Vehicle electronic component connection anomaly detection method, system and storage medium

Abstract

The present disclosure provides a vehicle electronic component connection anomaly detection method, system and storage medium, relating to the technical field of new energy vehicles, wherein the method at the vehicle end comprises: acquiring the current multi-dimensional operating parameters of the key nodes of the electronic component connection in the new energy vehicle; performing multi-parameter collaborative analysis based on the multi-dimensional operating parameters, acquiring multi-dimensional feature parameters for representing the electronic component connection state, and uploading to a cloud processor; after receiving the fault vehicle information of the existing electronic component connection anomaly predicted by the cloud processor based on the multi-dimensional feature parameters combined with the historical multi-dimensional feature parameters of different vehicles of the same type as the new energy vehicle under the current working condition for clustering analysis, when the new energy vehicle matches the fault vehicle information, push the warning information and repair information. The cooperative mechanism of vehicle end monitoring-cloud analysis can detect vehicle electronic component connection problems in advance and provide early warning, improving the reliability of vehicle electronic component connection.

Description

Technical Field

[0001] This disclosure belongs to the field of new energy vehicle technology, specifically relating to a method, system, and storage medium for detecting abnormal connections of vehicle electronic components. Background Technology

[0002] Abnormal contact in the electronic component connections (also known as connectors or connecting parts) of new energy vehicles, such as loose connections, can lead to increased contact resistance. In particular, high-voltage components, during prolonged charging and discharging, can experience abnormal temperature rise at their contact points, burning out nearby electronic components and potentially damaging the power battery or multi-function controller. The failure of these connecting parts during the operation of new energy vehicles severely restricts their driving safety, necessitating early detection of abnormal electronic component connections via software.

[0003] In related technologies, traditional methods for detecting abnormal connections in electronic components rely on periodic manual inspections during vehicle maintenance to assess whether high-voltage connections are loose. However, this method has limitations such as fixed inspection cycles, high workload, and insufficient reliability, resulting in the inability to identify high-voltage connector faults in a timely manner.

[0004] In related technologies, connector testing methods based on temperature rise parameters or pressure parameters generally have a relatively singular diagnostic dimension, resulting in insufficient accuracy of diagnostic results. Summary of the Invention

[0005] This disclosure provides a method, system, and storage medium for detecting abnormal connections of electronic components in vehicles, aiming to at least partially solve the technical problem that related technologies are unable to accurately detect abnormal connections of electronic components in new energy vehicles.

[0006] At least one embodiment of this disclosure provides a method for detecting abnormal connections of vehicle electronic components, applied to new energy vehicles communicating with cloud processors, including:

[0007] Obtain the current multi-dimensional operating parameters of key nodes connecting electronic components in new energy vehicles; Based on the multi-dimensional operating parameters, multi-parameter collaborative analysis is performed to obtain multi-dimensional feature parameters that characterize the connection status of electronic components, and the multi-dimensional feature parameters are uploaded to the cloud processor. After receiving information from the cloud processor about faulty vehicles with abnormal electronic component connections, based on the multi-dimensional feature parameters combined with historical multi-dimensional feature parameters of different vehicles of the same type as the new energy vehicle under the current operating conditions, and performing cluster analysis, the system pushes early warning information and maintenance information when the new energy vehicle matches the faulty vehicle information.

[0008] The above solution offers the following technical advantages: Addressing the challenge of accurately detecting abnormal connections in electronic components of new energy vehicles using existing technologies, this method proposes a detection approach capable of early detection and warning of connection problems. This approach is suitable for early warning of abnormal connections in high-voltage components on the vehicle side. By setting appropriate key nodes to collect multi-dimensional operating parameters of electronic components, and employing multi-parameter collaborative analysis on the vehicle side, cross-validation of faults is facilitated. Further comparison of multi-dimensional characteristic parameters of similar vehicles under the same operating conditions is conducted in the cloud, identifying vehicles with abnormal connections and reporting them back to the vehicle side. This method establishes a collaborative mechanism of vehicle-side monitoring and cloud-based analysis, effectively identifying abnormal temperature rises caused by increased contact resistance, preventing electronic component burnout, improving detection accuracy and reliability, and enabling proactive warning and precise maintenance of high-voltage connection faults.

[0009] In at least one embodiment of the method provided in this disclosure, the step of performing multi-parameter collaborative analysis based on the multi-dimensional operating parameters to obtain multi-dimensional feature parameters for characterizing the connection state of electronic components includes: The contact resistance of the key node is generated based on the voltage data and the current data; Determine whether there are any abnormal voltage jumps in the voltage data within a set time period, and count the number of abnormal voltage jumps; Determine whether there are any temperature anomalies or abnormal temperature rise rates in the temperature data within a set time period, and statistically analyze the abnormal temperature rise data for those with abnormal temperature or abnormal temperature rise rates; and, Generate multidimensional feature parameters that include the number of voltage anomaly jumps and the temperature rise anomaly data.

[0010] The above solution has the following technical effects: Based on the collected data, the vehicle end can report faults by collecting information such as temperature rise and voltage fluctuations and by coordinating with the cloud processor when there are abnormalities in the connection of electronic components, such as when the connection of high-voltage components is loose.

[0011] In at least one embodiment of the method provided in this disclosure, determining whether there is a temperature anomaly or an abnormal temperature rise rate in the temperature data within a set time period includes: The temperature rise rate data of the key node is generated based on the temperature data; The contact resistance data of the key node is generated based on the voltage data and the current data; Based on the contact resistance data and temperature rise rate data within a set time period, combined with a preset temperature compensation model, it is determined whether the temperature data shows an abnormal temperature rise rate due to increased contact resistance. The temperature compensation model includes different temperature rises caused by different contact stresses. Based on the contact resistance data and the temperature data, a preset resistance-temperature correlation model is used to determine whether there is a temperature anomaly caused by an increase in contact resistance. The resistance-temperature correlation model includes the relationship between the contact resistance data and the temperature rise rate data when there are no abnormalities in voltage jumps and temperature rise.

[0012] The above solution has the following technical effects: improving detection accuracy and reliability.

[0013] In the method provided in at least one embodiment of this disclosure, the key nodes include at least one of a connector terminal and a bolt connection, and the multi-dimensional operating parameters are collected synchronously; and, The number of abnormal voltage jumps is counted by determining whether the difference between the voltage data and the battery voltage exceeds a set threshold. The abnormal temperature rise data is statistically analyzed by determining whether the temperature data and its temperature rise rate data exceed their respective dynamic adaptive thresholds. The dynamic adaptive thresholds are related to the current operating conditions of the new energy vehicle and change with the current operating conditions.

[0014] The above solution has the following technical effects: improving detection accuracy and reliability.

[0015] At least one embodiment of this disclosure provides a method for detecting abnormal connections of vehicle electronic components, applied to a cloud processor communicating with new energy vehicles, including: After receiving the multi-dimensional feature parameters of key nodes sent by the new energy vehicle, a horizontal comparison process is executed. This process combines the multi-dimensional feature parameters with historical multi-dimensional feature parameters of different vehicles of the same type as the new energy vehicle under current operating conditions for cluster analysis to predict faulty vehicle information indicating abnormal electronic component connections. The faulty vehicle information predicted by the horizontal comparison process is sent to the new energy vehicle so that when the new energy vehicle matches the faulty vehicle information, early warning information and maintenance information are pushed.

[0016] The above solution has the following technical effects: it realizes a horizontal comparison mechanism, and filters out abnormal vehicles by comparing vehicle data with historical data of similar vehicles.

[0017] The method provided in at least one embodiment of this disclosure further includes: A longitudinal comparison process is executed, wherein the longitudinal comparison process is used to predict the changing trend of key parameters based on the multi-dimensional feature parameters combined with the historical multi-dimensional feature parameters of the same new energy vehicle, and to predict whether the new energy vehicle will become the faulty vehicle within a set time period based on the changing trend of the key parameters, and to generate new faulty vehicle information when the new energy vehicle becomes the faulty vehicle within the set time period; and The faulty vehicle information predicted by the longitudinal comparison process is sent to the new energy vehicle so that when the new energy vehicle matches the faulty vehicle information, early warning information and maintenance information are pushed.

