight value. The first anomaly set and the second anomaly set are fused according to a preset fusion method to obtain a comprehensive anomaly set, and the comprehensive anomaly set is analyzed to obtain a comprehensive diagnostic result.

2. The vehicle diagnostic method according to claim 1, characterized in that, The step of identifying the first abnormal information in the current diagnostic information and converting it into a first dimension value corresponding to the abnormal dimension to obtain the first abnormal set includes: Extract at least one fault code or abnormal data stream parameter from the current diagnostic information; Each extracted fault code or abnormal data stream parameter is mapped to one dimension of a preset multidimensional anomaly space; where each dimension corresponds to a fault type or component. Based on the severity level of the fault code or the degree to which the abnormal data stream parameters deviate from the standard threshold, an initial value is assigned to the mapped dimension to obtain the values ​​of each first dimension in the first abnormal set.

3. The vehicle diagnostic method according to claim 1, characterized in that, The step of identifying the second abnormal information in each historical diagnostic record, converting it into the second dimension value of the corresponding abnormal dimension, and then performing a weighted summation according to preset weights to obtain the comprehensive dimension value of each abnormal dimension, thereby obtaining the second abnormal set, includes: For each historical diagnostic record, a time weighting coefficient is assigned to it; wherein, the time weighting coefficient is calculated using a time decay function; For each abnormal dimension, the second dimension values ​​belonging to that abnormal dimension in all historical diagnostic records are multiplied by the time weight coefficient corresponding to each historical diagnostic record, summed, and then divided by the sum of all time weight coefficients involved in the summation to obtain the comprehensive dimension value of that abnormal dimension, thus obtaining the second abnormal set.

4. The vehicle diagnostic method according to claim 1, characterized in that, The step of fusing the first anomaly set and the second anomaly set according to a preset fusion method to obtain a comprehensive anomaly set includes: Iterate through the first exception set and the second exception set; If an abnormal dimension exists in both the first abnormal set and the second abnormal set, it is recorded as the same abnormal dimension. For any identical abnormal dimension, the dimension values ​​in the first abnormal set and the combined dimension values ​​in the second abnormal set are weighted and summed to obtain the fused dimension value; where the weight corresponding to the current diagnostic information is the first weight, and the weight corresponding to the historical diagnostic information is the second weight. The fused dimension values ​​are added to the comprehensive anomaly set as part of the comprehensive anomaly set.

5. The vehicle diagnostic method according to claim 4, characterized in that, The step of fusing the first anomaly set and the second anomaly set according to a preset fusion method to obtain a comprehensive anomaly set includes: Iterate through the first exception set and the second exception set; If an abnormal dimension exists only in the first abnormal set but not in the second abnormal set, it is denoted as the current abnormal dimension. For any current abnormal dimension, directly add the current abnormal dimension and its corresponding dimension value in the first abnormal set as a new abnormal item to the comprehensive abnormal set; If an abnormal dimension exists only in the second abnormal set and not in the first abnormal set, it is denoted as a historical abnormal dimension. The comprehensive dimension value of the historical anomaly dimension in the second anomaly set is calculated and compared with a preset historical legacy anomaly importance threshold; If the historical anomaly importance threshold is greater than or equal to the historical anomaly importance threshold, it is added to the comprehensive anomaly set as a historical continuation anomaly and marked. If the historical anomaly importance threshold is less than the historical anomaly importance threshold, the historical anomaly dimension is discarded.

6. The vehicle diagnostic method according to claim 1, characterized in that, After the step of fusing the first anomaly set and the second anomaly set according to a preset fusion method to obtain a comprehensive anomaly set, and analyzing the comprehensive anomaly set to obtain a comprehensive diagnostic result, the following steps are included: Obtain the target maintenance recommendations taken by the user based on the comprehensive diagnostic results, as well as the corresponding anomaly dimensions; Using the target repair suggestion action as a positive sample and other repair suggestion actions rejected or skipped by the user as negative samples, the preset weights and / or the preset fusion method are adjusted in reverse.

7. The vehicle diagnostic method according to claim 1, characterized in that, The step of retrieving multiple historical diagnostic records associated with the vehicle identification code from a preset server based on the vehicle identification code includes: Based on the vehicle identification code, retrieve all historical diagnostic records associated with the vehicle identification code and their timestamps from the preset server; Get the preset maximum number of records or the preset time backtracking range; Based on the preset maximum number of records or the preset time backtracking range, select the historical diagnostic records from all historical diagnostic records that are within the maximum number of records or the preset time backtracking range, and use them as the multiple historical diagnostic records.

8. A vehicle diagnostic device, characterized in that, The device includes: The module is used to establish a bus connection with the vehicle to be inspected in order to obtain the vehicle identification number of the vehicle to be inspected. The generation module is used to perform current vehicle diagnosis based on the vehicle identification code and generate current diagnostic information; The retrieval module is used to retrieve multiple historical diagnostic records associated with the vehicle identification code from a preset server based on the vehicle identification code. The identification module is used to identify the first abnormal information in the current diagnostic information and convert it into the first dimension value of the corresponding abnormal dimension to obtain the first abnormal set; The transformation module is used to identify the second abnormal information in each historical diagnostic record and transform it into the second dimension value of the corresponding abnormal dimension. The value is then weighted and summed according to preset weights to obtain the comprehensive dimension value of each abnormal dimension, thus obtaining the second abnormal set. The weight of each second dimension value is related to the historical time; the closer to the current time, the higher the weight value. The fusion module is used to fuse the first anomaly set and the second anomaly set according to a preset fusion method to obtain a comprehensive anomaly set, and to analyze the comprehensive anomaly set to obtain a comprehensive diagnostic result.

9. A computer-readable storage medium, characterized in that, The system contains a computer program that, when executed by a processor, causes the processor to perform the steps of the vehicle diagnostic method as described in any one of claims 1 to 7.

10. An electronic device, characterized in that, The device includes a memory and a processor, the memory storing a computer program that, when executed by the processor, causes the processor to perform the steps of the vehicle diagnostic method as described in any one of claims 1 to 7.

Images