Power failure monitoring and power consumption optimization method and system applied to vehicle-mounted terminal

Abstract

The invention discloses a power failure monitoring and power utilization optimization method and system applied to a vehicle-mounted terminal, and relates to the technical field of power utilization optimization, and the method comprises the steps: obtaining real-time electric quantity information as power failure tags, carrying out the preprocessing and normalization of vehicle data, obtaining vehicle features, dividing the power failure tags into to-be-confirmed fault power failure tags and non-fault power failure tags, and carrying out the verification of the fault power failure tags and the non-fault power failure tags. The method comprises the steps of distinguishing a power-down speed undetermined normal interval and a power-down speed undetermined abnormal interval, carrying out secondary distinguishing on the undetermined normal interval and the power-down speed undetermined abnormal interval, monitoring a power-down label and an abnormal label in real time when a vehicle runs in real time, and finally judging that power utilization optimization of a user is completed. According to the method, secondary interval distinguishing is completed by introducing feature deviation degree analysis, the technical limitation that an intersection area cannot be defined in traditional interval judgment is broken through, power utilization optimization is changed from passive response to active response, and effectiveness and pertinence of optimization operation are improved.

Description

Technical Field

[0001] This invention relates to the field of power consumption optimization technology, specifically to a method and system for power failure monitoring and power consumption optimization applied to vehicle-mounted terminals. Background Technology

[0002] Current vehicle-mounted terminal power loss monitoring solutions mostly adopt a threshold comparison mode based on a single power parameter, relying solely on a fixed rate of power loss to determine the battery's operating status, without establishing a correlation analysis logic between power information and the actual operating characteristics of the vehicle. Such solutions neither classify and filter power loss tags based on fault correlation nor differentiate power loss attributes based on vehicle operating conditions. They often misjudge reasonable power loss under normal operating conditions as fault anomalies, and may also miss abnormal power loss caused by latent faults. A large number of meaningless warning messages not only interfere with the normal operation of vehicle-mounted terminals but also reduce user trust in warning prompts, failing to provide effective data support for battery fault diagnosis and failing to meet the actual needs of accurate monitoring. Existing technologies, after identifying abnormal power loss, mostly only provide warnings at the phenomenological level, without delving into the underlying vehicle characteristics that cause the anomaly. When there are ambiguous judgment scenarios with overlapping power loss rate ranges, traditional solutions, lacking secondary differentiation techniques, cannot accurately define the attributes of power loss data within the range, let alone pinpoint the core characteristics causing the anomaly. This forces users to resort to haphazard troubleshooting when optimizing battery usage, making it difficult to quickly pinpoint the root cause of anomalies or effectively intervene in key influencing factors. This not only wastes significant time and effort but also severely diminishes the effectiveness of power optimization, failing to fundamentally solve the problem of abnormal battery power loss. Traditional vehicle power loss monitoring systems often use static storage databases without a regular data update mechanism, making them unable to adapt to parameter changes caused by dynamic factors such as battery aging, changes in operating conditions, and altered usage environments. Over time, the monitoring model may become out of sync with actual operating conditions, leading to the gradual invalidation of monitoring judgments. Furthermore, existing solutions lack a post-warning effectiveness verification process. They can only issue alerts when an anomaly occurs but cannot reassess the battery status after user optimization operations. This makes it difficult to form a complete closed loop from anomaly monitoring and early warning handling to effectiveness verification. It fails to guarantee a complete resolution of short-term anomalies and is detrimental to long-term battery stability maintenance, making it difficult to meet the monitoring needs of the entire lifecycle of the vehicle power system. Summary of the Invention

[0003] The purpose of this invention is to provide a method and system for power failure monitoring and power consumption optimization applied to vehicle terminals, so as to solve the problems mentioned in the background art.

[0004] To address the aforementioned technical problems, this invention provides the following technical solution: a method for power failure monitoring and power consumption optimization applied to vehicle-mounted terminals, comprising the following steps: S1. Obtain real-time battery information as a power loss tag, preprocess and normalize vehicle data to obtain vehicle features, and establish a vehicle feature database based on the vehicle features. S2. For power failure tags, divide power failure tags into power failure tags for pending faults and power failure tags for non-faults; S3. Analyze the power outage tags for unconfirmed faults and non-fault power outage tags, and distinguish between the normal range of undetermined power outage speed and the abnormal range of undetermined power outage speed. S4. Based on the relationship between the undetermined normal range and the undetermined abnormal range of power loss speed, a second distinction is made between the undetermined normal range and the undetermined abnormal range of power loss speed. For power loss tags whose power loss speed is in the abnormal range of power loss speed, the power loss tag is marked as an abnormal tag. S5. During real-time vehicle operation, monitor the power failure tags of the vehicle, prompt abnormal tags, and finally determine that the user has completed the power optimization. S6. Update the historical database based on the preset update cycle.