[0018] The above solution has the following technical effects: it realizes a longitudinal comparison mechanism, realizes fault pre-diagnosis by comparing vehicle data with historical data of the same vehicle, and filters out vehicles that are about to experience abnormalities.

[0019] In at least one embodiment of the method provided in this disclosure, the multidimensional feature parameters include temperature value, voltage value, number of abnormal voltage jumps, and abnormal temperature rise data; and, The horizontal comparison process includes: The first K-Means clustering model is used to determine whether the multidimensional feature parameters meet a preset first-type judgment condition, wherein the first-type judgment condition includes that the current temperature value is higher than the temperature values ​​of different vehicles of the same type and that the number of abnormal voltage jumps is multiple; and... If so, determine whether the new energy vehicle is a faulty vehicle with abnormal electronic component connections; The longitudinal comparison process includes: The second K-Means clustering model is used to determine whether the multidimensional feature parameters meet the preset second type of judgment conditions. The second type of judgment conditions include that the current temperature change trend is different from the historical temperature change trend of the same new energy vehicle, and the current voltage value is different from the historical voltage value of the same new energy vehicle, and the current temperature value is higher than each of the historical temperature values ​​of the same new energy vehicle, and the number of abnormal voltage jumps is multiple. If so, the information indicates that the new energy vehicle is a faulty vehicle with abnormal connection of electronic components.

[0020] The above solution has the following technical effects: improving detection accuracy and reliability.

[0021] At least one embodiment of this disclosure also provides a vehicle electronic component connection anomaly detection system, applied to new energy vehicles communicating with cloud processors, including: The vehicle-side acquisition unit is configured to acquire the current multi-dimensional operating parameters of key nodes connecting electronic components in new energy vehicles. The vehicle-side first-level processing unit is configured to perform multi-parameter collaborative analysis based on the multi-dimensional operating parameters, obtain multi-dimensional feature parameters characterizing the connection status of electronic components, and upload the multi-dimensional feature parameters to the cloud processor; and, The vehicle-side second-level processing unit is configured to receive feedback from the cloud processor, based on the multi-dimensional feature parameters combined with the historical multi-dimensional feature parameters of different vehicles of the same type as the new energy vehicle under the current operating conditions, and perform cluster analysis to predict the presence of faulty vehicles with abnormal electronic component connections. When the new energy vehicle matches the faulty vehicle information, it pushes early warning information and maintenance information.

[0022] At least one embodiment of this disclosure also provides a vehicle electronic component connection anomaly detection system, applied to a cloud processor communicating with new energy vehicles, including: The cloud-based diagnostic unit is configured to receive multi-dimensional feature parameters of key nodes sent by the new energy vehicle, and then execute a horizontal comparison process. This horizontal comparison process combines the multi-dimensional feature parameters with historical multi-dimensional feature parameters of different vehicles of the same type as the new energy vehicle under current operating conditions for cluster analysis to predict faulty vehicle information related to abnormal electronic component connections; and... The cloud-based sending unit is configured to send the faulty vehicle information predicted by the horizontal comparison process to the new energy vehicle, so as to push early warning information and maintenance information when the new energy vehicle matches the faulty vehicle information.

[0023] At least one embodiment of this disclosure also provides a storage medium storing a program or instructions, wherein the program or instructions, when executed by a processor, implement the steps of the method provided in any embodiment of this disclosure.

[0024] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and are not intended to limit this disclosure. Attached Figure Description

[0025] To more clearly illustrate the technical solutions in the embodiments of this disclosure, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this disclosure. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0026] Figure 1 A flowchart of a vehicle-side electronic component connection anomaly detection method provided for at least one embodiment of this disclosure; Figure 2 Flowchart of a multi-dimensional operating parameter acquisition scheme provided for at least one embodiment of this disclosure; Figure 3 A flowchart of a multi-parameter collaborative analysis scheme provided for at least one embodiment of this disclosure; Figure 4 Flowchart of another multi-parameter collaborative analysis scheme provided for at least one embodiment of this disclosure; Figure 5 A flowchart of a cloud-based method for detecting abnormal connections of vehicle electronic components, provided for at least one embodiment of this disclosure; Figure 6 Flowchart of another cloud-based method for detecting abnormal connections of vehicle electronic components provided in at least one embodiment of this disclosure; Figure 7 A flowchart illustrating an example of vehicle electronic component connection anomaly detection provided in at least one embodiment of this disclosure; Figure 8 A structural block diagram of a vehicle electronic component connection anomaly detection system provided in at least one embodiment of this disclosure; Figure 9 A structural block diagram of a cloud-based vehicle electronic component connection anomaly detection system provided for at least one embodiment of this disclosure; Figure 10 A structural block diagram of a program product provided for at least one embodiment of this disclosure.

[0027] Figure label: 100 - Vehicle-side electronic component connection anomaly detection system; 101 - Vehicle-side acquisition unit; 102 - Vehicle-side first-level processing unit; 103 - Vehicle-side second-level processing unit; 200 - Cloud-based vehicle electronic component connection anomaly detection system; 201 - Cloud-based diagnostic unit; 202 - Cloud-based transmission unit; 301 - Processor; 302 - Memory; 303 - Input device; 304 - Output device. Detailed Implementation

[0028] The present disclosure will now be described in further detail with reference to the accompanying drawings and embodiments. It should be particularly noted that the following embodiments are for illustrative purposes only and do not limit the scope of the disclosure. Similarly, the following embodiments are only some, not all, embodiments of the present disclosure, and all other embodiments obtained by those skilled in the art without inventive effort are within the scope of protection of this disclosure.

[0029] The terms "first," "second," and "third" used in the embodiments of this disclosure are for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Therefore, a feature defined as "first," "second," and "third" may explicitly or implicitly include at least one of that feature.

[0030] In the description of this disclosure, "multiple" means at least two, such as two or three, unless otherwise expressly and specifically limited.

[0031] In this disclosure, the terms "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., refer to a specific feature, structure, material, or characteristic described in connection with that embodiment or example, which is included in at least one embodiment or example of this disclosure. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0032] The terms “comprising” and “having”, and any variations thereof, used in this disclosure are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or device that includes a series of steps or units is not limited to the steps or units listed, but may optionally include steps or units not listed, or may optionally include other steps or components inherent to such processes, methods, products, or devices.

[0033] In this disclosure, the term "electronic component" includes, but is not limited to, high-voltage components whose supply voltage is higher than a set voltage. High-voltage components include power semiconductor devices, high-voltage relays, high-voltage distribution boxes, and DC-DC converters, etc.

[0034] The term "multi-parameter collaborative analysis" in this disclosure refers to an analytical method that comprehensively collects multiple parameters such as voltage, temperature, and current data from vehicle electronic component connections, and simultaneously analyzes and correlates these parameters using a pre-defined algorithm model to identify the coupling relationships and abnormal change trends between the parameters, thereby accurately determining whether there are any anomalies in the electronic component connections. This method overcomes the limitations of single-parameter detection and improves the accuracy and robustness of anomaly detection through the collaborative fusion of multi-dimensional data.

[0035] In this disclosure, the term "cluster analysis" refers to a data analysis method that automatically groups and classifies multidimensional feature parameters according to the similarity of data features during the detection of abnormal connections in vehicle electronic components. This method, through unsupervised learning, aggregates samples with similar data distributions or abnormal features into different clusters, effectively uncovering hidden patterns and associations within the data set. This helps the detection system quickly distinguish between normal and abnormal data clusters, thereby improving the efficiency and accuracy of anomaly identification and providing a clear classification basis for subsequent anomaly cause analysis.

[0036] The term "K-Means clustering model" in this disclosure refers to an unsupervised clustering algorithm that automatically groups data samples based on a distance metric (such as Euclidean distance). Its core is to divide the data into K preset clusters through an iterative optimization process, maximizing the similarity of samples within each cluster and minimizing the similarity between samples in different clusters. In the scenario of detecting abnormal connections in vehicle electronic components, this model can perform clustering processing on collected multi-dimensional operating parameters such as voltage, temperature, and current. By setting a reasonable K value, normal data and abnormal data with similar characteristics are aggregated into different clusters, effectively helping the detection system quickly identify abnormal states in the connections of electronic components.

[0037] The term "first K-Means clustering model" in this embodiment refers to a clustering analysis performed on the multidimensional feature parameters of a group of vehicles of the same type, dividing the group into a common normal feature cluster of the vehicle type. This cluster serves as a benchmark for horizontal comparison and can quickly screen out faulty vehicles in the group whose features deviate from the common normal range.

[0038] The term "second K-Means clustering model" in this embodiment refers to the construction of a unique normal feature cluster for the same new energy vehicle under similar operating conditions based on its historical operating data. This cluster serves as a benchmark for longitudinal comparison and can quickly screen out abnormal parameter points in the current operating data of the vehicle that deviate from its unique normal feature cluster, thereby enabling longitudinal tracking of feature changes in an individual vehicle over a long period of time.

[0039] The term "abnormal temperature rise" in the embodiments of this disclosure includes two types: abnormal temperature rise rate and abnormal temperature.

[0040] The term "abnormal temperature rise rate" in this embodiment refers to the judgment of abnormal temperature rise based on different current values. The normal temperature rise rate corresponding to a certain current value needs to be obtained based on experimental data and supplemented by the database.

[0041] In this disclosure, the term "temperature anomaly" refers to a situation where the temperature value actually measured at the connection point of a vehicle's electronic components exceeds the normal temperature threshold range under the same operating conditions determined based on test data or historical normal operation data.

[0042] The technical approach involved in this disclosure will be briefly described below.

[0043] In related technologies, it is not uncommon for high-voltage components in new energy vehicles to burn out due to loose connections after prolonged operation.

[0044] To address the technical challenge of accurately detecting abnormal connections in electronic components of new energy vehicles using existing technologies, this disclosure proposes a detection method capable of early detection and warning of such connection problems. This method is applicable to early warning of abnormal connections in high-voltage components on the vehicle side. The method collects multi-dimensional operating parameters of electronic components by setting appropriate key nodes. Multi-parameter collaborative analysis is employed at the vehicle end to facilitate cross-verification of faults. Further comparison of multi-dimensional characteristic parameters of similar vehicles under the same operating conditions is conducted in the cloud, identifying vehicles with abnormal connections and reporting this information back to the vehicle end. This method establishes a collaborative mechanism of vehicle-side monitoring and cloud-based analysis, effectively identifying abnormal temperature rises caused by increased contact resistance, preventing electronic component burnout, improving detection accuracy and reliability, and enabling proactive warning and precise maintenance of high-voltage connection faults.

[0045] Based on this, the method disclosed herein sets reasonable deployment points for temperature sensors, voltage sensors, and current sensors, and collects voltage, temperature, and current data of key nodes in real time and synchronously. In multi-parameter collaborative analysis, based on the synchronously collected voltage, temperature, and current data of electronic components, combined with the contact resistance calculation model, cross-verification of faults is achieved, thereby improving detection accuracy.

[0046] Based on this, the method disclosed herein combines the voltage data (including voltage jump information), temperature data, and contact resistance data of key nodes with the resistance-temperature correlation model and temperature compensation model established by the experiment to achieve joint electrical-thermal diagnosis, so as to count the number of abnormal voltage jumps and abnormal temperature rise data in real time at the vehicle end, and upload them to the cloud as multi-dimensional feature parameters.

[0047] Building upon this, the method disclosed herein implements a horizontal comparison mechanism, which compares real-time vehicle data with historical data of similar vehicles to filter out faulty vehicles. The cloud further utilizes the first K-Means clustering model to horizontally compare the operating conditions of similar vehicles, filtering out abnormal vehicles and pushing them for repair, thus forming a collaborative mechanism between vehicle-side monitoring and cloud-based analysis.

[0048] Building upon this foundation, the method disclosed herein implements a longitudinal comparison mechanism, which compares real-time vehicle data with historical data of the same vehicle to achieve fault pre-diagnosis and further filter faulty vehicles. The cloud-based system further utilizes a second K-Means clustering model to longitudinally compare the historical operating conditions of the new energy vehicle. This longitudinal comparison mechanism, in conjunction with the horizontal comparison mechanism, constructs a two-layer detection network covering the entire lifecycle of a single vehicle and similar operating conditions for multiple vehicles. This further enhances the reliability of vehicle-cloud collaborative diagnosis, facilitating early detection and timely intervention of abnormal connections in electronic components of new energy vehicles.

[0049] Based on this, the method disclosed herein reduces misjudgments due to environmental interference by integrating data-driven approaches with physical models, enabling proactive early warning and precise operation and maintenance of high-voltage connection faults. This method requires no manual commands, achieving closed-loop management across the entire chain from data acquisition and interference elimination to early warning push notifications.

[0050] Figure 1 This is a flowchart illustrating a method for detecting abnormal connections of vehicle electronic components in a vehicle, provided in at least one embodiment of this disclosure. This method can be applied to new energy vehicles communicating with cloud processors, including but not limited to vehicle-side controllers. Figure 1 As shown, the method may include the following steps S10-S30.

[0051] Step S10: Obtain the current multi-dimensional operating parameters of the key nodes connecting electronic components in the new energy vehicle.

[0052] Step S20: Perform multi-parameter collaborative analysis based on multi-dimensional operating parameters to obtain multi-dimensional feature parameters used to characterize the connection status of electronic components, and upload the multi-dimensional feature parameters to the cloud processor.

[0053] Step S30: After receiving the fault vehicle information with abnormal electronic component connection based on the cluster analysis of the multi-dimensional feature parameters and the historical multi-dimensional feature parameters of different vehicles of the same type as new energy vehicles under the current working conditions, which is based on the feedback from the cloud processor, when the information of new energy vehicles and fault vehicles is matched, the warning information and maintenance information are pushed.

[0054] It should be noted that the cloud processor stores historical multidimensional feature parameters of different vehicles under different operating conditions, which can be accessed. The historical multidimensional feature parameters of the current operating condition can be the historical multidimensional feature parameters of the current operating condition point, or the historical multidimensional feature parameters of other operating condition points similar to the current operating condition point.

[0055] During implementation, through steps S10-S30, the vehicle-side monitors the system, while the cloud uses cluster analysis to filter out faulty vehicles. By comparing the filtered operating conditions horizontally, diagnosis is achieved. The cloud can also diagnose abnormalities in the connections of electronic components exhibiting abnormal temperatures under similar operating conditions.

[0056] In the above scheme, this disclosure does not limit the multi-dimensional operating parameters and their acquisition scheme in step S10. In practical application scenarios, in addition to the schemes described in the following embodiments, supplementary parameters such as pressure data and vibration amplitude data of key nodes can be collected synchronously through the vehicle CAN bus system, or a distributed fiber optic sensor array can be used to continuously monitor the connection parts of electronic components to obtain their strain data and micro-displacement data. In addition, in some scenarios, the original collected data can be preprocessed by combining the vehicle edge computing unit to perform noise reduction, normalization and other preprocessing operations, and then the processed effective parameters can be uploaded to the subsequent analysis module to adapt to the hardware configuration and monitoring needs of different vehicle models. At the same time, for monitoring scenarios in extreme environments, wireless sensor network technology can be introduced to realize the collection of key node parameters in areas where wiring is difficult, further expanding the coverage and flexibility of multi-dimensional operating parameter acquisition. When the system executes step S10, it can dynamically select appropriate multi-dimensional operating parameters and their acquisition schemes according to the real-time operating status of the vehicle and the preset monitoring strategy.