[0005] Furthermore, in step S1, after authorization, real-time battery information is acquired. This battery information includes vehicle battery level and battery depletion rate information. The battery depletion rate is the ratio of the difference between the vehicle's battery level before a power outage monitoring cycle and its current battery level to the vehicle's total battery level. The power outage monitoring cycle is a preset time length. The battery information is used as a power outage tag for the monitoring cycle. Vehicle data is collected within the monitoring cycle, including internal vehicle operation data and environmental data. The vehicle data is preprocessed and normalized to obtain a set of vehicle features. A vehicle feature database is established based on these features. In this database, vehicle features are classified based on the tags from the power outage monitoring cycle. By collecting multi-dimensional battery information and vehicle data after authorization, a correlation is established between battery features and vehicle operation and environmental features, overcoming the limitations of traditional monitoring that relies solely on a single battery parameter. The standardized vehicle features obtained through preprocessing and normalization eliminate interference caused by differences in data units, laying a solid data foundation for subsequent tag classification and anomaly analysis. At the same time, the feature is classified and database is built according to the power failure label, realizing the structured archiving of data. This avoids the problem of inefficient subsequent retrieval caused by messy data accumulation, and greatly improves the utilization value of vehicle feature data and the convenience of subsequent analysis.

[0006] Furthermore, in step S2, for any type of power failure tag, there are M groups of vehicle features corresponding to the power failure tag. The m-th group of vehicle features is analyzed. If the vehicle is judged to have a battery fault within a preset fault cycle after the m-th group of vehicle features is collected, the m-th group of vehicle features is judged to be a fault manifestation feature group; otherwise, the m-th group of vehicle features is judged not to be a fault manifestation feature group. Substituting m=1,2,…,M one by one, the number of fault manifestation feature groups is M1, and then the battery failure rate A=M1 / M of the power failure tag is obtained. If A is greater than A0, the power failure tag is judged to be a power failure tag to be confirmed, where A0 is a preset battery failure rate threshold; otherwise, the power failure tag is judged not to be a non-fault power failure tag, and then the power failure tags are classified into power failure tags to be confirmed and non-fault power failure tags. The association judgment logic between power failure tags, vehicle features and battery faults is constructed. By analyzing the battery fault situation within a specific cycle after the feature group is collected, the fault manifestation feature group and the non-fault feature group are accurately distinguished. The initial classification of power outage tags is completed based on the failure rate associated with feature groups, effectively filtering out power outage tags that need to be confirmed as faults and excluding non-fault tags, thus eliminating invalid data interference from the source. This classification method lays an accurate data foundation for subsequent power outage rate range analysis, avoids subsequent judgment bias caused by mixed tag attributes, and significantly improves the quality of early data processing in power outage anomaly monitoring.