[0057] In the above scheme, this disclosure does not limit the multi-parameter collaborative analysis scheme in step S20. In practical application scenarios, in addition to the schemes described in the following embodiments, collaborative analysis can also be achieved by combining multi-source parameter feature fusion with machine learning models: for example, extracting and fusing the time-domain features (such as peak or root mean square) and frequency-domain features (such as characteristic frequencies) of current data, temperature data, voltage data, pressure data, and vibration amplitude data, and inputting them into a pre-trained random forest or support vector machine model, and achieving rapid identification of connection anomalies by learning the parameter distribution law under normal working conditions; or, for continuously monitored micro-displacement data, using time series prediction algorithms (such as LSTM neural networks) to build a trend model, when the deviation between the predicted value and the actual collected value exceeds a preset threshold, it is judged as a potential anomaly; in addition, a parameter correlation analysis mechanism can be introduced, if a sudden change in vibration amplitude and strain data deviation occur simultaneously, or if pressure data fluctuation and micro-displacement change show a strong correlation, triggering multi-level anomaly warning, further improving the accuracy and timeliness of detection. When executing step S20, the system can dynamically select a single or combined multi-parameter collaborative analysis scheme according to the complexity and accuracy requirements of the actual monitoring scenario, so as to adapt to the monitoring needs of different vehicle electronic component connection structures.

[0058] In the above scheme, this disclosure does not limit the prediction scheme based on cluster analysis in step 30. In practical application scenarios, in addition to the scheme described in the following embodiments, the DBSCAN algorithm can also be used to perform density clustering on multi-dimensional monitoring parameters. By setting a reasonable neighborhood radius and minimum number of points, outliers that deviate significantly from the normal operating condition clusters can be automatically identified and marked as candidates for potential connectivity anomalies. At the same time, the time series characteristics of the monitoring parameters can be combined to track and analyze the sequence of continuously occurring outliers. If the duration of outliers exceeds a preset duration or they appear densely in a short period of time, it is determined that there is a faulty vehicle, further reducing the false judgment rate and ensuring the reliability and adaptability of the prediction scheme. When the system executes step S30, it can select a suitable prediction scheme based on cluster analysis according to the complexity and accuracy requirements of the actual monitoring scenario to adapt to the monitoring needs of different vehicle electronic component connection structures.

[0059] Some embodiments of this disclosure also provide systems, storage media, and program products corresponding to the methods described above.

[0060] The method provided by at least one embodiment of this disclosure is applicable to any existing new energy vehicle application scenario that requires the detection of abnormal connections in high-voltage components. For example, in the scenario of monitoring the high-voltage wiring harness connection of the battery pack in a pure electric passenger vehicle, the high-voltage wiring harness between the battery pack and the motor controller is prone to loosening due to long-term vibration. By using the method of this disclosure to perform density clustering on multi-dimensional parameters such as voltage fluctuations and current stability, abnormal parameter outliers caused by loose connections can be quickly identified. Another example is the monitoring of loose bolts at the connection between the motor and the gearbox in a hybrid commercial vehicle. By using time series tracking to identify continuously occurring abnormal torque outliers, if the duration exceeds a set time, the vehicle is determined to be faulty. Yet another example is the detection of copper busbar connections inside the high-voltage distribution box of a new energy bus. By setting an appropriate neighborhood radius and minimum number of points, parameter deviations caused by increased contact resistance due to oxidation can be accurately captured, effectively preventing circuit faults caused by abnormal connections.

[0061] In some embodiments, Figure 1 Based on the proposed solution, key nodes in step S10 may include connector terminals or bolted connections. Connector terminals are prone to wear, oxidation, or deformation of the contact surface due to prolonged insertion and removal, leading to decreased contact pressure and increased contact resistance. Bolted connections may loosen due to vibration, thermal expansion and contraction, reducing the contact area or causing insufficient contact pressure, which can also result in abnormally high contact resistance. For these key nodes, the system can deploy appropriate sensors to ensure effective monitoring of the connection status of electronic components.

[0062] In the above scheme, critical nodes need to cover parts of electronic components that are prone to loose connections or poor contact. One or more critical nodes may be included.

[0063] In some embodiments, Figure 1 Based on the scheme, the multi-dimensional operating parameters in step S10 include real-time and synchronously acquired voltage, temperature, and current data. The multi-dimensional characteristic parameters in step S20 may include, but are not limited to, a first parameter related to temperature data, a second parameter related to voltage data and used to characterize voltage fluctuations, and a third parameter related to current data. Voltage data directly reflects changes in contact resistance at key locations; when terminals or bolts become loose, increased contact resistance leads to abnormal fluctuations in local voltage drops. Temperature data captures the heating state of connection points; loose connections can cause poor contact, resulting in localized overheating, with temperature values ​​and fluctuation amplitudes deviating from the normal operating range. Current data can be used in conjunction with voltage and temperature data for verification, and the three together constitute a multi-dimensional monitoring matrix.

[0064] Figure 2 A flowchart illustrating a multi-dimensional operational parameter acquisition scheme provided for at least one embodiment of this disclosure. Figure 1 Based on the existing plan, in order to improve the sensitivity and accuracy of the detection, such as Figure 2 As shown, step S10 may further include the following sub-steps S101-S103.

[0065] Sub-step S101: Collect voltage data by deploying voltage sensors at key nodes.

[0066] Sub-step S102: Collect temperature data by deploying temperature sensors along the relevant temperature conduction paths at key nodes.

[0067] Sub-step S103: Acquire current data by using a current sensor connected in series with the circuit containing the electronic component.

[0068] In particular, sub-steps S101-S103 provide comprehensive and reliable data support for the accurate detection of abnormal connections of vehicle electronic components, effectively improving the sensitivity and accuracy of detection, helping to discover potential faults in a timely manner and take corresponding maintenance measures, and ensuring the safety and stability of vehicle operation.

[0069] In the above scheme, the temperature conduction path must cover the electronic component body, connection contact points, and adjacent heat-conducting structures to ensure that the collected temperature data can accurately reflect the thermal anomaly state of the connection node. The current sensor must be a high-precision, wide-range model, and its installation must ensure reliable series connection with the circuit without introducing additional impedance interference.

[0070] In some embodiments, Figure 1 or Figure 2Based on the proposed solution, to improve the coverage and accuracy of connection anomaly detection, the multi-dimensional feature parameters in step S20 include at least the number of voltage anomaly jumps and abnormal temperature rise data. These two feature parameters, from the perspectives of transient changes in electrical signals and cumulative effects of thermal states, jointly characterize the abnormal features of electronic component connections. Combined with previously collected current data, this collaborative analysis further improves the coverage and accuracy of connection anomaly detection, avoiding potential missed or false positives that might occur with single-parameter detection.

[0071] Figure 3 A flowchart illustrating a multi-parameter collaborative analysis scheme provided for at least one embodiment of this disclosure. Figure 1 or Figure 2 Based on the existing solution, in order to improve the accuracy of detecting abnormal connections in vehicle electronic components, such as... Figure 3 As shown, step S20 may further include the following sub-steps S201-S204.

[0072] Sub-step S201: Generate the contact resistance of key nodes based on voltage and current data.

[0073] Sub-step S202: Determine whether there are abnormal voltage jumps in the voltage data within the set time period, and count the number of abnormal voltage jumps.

[0074] Sub-step S203: Determine whether there are temperature anomalies or temperature rise rate anomalies in the temperature data within the set time period, and count the abnormal temperature rise data of temperature anomalies or temperature rise rate anomalies.

[0075] Sub-step S204: Generate multi-dimensional feature parameters containing data on the number of voltage anomaly jumps and temperature rise anomalies.

[0076] It should be noted that abnormal temperature rise data includes, but is not limited to, the number of abnormal temperature rises.

[0077] Step S201 aims to calculate contact resistance, sub-step S202 aims to perform voltage anomaly statistics, and sub-step S203 aims to perform temperature rise rate and temperature anomaly statistics at the connection point. Sub-steps S201-S204 construct a multi-dimensional feature system for the connection status of key nodes, organically combining the static characteristics of contact resistance, the dynamic fluctuation characteristics of voltage jumps, and the thermal characteristics of temperature anomalies. This provides comprehensive and discriminative feature inputs for subsequent intelligent judgment of connection anomalies based on machine learning models or preset rule engines. These multi-dimensional feature parameters can more accurately depict the gradual transition of connection nodes from normal to abnormal, effectively capturing potential abnormal signals that are easily overlooked by traditional single-parameter detection. This further improves the accuracy and robustness of connection anomaly detection for vehicle electronic components, reduces false positives and false negatives, and provides more reliable technical protection for vehicle operation safety.

[0078] In some embodiments, Figure 3 Based on the proposed solution, to achieve rapid detection, sub-step S202 counts of abnormal voltage jumps by determining whether the difference between the voltage data and the battery voltage exceeds a set threshold. This set threshold can be determined comprehensively based on the rated operating voltage range of the vehicle's electronic components, historical fault data, and safety redundancy requirements. Simultaneously, during the statistical process, transient voltage fluctuations caused by transient conditions such as vehicle start-up and rapid acceleration must be excluded. This can be achieved by introducing a time window filtering mechanism to filter out invalid interference data. For example, only when the voltage difference exceeds the set threshold for three consecutive sampling periods is it considered a valid abnormal jump, further improving the robustness of the overall detection solution.

[0079] In the above scheme, the threshold is set to dynamically adapt to the normal operating conditions, and the threshold for different operating conditions can be obtained through calibration.

[0080] In some embodiments, Figure 3 Based on the scheme, in order to improve the accuracy of temperature rise anomaly judgment, the temperature rise anomaly judgment scheme of sub-step S203 further includes the following sub-steps 203a-203d.

[0081] Sub-step 203a: Generate temperature rise rate data for key nodes based on temperature data.

[0082] Sub-step 203b: Generate contact resistance data for key nodes based on voltage and current data.

[0083] Sub-step 203c: Based on the contact resistance data and temperature rise rate data within a set time period, combined with the preset temperature compensation model, determine whether there is an abnormal temperature rise rate caused by the increase in contact resistance. The temperature compensation model includes different temperature rises caused by different contact stresses.

[0084] Sub-step 203d: Based on the contact resistance data and temperature data, combined with the preset resistance-temperature correlation model, determine whether there is a temperature anomaly caused by the increase in contact resistance. The resistance-temperature correlation model includes the relationship between contact resistance data and temperature rise rate data when there are no abnormalities in voltage jumps and temperature rise.

[0085] It should be noted that the temperature compensation model is based on the temperature data collected from experiments. The resistance-temperature correlation model is based on the contact resistance data collected from experiments combined with temperature data.

[0086] Through the coordinated operation of sub-steps 203a and 203d, multi-dimensional cross-verification of temperature rise anomalies at key nodes can be achieved from two core dimensions: temperature rise rate and contact resistance. This composite judgment method not only captures the dynamic trend of temperature changes through temperature rise rate data, but also traces the root cause of temperature rise anomalies using contact resistance data. Simultaneously, it uses a temperature compensation model to correct temperature rise deviations under different contact stresses and eliminates interference from voltage jumps without anomalies through a resistance-temperature correlation model. This effectively improves the accuracy of temperature rise anomaly detection, reduces the risk of missed or false alarms due to misjudgment of a single indicator, and provides more reliable technical support for real-time monitoring and anomaly early warning of electronic component connection status.

[0087] In the above scheme, the temperature compensation model can be simulated using a test bench to study the different temperature rises caused by electronic components under different contact stresses. For example, due to different contact stresses, a 10Ω resistor corresponds to a temperature rise of 20℃ and 30℃, respectively.

[0088] In some embodiments, Figure 3 Based on the scheme, in sub-step S203, abnormal temperature rise data is statistically analyzed by determining whether the temperature data and its rate of temperature rise exceed their respective dynamic adaptive thresholds. The dynamic adaptive threshold is related to the current operating conditions of the new energy vehicle and changes with these conditions. The current operating conditions specifically include key parameters such as vehicle speed, power load level, remaining battery charge, ambient temperature, and air conditioning operation status. For example, when the vehicle is under high-load conditions such as high-speed cruising or rapid acceleration, the normal heat generation at the electronic component connections will increase significantly. In this case, the dynamic adaptive threshold will float upwards according to a preset operating condition-threshold mapping relationship to adapt to the reasonable temperature rise range under this condition, avoiding misjudging temperature fluctuations under normal operating conditions as abnormal. Conversely, under light-load conditions such as low-speed driving or idling, the threshold will be lowered accordingly to ensure timely detection of minor temperature rise anomalies. Furthermore, the dynamic adaptive threshold is updated in real-time. By embedding an edge computing module into the vehicle controller, and combining typical scenario data from the historical operating condition database with multi-dimensional parameters collected by current sensors, the threshold setting is continuously optimized to ensure it always matches the actual operating state of the vehicle. This dynamic adjustment mechanism complements the aforementioned resistance-temperature correlation model and temperature compensation model, further enhancing the effect of multi-dimensional cross-validation and ensuring the accuracy and robustness of abnormal temperature rise detection under complex and ever-changing vehicle operating conditions.

[0089] In the above scheme, the dynamic adaptive threshold is dynamically adapted according to the normal operating conditions, and the dynamic adaptive threshold for different operating conditions can be obtained through calibration.

[0090] Figure 4 A flowchart illustrating another multi-parameter collaborative analysis scheme provided for at least one embodiment of this disclosure. Figure 1 or Figure 2 Based on the solution, in order to improve the accuracy of detecting abnormal connections of vehicle electronic components, step S20 may further include the following sub-step S201. - Sub-step S205 .

[0091] Sub-step S201 : Generate the contact resistance of key nodes based on voltage and current data.

[0092] Sub-step S202 : Determine whether there are abnormal voltage jumps in the voltage data within a set time period, and count the number of abnormal voltage jumps.

[0093] Sub-step S203 : Determine whether there are temperature anomalies or abnormal temperature rise rates in the temperature data within a set time period, and count the abnormal temperature rise data for those with abnormal temperature or abnormal temperature rise rates.

[0094] Sub-step S204 : Determine whether the contact resistance exceeds the set threshold within a set time period, and count the number of contact resistance anomalies where the contact resistance exceeds the set threshold.

[0095] Sub-step S205 Generate multidimensional feature parameters that include the number of abnormal voltage jumps, abnormal temperature rise data, and the number of abnormal contact resistances.

[0096] It should be noted that step S204 The purpose is to statistically analyze abnormal contact resistances exceeding a set threshold.

[0097] Among them, through sub-step S201 - Sub-step S205 The constructed multi-dimensional feature parameters can comprehensively integrate abnormal information from key nodes in terms of electrical characteristics (voltage, current, contact resistance) and thermal characteristics (temperature, rate of temperature rise), providing multi-dimensional and highly correlated analytical basis for subsequent anomaly judgment. Compared with detection schemes relying on a single parameter, this multi-parameter collaborative fusion approach can effectively reduce the risk of misjudgment caused by accidental fluctuations in a single indicator and improve the accuracy of anomaly identification. For example, when the number of abnormal contact resistance events exceeds a set threshold and is accompanied by abnormal temperature rise data, the system can quickly identify it as a high-risk connection anomaly and trigger a real-time warning signal, so that maintenance personnel can take timely intervention measures to ensure the stability of vehicle electronic component connections and driving safety.

[0098] Figure 5This is a flowchart illustrating a cloud-based method for detecting abnormal connections of vehicle electronic components, provided in at least one embodiment of this disclosure. This method can be applied to cloud processors used in communication for new energy vehicles. Figure 5 As shown, the method includes the following steps S21-S22.

[0099] Step S21: After receiving the multi-dimensional feature parameters of the key nodes sent by the new energy vehicle, execute the horizontal comparison process. The horizontal comparison process is used to combine the multi-dimensional feature parameters with the historical multi-dimensional feature parameters of different vehicles of the same type as the new energy vehicle under the current working conditions for cluster analysis to predict the fault vehicle information of abnormal electronic component connection.

[0100] Step S22: The faulty vehicle information predicted by the horizontal comparison process is sent to the new energy vehicle, so that when the information of the new energy vehicle matches that of the faulty vehicle, the warning information and maintenance information are pushed.

[0101] The aforementioned horizontal comparison process can filter out abnormal vehicles by comparing data from multiple vehicles.