[0007] In step S3, the power loss tags are analyzed, and the battery level information in the tags is retrieved. For each battery level information, the corresponding power loss tags are retrieved. These tags with the same battery level information include X non-faulty power loss tags and Y unconfirmed faulty power loss tags. The vehicle battery level of the X non-faulty power loss tags decreases at a rate of {A1, A2, ..., A...}. x ,…,A X}, where A x Let X represent the vehicle battery rate of the xth non-fault power loss tag. Remove impurities from the vehicle battery rate of the X non-fault power loss tags to obtain the undetermined normal range of power loss rate. The upper limit of the undetermined normal range of power loss rate is the maximum value of the vehicle battery rate of the X non-fault power loss tags after removing impurities, and the lower limit is the minimum value of the vehicle battery rate of the X non-fault power loss tags after removing impurities. The battery level of the vehicles with Y pending fault power loss tags decreases at a rate of {B1, B2, ..., B}. y ,…,B Y}, where B yLet represent the vehicle battery depletion rate of the y-th pending fault power loss tag. After removing nodules from the vehicle battery depletion rates of the X pending fault power loss tags, an undetermined abnormal range for power loss rate is obtained. The upper limit of this undetermined abnormal range is the maximum value of the vehicle battery depletion rate of the X pending fault power loss tags after nodules are removed, and the lower limit is the minimum value of the vehicle battery depletion rate of the X pending fault power loss tags after nodules are removed. Therefore, for any given battery level information, a normal undetermined range and an abnormal undetermined range for power loss rate are obtained. For any given battery level information, if the undetermined normal range for power loss rate is... If there is no overlap between the interval and the undetermined abnormal interval of power loss rate, the undetermined normal interval of power loss rate is determined as the normal interval of power loss rate, and the undetermined abnormal interval of power loss rate is determined as the abnormal interval of power loss rate. Otherwise, a secondary analysis is performed on the undetermined interval, which is the intersection of the undetermined normal interval and the undetermined abnormal interval of power loss rate. By classifying power loss tags according to the same power information, and distinguishing between non-fault and unconfirmed fault tags, power loss rate noise removal is performed to accurately construct the undetermined normal and abnormal intervals, effectively avoiding the interval definition deviation caused by data mixing under different power scenarios. The separate processing for intervals with and without overlap enables rapid and accurate determination of non-overlapping scenarios, while reserving space for secondary analysis of fuzzy scenarios with overlap, significantly improving the scientificity and reliability of power loss rate interval division, and providing a solid foundation for accurate interval determination of subsequent abnormal tags.

[0008] Furthermore, in step S4, when there is an intersection between the normal range and the abnormal range of the undetermined power loss rate, a secondary analysis is performed on the undetermined range. Within the undetermined range, for the vehicle battery depletion rate information α, N sets of vehicle features from the power loss tag are retrieved, where the nth set of vehicle features is {C}. 1_n C 2_n ,…,C q_n ,…,C Q_n}, where C q_n Let Q represent the q-th vehicle feature in the n-th vehicle feature group, where Q represents the number of vehicle features in the group. Substituting each value into n=1,2,…,N, we obtain the q-th vehicle feature in the N-th vehicle feature group. The q-th vehicle feature in the N-th vehicle feature group is {C}. q_1 C q_2 ,…,C q_n ,…,C q_N}, and then obtain the feature deviation D of the q-th vehicle feature when considering the vehicle's battery depletion rate information. q : ; Where E q E represents the q-th standard vehicle feature when considering the vehicle's battery depletion rate information. q Let D be the average value of the q-th vehicle feature in all feature groups within the normal interval.q If the deviation of the qth vehicle feature exceeds the preset deviation threshold, the qth vehicle feature is determined to be abnormal and is considered an abnormal vehicle feature. Otherwise, the qth vehicle feature is determined to be normal. Then, the vehicle battery depletion rate information α is assessed. If an abnormal vehicle feature exists, the vehicle battery depletion rate information α is determined to be within the abnormal power loss rate range; otherwise, the vehicle battery depletion rate information α is determined to be within the normal power loss rate range. This process is repeated for all vehicle battery depletion rate information within the specified range, thus distinguishing between the abnormal and normal power loss rate ranges. After confirming the abnormal power loss rate range, any power loss label within this range is marked as an abnormal label. If the qth vehicle feature is an abnormal vehicle feature, it is marked as a key observation feature. For fuzzy judgment scenarios involving the intersection of power loss rate ranges, feature deviation analysis is innovatively introduced to complete a secondary distinction, overcoming the technical limitation of traditional range judgments that cannot define the attributes of intersecting data. This step relies on the average value of standard features to construct a judgment benchmark. By accurately identifying the characteristics of abnormal vehicles through the degree of feature deviation, it not only achieves precise division of the power loss speed attribute within the undetermined range, but also simultaneously identifies and marks the core features that cause the anomaly as key observation features. This design not only completes the precise definition of anomaly labels, but also provides clear problem directions for subsequent anomaly warnings, avoiding the drawback of merely identifying anomalies without being able to pinpoint the root cause. It significantly improves the accuracy and guidance of power loss anomaly monitoring, laying a solid foundation for users to optimize their subsequent power usage.