[0102] Specifically, steps S21-S22 enable cross-vehicle lateral monitoring of the connection status of new energy electronic components. By clustering and comparing operating data from similar vehicles, vehicles with potential connection anomalies can be quickly located, thus triggering an early warning mechanism. This cloud-vehicle collaborative approach not only improves the timeliness and accuracy of anomaly detection but also provides timely maintenance guidance for vehicle users, reducing potential safety risks caused by abnormal electronic component connections and ensuring the operational stability of new energy vehicles. Furthermore, this process optimizes the clustering model by incorporating historical fault data from similar vehicles, continuously improving the accuracy of predictions and providing data support for proactive prevention and control of subsequent vehicle electronic component connection anomalies.

[0103] In the above scheme, the horizontal comparison process is configured to statistically analyze the temperature difference and voltage jump difference at the high-voltage connection point between the new energy vehicle and different vehicles of the same type under similar operating conditions. If the temperature is significantly higher than the overall data baseline and the voltage jumps multiple times, the new energy vehicle is determined to be a faulty vehicle with abnormal electronic component connections. Similar operating conditions can include multiple other operating conditions that are similar to the current operating condition. The similarity here can be defined by the similarity of key operating parameters such as vehicle speed range, remaining battery charge, ambient temperature range, and type of road segment. For example, if the current operating condition is constant speed driving on a highway, speed 80~100km / h, remaining battery charge 40%~60%, and ambient temperature 15~20℃, then similar operating conditions can cover other operating conditions within the range of speed 75-105km / h, SOC 38%-62%, and temperature 13-22℃.

[0104] Figure 6 A flowchart illustrating another cloud-based method for detecting abnormal connections of vehicle electronic components, provided for at least one embodiment of this disclosure. Figure 5 Based on the existing plan, in order to compensate for the shortcomings of horizontal comparison in monitoring changes in individual vehicles, such as... Figure 6 As shown, the method also includes steps S23-S24.

[0105] Step S23: Execute the longitudinal comparison process, wherein the longitudinal comparison process is used to predict the changing trend of key parameters based on multi-dimensional feature parameters combined with the historical multi-dimensional feature parameters of the same new energy vehicle, and to predict whether the new energy vehicle will become a faulty vehicle within a set time period based on the changing trend of key parameters, and to generate new faulty vehicle information when the new energy vehicle becomes a faulty vehicle within the set time period.

[0106] Step S24: The faulty vehicle information predicted by the longitudinal comparison process is sent to the new energy vehicle, so that when the information of the new energy vehicle matches that of the faulty vehicle, the warning information and maintenance information are pushed.

[0107] It should be noted that the above longitudinal comparison process can achieve fault pre-diagnosis by comparing the vehicle's historical data.

[0108] Steps S23-S24 can be executed in parallel with steps S21-S22, or placed after steps S21-S22. Steps S23-S24 enable longitudinal tracking of the connection status of electronic components in a single new energy vehicle over time. By clustering and comparing multi-dimensional feature parameters of the same vehicle at different times, progressive anomalies or potential degradation trends in the connection status of electronic components can be accurately captured, compensating for the shortcomings of horizontal comparison in monitoring individual changes in a single vehicle. This collaborative mechanism of longitudinal tracking and horizontal comparison further improves the anomaly detection method, enabling the discovery of common connection problems across vehicles as well as identifying individual hidden dangers in a single vehicle, significantly improving the coverage and accuracy of anomaly detection. Simultaneously, the historical trend data generated by the longitudinal comparison process can be deeply integrated with the horizontal comparison data, providing richer sample dimensions for the iterative optimization of the clustering model. This helps build a more robust anomaly prediction model, providing solid technical support for the full-cycle prevention and control of connection anomalies in electronic components of new energy vehicles, further ensuring the safety and stability of vehicle operation.

[0109] In the above scheme, the longitudinal comparison process can be configured to statistically analyze the temperature and voltage differences at the high-voltage connection under similar operating conditions in the vehicle's historical data. If the temperature is significantly higher than the historical benchmark and the voltage jumps multiple times, the new energy vehicle is determined to be a faulty vehicle with abnormal electronic component connections.

[0110] In some embodiments, Figure 5 or Figure 6 Based on the solution, in order to improve the accuracy of cloud-based diagnosis, multi-dimensional feature parameters include temperature value, voltage value, number of abnormal voltage jumps and abnormal temperature rise data. Furthermore, the horizontal comparison process in step S21 may further include the following sub-steps S211-S212.

[0111] Sub-step S211: Determine whether the multidimensional feature parameters meet the preset first-class judgment conditions through the first K-Means clustering model. The first-class judgment conditions include that the current temperature value is higher than the temperature values ​​of different vehicles of the same type and the number of abnormal voltage jumps is multiple.

[0112] Sub-step S212: If yes, determine whether the new energy vehicle is a faulty vehicle with abnormal electronic component connections.

[0113] Specifically, sub-steps S211-S212 perform a horizontal comparative screening of the connection status of electronic components among vehicles of the same type. The first K-Means clustering model is used to cluster the multidimensional feature parameters of vehicles of the same type under similar operating conditions, dividing them into normal and abnormal feature clusters. When the multidimensional feature parameters of the vehicle under test fall into the abnormal feature cluster, and simultaneously meet the conditions of a temperature value higher than the average of vehicles of the same type and a voltage abnormal jump count reaching a preset threshold, the system combines the clustering results of the horizontal comparison with preset anomaly judgment rules to generate fault vehicle information including the fault vehicle ID, abnormal feature parameter values, and the time period of the anomaly.

[0114] In some embodiments, Figure 6 Based on the solution, in order to further improve the accuracy of cloud-based diagnosis, the longitudinal comparison process in step S23 may further include the following sub-steps S231-S232.

[0115] Sub-step S231: Use the second K-Means clustering model to determine whether the multidimensional feature parameters meet the preset second-class judgment conditions. The second-class judgment conditions include that the current temperature change trend is different from the historical temperature change trend of the same new energy vehicle, and the current voltage value is different from the historical voltage value of the same new energy vehicle, and the current temperature value is higher than each of the historical temperature values ​​of the same new energy vehicle, and the number of abnormal voltage jumps is multiple.

[0116] Sub-step S232: If yes, determine that the new energy vehicle is a faulty vehicle with abnormal connection of electronic components.

[0117] The system employs sub-steps S231 and S232 to perform a longitudinal comparative screening of the connection status of electronic components in the same new energy vehicle. A second K-Means clustering model is used to cluster the historical multidimensional feature parameters of the same vehicle under similar operating conditions, identifying a unique normal feature cluster for that vehicle. When the matching degree between the multidimensional feature parameters of the vehicle under test and the historical normal feature cluster is lower than a preset threshold, and simultaneously meets the following conditions: a significant difference between the temperature change trend and the historical trend; a deviation between the current voltage value and the historical average value exceeding a set value; a current temperature value higher than the temperature peak value in all historical records for that vehicle; and the number of abnormal voltage jumps reaching a preset threshold, the system combines the longitudinal comparison clustering results with preset anomaly judgment rules to generate new fault vehicle information, including the fault vehicle ID, abnormal feature parameter values, and the time period of the anomaly. This longitudinal comparison method effectively captures feature changes caused by the gradual deterioration of the connection status of electronic components in the same vehicle, complementing the horizontal comparative screening and further improving the accuracy and reliability of cloud-based diagnostics.

[0118] In the above scheme, the first K-Means clustering model and the second K-Means clustering model do not exist in isolation, but rather form a collaborative diagnostic mechanism of horizontal clustering screening and vertical clustering tracking. The first K-Means clustering model performs cluster analysis on the multi-dimensional feature parameters of a group of vehicles of the same type, dividing them into common normal feature clusters. This serves as a benchmark for horizontal comparison, quickly screening out faulty vehicles within the group whose features deviate from the common normal range. The second K-Means clustering model, on the other hand, constructs a vehicle-specific normal feature cluster based on the historical operating data of a single vehicle under similar operating conditions, enabling vertical tracking of feature changes in individual vehicles over long-term operation. The two models complement each other, capable of identifying sudden anomalies that are significantly different from other vehicles of the same type at the group level, and capturing slow feature changes caused by the gradual deterioration of electronic component connections at the individual level. This comprehensively covers different manifestations of electronic component connection anomalies in vehicles, providing comprehensive and accurate decision support for cloud-based diagnostics.

[0119] In some embodiments, Figure 5 or Figure 6 In order to achieve accurate vehicle early warning and advance maintenance scheduling based on any of the schemes, the method also includes the following step S25.

[0120] Step S25: Send the faulty vehicle information to the maintenance unit of the new energy vehicle so that the maintenance unit can perform fault inspection in advance.

[0121] Step S25 can be placed after step S22 or step S24. Step S25 enables accurate early warning and proactive repair scheduling for faulty vehicles. Based on the received faulty vehicle information, the repair unit can quickly retrieve the corresponding multi-dimensional characteristic parameters, fault judgment criteria, and historical operating data, and prepare suitable electronic components and professional repair tools in advance, effectively shortening the waiting time for repairs after the vehicle arrives at the shop and improving overall repair efficiency and user experience. Furthermore, after completing the repair, the repair unit can feed back information such as the actual fault cause, repair process, and results to the detection system. This is used to iteratively optimize the preset judgment conditions of the first and second K-Means clustering models, further enhancing the accuracy and robustness of fault detection and forming a closed-loop management mechanism of detection-early warning-repair-feedback-optimization.

[0122] In the above scheme, the cloud pushes the faulty vehicle to the maintenance unit or the vehicle terminal for early fault inspection and repair.

[0123] Figure 7 A flowchart illustrating an example of detecting abnormal connections of vehicle electronic components, provided for at least one embodiment of this disclosure. Figure 7 As shown, the method includes: 1) Establishment of basic data: For high-voltage component products, select products from the same batch to conduct charge and discharge tests with different connection reliability, and collect temperature data, voltage data and current data as basic datasets. Establish resistance-temperature correlation model and resistance compensation model through laboratory data; the vehicle end will transmit the real-time collected temperature data, voltage data and current data back to the cloud. 2) Vehicle-side calculation: The vehicle-side system determines whether there are abnormal jump values ​​in the high-voltage sampling. When the changes in the sampled voltage and battery voltage exceed the set threshold, the number of times is counted. In addition, the temperature and temperature rise rate at the high-voltage components are estimated based on the sampling temperature. The abnormal number of times is counted by comparing the normal contact resistance temperature and temperature rise. 3) Cloud computing: The cloud uses the K-Means clustering algorithm to screen abnormal vehicles and calculates the temperature change trend based on the vehicle's historical information. Finally, the screened abnormal vehicles are pushed to the maintenance unit or the vehicle end for early fault inspection.

[0124] Figure 8 This is a structural block diagram of a vehicle-side electronic component connection anomaly detection system provided for at least one embodiment of this disclosure. This system can be applied to new energy vehicles that communicate with cloud processors. Figure 8 As shown, the vehicle electronic component connection anomaly detection system 100 integrates a vehicle-side acquisition unit 101, a vehicle-side first-level processing unit 102, and a vehicle-side second-level processing unit 103.

[0125] The vehicle-side acquisition unit 101 (also known as the vehicle-side data acquisition unit) is configured to acquire the current multi-dimensional operating parameters of key nodes connecting electronic components in new energy vehicles.

[0126] The vehicle-side first-level processing unit 102 (also known as the vehicle-side data diagnostic unit) is configured to perform multi-parameter collaborative analysis based on multi-dimensional operating parameters, obtain multi-dimensional feature parameters used to characterize the connection status of electronic components, and upload the multi-dimensional feature parameters to the cloud processor.

[0127] The vehicle-side second-level processing unit 103 (also known as the vehicle-side fault reporting unit) is configured to receive feedback from the cloud processor, based on the multi-dimensional feature parameters combined with the historical multi-dimensional feature parameters of different vehicles of the same type as new energy vehicles under the current working conditions, and perform cluster analysis to predict the presence of faulty vehicles with abnormal electronic component connections. When the information of new energy vehicles and faulty vehicles is matched, warning information and maintenance information are pushed.

[0128] The specific execution methods of each unit in the above system embodiments have been described in detail in the embodiments related to the method, and will not be elaborated here.

[0129] In some embodiments, Figure 8 Based on the solution, the vehicle-side acquisition unit 101 can be implemented through corresponding sensors, and the vehicle-side first-level processing unit 102 and the vehicle-side second-level processing unit 103 can be implemented through a controller with corresponding programs.

[0130] In some embodiments, Figure 8 Based on this solution, the vehicle-side acquisition unit 101 includes a voltage sensor, a temperature sensor, and a current sensor (also known as a current detection unit). The voltage sensor is deployed at key nodes of high-voltage components, such as connector terminals or bolt connections, to measure real-time voltage data. The temperature sensor is deployed along relevant temperature conduction paths to monitor temperature. The current sensor is connected in series in the circuit to calculate the contact resistance using Ohm's law as shown in the following formula. :

[0131] In the formula, This indicates the voltage at the critical node (also known as the positive voltage detection point in the circuit) relative to the negative voltage of the battery. This indicates the current current value.

[0132] Figure 9 This is a structural block diagram of a cloud-based vehicle electronic component connectivity anomaly detection system provided in at least one embodiment of this disclosure. The system can be applied to systems communicating with cloud processors. New energy vehicles. For example... Figure 9As shown, the cloud-based vehicle electronic component connection anomaly detection system 200 integrates a cloud diagnostic unit 201 and a cloud transmission unit 202.

[0133] The cloud-based diagnostic unit 201 is configured to receive multi-dimensional feature parameters of key nodes sent by the new energy vehicle and then perform a horizontal comparison process. The horizontal comparison process is used to combine the multi-dimensional feature parameters with the historical multi-dimensional feature parameters of different vehicles of the same type as the new energy vehicle under the current operating conditions for cluster analysis to predict fault vehicle information with abnormal electronic component connections.

[0134] The cloud-based sending unit 202 is configured to send faulty vehicle information predicted by the horizontal comparison process to the new energy vehicles, so as to push early warning information and maintenance information when the information of the new energy vehicles matches that of the faulty vehicles.

[0135] The specific execution methods of each unit in the above system embodiments have been described in detail in the embodiments related to the method, and will not be elaborated here.

[0136] In some embodiments, Figure 9 Based on the solution, the cloud diagnostic unit 201 can be implemented through a controller with corresponding programs.

[0137] This disclosure also provides a storage medium storing a program or instructions that, when executed by a processor, implement the steps of the method embodiments described above.

[0138] This disclosure also provides a program product, such as... Figure 10 As shown, the program product includes one or more processors 301 and memory 302. Figure 10 Take processor 301 as an example.

[0139] The controller may also include an input device 303 and an output device 304.

[0140] The processor 301, memory 302, input device 303, and output device 304 can be connected via a bus or other means. Figure 10 Taking the example of a connection between China and Israel via a bus.

[0141] Processor 301 can be a central processing unit (CPU), or it can be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, or combinations of the above types of chips. The general-purpose processor can be a microprocessor or any conventional processor.

[0142] The memory 302, as a non-transitory computer-readable storage medium, can be used to store non-transitory software programs, non-transitory computer-executable programs, and modules, such as the program instructions / modules corresponding to the methods in the embodiments of this disclosure. The processor 301 executes various functional applications and data processing of the server by running the non-transitory software programs, instructions, and modules stored in the memory 302, thereby implementing the steps of the above-described method embodiments.

[0143] The memory 302 may include a program storage area and a data storage area. The program storage area may store the operating system and applications required for at least one function; the data storage area may store data created by the use of the processing device operated by the server. Furthermore, the memory 302 may include high-speed random access memory and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some embodiments, the memory 302 may optionally include memory remotely located relative to the processor 301, and these remote memories can be connected to a network connection device via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

[0144] Input device 303 can receive input digital or character information, and generate key signal inputs related to driver settings and function control of the server's processing unit. Output device 304 may include display devices such as a display screen.