[0009] Furthermore, in step S5, during real-time vehicle operation, the power failure tag is monitored. If the power failure tag is an abnormal tag, the user is prompted to check the vehicle, and key observation features are highlighted. If the vehicle characteristics change within a preset reminder period after the reminder, a new power failure tag is generated. If the power failure tag is again determined to be an abnormal tag, the user is prompted to check the vehicle again; otherwise, it is determined that the user has completed power optimization. This establishes a closed loop for real-time monitoring and optimization effect verification of vehicle terminal power failure anomalies, achieving timely response to abnormal situations and accurate judgment of handling effects. When an abnormal tag is detected, this step not only pushes a vehicle inspection reminder to the user but also simultaneously informs them of key observation features, providing clear guidance for the user to troubleshoot problems and avoiding the drawback of traditional early warnings that only indicate abnormalities without specific troubleshooting directions. Meanwhile, by monitoring feature changes and re-judging tags within the alert period, the actual effectiveness of users' electricity optimization operations can be verified. This not only prevents repeated anomalies caused by incomplete optimization, but also allows for timely confirmation of completion status after optimization targets are met, forming a complete link from anomaly warning to handling verification. This significantly improves the effectiveness of users' electricity optimization and the closed-loop nature of the monitoring process.

[0010] Furthermore, in step S6, after a preset update cycle, the historical data in the database is updated.

[0011] A power failure monitoring and power consumption optimization system for vehicle terminals includes: a vehicle feature database construction module, a power failure tag initial classification module, a power failure interval initial delineation module, a power failure tag re-labeling module, a real-time anomaly early warning optimization module, and a historical database update module. The vehicle feature database construction module is used to obtain real-time power information as a power loss tag, preprocess and normalize vehicle data to obtain vehicle features, and establish a vehicle feature database based on the vehicle features. The power failure tag initial classification module classifies power failure tags into power failure tags with pending faults and non-fault power failure tags. The power outage interval initial delineation module is used to analyze the power outage tags to be confirmed and the non-fault power outage tags, and to distinguish between the normal range of undetermined power outage speed and the abnormal range of undetermined power outage speed. The power failure tag relabeling module is used to perform a second distinction between the undetermined normal range and the undetermined abnormal range of power failure speed based on the relationship between the undetermined normal range and the undetermined abnormal range of power failure speed. For power failure tags whose power failure speed is in the abnormal range of power failure speed, the power failure tag is labeled as an abnormal tag. The real-time anomaly warning and optimization module is used to monitor the power loss tag of the vehicle in real time during vehicle operation, prompt the anomaly tag, and finally determine that the user has completed the power optimization. The historical database update module is used to update the historical database based on a preset update cycle.

[0012] Compared with existing technologies, the beneficial effects achieved by this invention are as follows: On the one hand, by using a hierarchical power loss tag classification logic, it first completes the preliminary classification of tags by combining battery fault association features, and then delineates the initial interval based on the power loss rate distribution under the same power information, thus avoiding the misjudgment problem caused by single-dimensional judgment in traditional monitoring from the source. It abandons the crude threshold comparison mode and relies on the correlation analysis between vehicle features and power loss tags to achieve accurate definition of tag attributes. It can effectively screen out tags related to the fault to be confirmed, and also filter out normal power loss data that is not faulty. This greatly reduces the interference of meaningless warnings on the operation of vehicle terminals and user use, making the identification results of abnormal power loss more consistent with the actual operating state of the vehicle battery, and laying a solid data judgment foundation for subsequent fault investigation.

[0013] On the one hand, for ambiguous scenarios where power outage speed ranges overlap, the method innovatively introduces feature deviation analysis to perform secondary segmentation of the ranges, overcoming the technical limitation of traditional range determination that the overlapping area cannot be defined. While completing anomaly labeling, it can also simultaneously identify key observation features that trigger power outage anomalies, moving beyond simply issuing warnings of abnormal phenomena to delving into the root cause at the feature level. This design guides users to focus on core anomaly factors to optimize power usage, avoiding the time and effort wasted on blind investigations, and shifting power optimization from passive response to proactive and precise intervention, significantly improving the effectiveness and relevance of optimization operations.