[0145] One or more modules are stored in memory 302, and when executed by one or more processors 301, they perform actions such as... Figure 1 The method shown.

[0146] Those skilled in the art will understand that all or part of the processes in the above method embodiments can be implemented by a computer program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, it can include the processes described in the above method embodiments. The storage medium can be a magnetic disk, optical disk, read-only memory (ROM), random access memory (RAM), flash memory (FM), hard disk drive (HDD), or solid-state drive (SSD), etc.; the storage medium can also include combinations of the above types of memory.

[0147] Although embodiments of the present disclosure have been described in conjunction with the accompanying drawings, those skilled in the art can make various modifications and variations without departing from the spirit and scope of the present disclosure, and all such modifications and variations fall within the scope defined by the appended claims.

[0148] Although embodiments of the present disclosure have been shown and described above, it is to be understood that the above embodiments are exemplary and should not be construed as limiting the present disclosure. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of the present disclosure.

Claims

1. A method for detecting abnormal connections of vehicle electronic components, applied to new energy vehicles communicating with cloud processors, characterized in that, include: Obtain the current multi-dimensional operating parameters of key nodes connecting electronic components in new energy vehicles; Based on the multi-dimensional operating parameters, multi-parameter collaborative analysis is performed to obtain multi-dimensional feature parameters that characterize the connection status of electronic components, and the multi-dimensional feature parameters are uploaded to the cloud processor. as well as, After receiving information from the cloud processor about faulty vehicles with abnormal electronic component connections, based on the multi-dimensional feature parameters combined with historical multi-dimensional feature parameters of different vehicles of the same type as the new energy vehicle under the current operating conditions, and performing cluster analysis, the system pushes early warning information and maintenance information when the new energy vehicle matches the faulty vehicle information.

2. The method according to claim 1, characterized in that, The multidimensional operating parameters include voltage data, temperature data, and current data. Furthermore, the multi-parameter collaborative analysis based on these multidimensional operating parameters to obtain multidimensional feature parameters characterizing the connection status of electronic components includes: The contact resistance of the key node is generated based on the voltage data and the current data; Determine whether there are any abnormal voltage jumps in the voltage data within a set time period, and count the number of abnormal voltage jumps; Determine whether there are any temperature anomalies or abnormal temperature rise rates in the temperature data within a set time period, and statistically analyze the abnormal temperature rise data for those with abnormal temperature or abnormal temperature rise rates; and, Generate multidimensional feature parameters that include the number of voltage anomaly jumps and the temperature rise anomaly data.

3. The method according to claim 2, characterized in that, The determination of whether the temperature data within the set time period is abnormal or whether the temperature rise rate is abnormal includes: The temperature rise rate data of the key node is generated based on the temperature data; The contact resistance data of the key node is generated based on the voltage data and the current data; Based on the contact resistance data and temperature rise rate data within a set time period, combined with a preset temperature compensation model, it is determined whether the temperature data shows an abnormal temperature rise rate due to increased contact resistance. The temperature compensation model includes different temperature rises caused by different contact stresses. Based on the contact resistance data and the temperature data, a preset resistance-temperature correlation model is used to determine whether there is a temperature anomaly caused by an increase in contact resistance. The resistance-temperature correlation model includes the relationship between the contact resistance data and the temperature rise rate data when there are no abnormalities in voltage jumps and temperature rise.

4. The method according to claim 2 or 3, characterized in that, The key nodes include at least one of the connector terminals and bolt connections, and the multi-dimensional operating parameters are collected synchronously; furthermore... The number of abnormal voltage jumps is counted by determining whether the difference between the voltage data and the battery voltage exceeds a set threshold. The abnormal temperature rise data is statistically analyzed by determining whether the temperature data and its temperature rise rate data exceed their respective dynamic adaptive thresholds. The dynamic adaptive thresholds are related to the current operating conditions of the new energy vehicle and change with the current operating conditions.

5. A method for detecting abnormal connections of vehicle electronic components, applied to a cloud processor communicating with new energy vehicles, characterized in that, include: After receiving the multi-dimensional feature parameters of key nodes sent by the new energy vehicle, a horizontal comparison process is executed. The horizontal comparison process is used to combine the multi-dimensional feature parameters with the historical multi-dimensional feature parameters of different vehicles of the same type as the new energy vehicle under the current working conditions for cluster analysis to predict the fault vehicle information of abnormal electronic component connection. as well as, The faulty vehicle information predicted by the horizontal comparison process is sent to the new energy vehicle so that when the new energy vehicle matches the faulty vehicle information, early warning information and maintenance information are pushed.

6. The method according to claim 5, characterized in that, Also includes: The longitudinal comparison process is executed, wherein the longitudinal comparison process is used to predict the change trend of key parameters based on the multidimensional feature parameters combined with the historical multidimensional feature parameters of the same new energy vehicle, and to predict whether the new energy vehicle will become the faulty vehicle within a set time period based on the change trend of the key parameters, and to generate new faulty vehicle information when the new energy vehicle becomes the faulty vehicle within the set time period. as well as, The faulty vehicle information predicted by the longitudinal comparison process is sent to the new energy vehicle so that when the new energy vehicle matches the faulty vehicle information, early warning information and maintenance information are pushed.

7. The method according to claim 6, characterized in that, The multidimensional feature parameters include temperature value, voltage value, number of abnormal voltage jumps, and abnormal temperature rise data; and... The horizontal comparison process includes: The first K-Means clustering model is used to determine whether the multidimensional feature parameters meet a preset first-type judgment condition, wherein the first-type judgment condition includes that the current temperature value is higher than the temperature values ​​of different vehicles of the same type and that the number of abnormal voltage jumps is multiple; and... If so, information on whether the new energy vehicle is a faulty vehicle with abnormal connection of electronic components; The longitudinal comparison process includes: The second K-Means clustering model is used to determine whether the multidimensional feature parameters meet the preset second type of judgment conditions. The second type of judgment conditions include that the current temperature change trend is different from the historical temperature change trend of the same new energy vehicle, and the current voltage value is different from the historical voltage value of the same new energy vehicle, and the current temperature value is higher than each of the historical temperature values ​​of the same new energy vehicle, and the number of abnormal voltage jumps is multiple. If so, the information indicates that the new energy vehicle is a faulty vehicle with abnormal connection of electronic components.

8. A vehicle electronic component connection anomaly detection system, applied to new energy vehicles communicating with cloud processors, characterized in that, include: The vehicle-side acquisition unit is configured to acquire the current multi-dimensional operating parameters of key nodes connecting electronic components in new energy vehicles. The vehicle-side first-level processing unit is configured to perform multi-parameter collaborative analysis based on the multi-dimensional operating parameters, obtain multi-dimensional feature parameters for characterizing the connection status of electronic components, and upload the multi-dimensional feature parameters to the cloud processor. and, The vehicle-side second-level processing unit is configured to receive feedback from the cloud processor, based on the multi-dimensional feature parameters combined with the historical multi-dimensional feature parameters of different vehicles of the same type as the new energy vehicle under the current operating conditions, and perform cluster analysis to predict the presence of faulty vehicles with abnormal electronic component connections. When the new energy vehicle matches the faulty vehicle information, it pushes early warning information and maintenance information.

9. A vehicle electronic component connection anomaly detection system, applied to a cloud processor communicating with new energy vehicles, characterized in that, include: The cloud-based diagnostic unit is configured to receive multi-dimensional feature parameters of key nodes sent by the new energy vehicle and then perform a horizontal comparison process. The horizontal comparison process is used to combine the multi-dimensional feature parameters with the historical multi-dimensional feature parameters of different vehicles of the same type as the new energy vehicle under the current operating conditions for cluster analysis to predict fault vehicle information with abnormal electronic component connections. and, The cloud-based sending unit is configured to send the faulty vehicle information predicted by the horizontal comparison process to the new energy vehicle, so as to push early warning information and maintenance information when the new energy vehicle matches the faulty vehicle information.

10. A storage medium, characterized in that, The storage medium stores a program or instructions, wherein the program or instructions, when executed by a processor, implement the steps of the method as described in any one of claims 1 to 7.

Images