[0014] On the other hand, relying on a pre-set database update mechanism, continuous iteration of vehicle characteristics and power loss tag data is achieved. This allows the monitoring model to dynamically adapt to factors such as battery aging and changes in operating conditions, breaking the long-term judgment failure problem caused by data stagnation in traditional monitoring solutions. Simultaneously, the real-time monitoring and post-warning review process forms a complete closed loop. This not only promptly alerts users to investigate when anomalies occur but also verifies the optimization effect after users complete their operations, ensuring a traceable and complete power optimization chain. This closed-loop design not only guarantees timely handling of short-term power loss anomalies but also maintains the long-term stability of battery operation, reduces the probability of battery failure, and extends the overall lifespan of the vehicle power system. Attached Figure Description

[0015] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings: Figure 1 This is a structural diagram of a power failure monitoring and power consumption optimization system for vehicle-mounted terminals according to the present invention; Figure 2 This is a flowchart of a power failure monitoring and power consumption optimization method applied to an in-vehicle terminal according to the present invention. Detailed Implementation

[0016] 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.

[0017] Please see Figure 1 and Figure 2 The present invention provides a technical solution: a method for power failure monitoring and power consumption optimization applied to vehicle-mounted terminals, comprising the following steps: S1. Obtain real-time battery information as a power loss tag, preprocess and normalize vehicle data to obtain vehicle features, and establish a vehicle feature database based on the vehicle features. S2. For power failure tags, divide power failure tags into power failure tags for pending faults and power failure tags for non-faults; S3. Analyze the power outage tags for unconfirmed faults and non-fault power outage tags, and distinguish between the normal range of undetermined power outage speed and the abnormal range of undetermined power outage speed. S4. Based on the relationship between the undetermined normal range and the undetermined abnormal range of power loss speed, a second distinction is made between the undetermined normal range and the undetermined abnormal range of power loss speed. For power loss tags whose power loss speed is in the abnormal range of power loss speed, the power loss tag is marked as an abnormal tag. S5. During real-time vehicle operation, monitor the power failure tags of the vehicle, prompt abnormal tags, and finally determine that the user has completed the power optimization. S6. Update the historical database based on the preset update cycle.

[0018] In step S1, after authorization, real-time battery information is acquired. This information includes vehicle battery level and battery depletion rate. The battery depletion rate is the ratio of the difference between the vehicle's battery level before a power outage monitoring cycle and its current battery level to the vehicle's total battery level. The power outage monitoring cycle is a preset time length. The battery information is used as a power outage tag for the monitoring cycle. Vehicle data is collected within the monitoring cycle, including internal vehicle operation data and environmental data. The vehicle data is preprocessed and normalized to obtain a set of vehicle features. A vehicle feature database is established based on these features. Within the database, vehicle features are classified according to the tags of the power outage monitoring cycle. By collecting multi-dimensional battery information and vehicle data after authorization, a correlation is established between battery features and vehicle operation and environmental features, overcoming the limitations of traditional monitoring that relies solely on a single battery parameter. The standardized vehicle features obtained through preprocessing and normalization eliminate interference caused by differences in data units, laying a solid data foundation for subsequent tag classification and anomaly analysis. At the same time, the feature is classified and database is built according to the power failure label, realizing the structured archiving of data. This avoids the problem of inefficient subsequent retrieval caused by messy data accumulation, and greatly improves the utilization value of vehicle feature data and the convenience of subsequent analysis.

[0019] In step S2, for any type of power failure tag, there are M groups of vehicle features corresponding to the power failure tag. The m-th group of vehicle features is analyzed. If the vehicle is judged to have a battery fault within a preset fault cycle after the m-th group of vehicle features is collected, the m-th group of vehicle features is judged to be a fault manifestation feature group; otherwise, the m-th group of vehicle features is judged not to be a fault manifestation feature group. Substituting m=1,2,…,M one by one, the number of fault manifestation feature groups is M1, and then the battery failure rate A=M1 / M of the power failure tag is obtained. If A is greater than A0, the power failure tag is judged to be a power failure tag to be confirmed, where A0 is a preset battery failure rate threshold; otherwise, the power failure tag is judged not to be a non-fault power failure tag, and then the power failure tags are classified into power failure tags to be confirmed and non-fault power failure tags. The association judgment logic between power failure tags, vehicle features and battery faults is constructed. By analyzing the battery fault situation within a specific cycle after the feature group is collected, the fault manifestation feature group and the non-fault feature group are accurately distinguished. The initial classification of power outage tags is completed based on the failure rate associated with feature groups, effectively filtering out power outage tags that need to be confirmed as faults and excluding non-fault tags, thus eliminating invalid data interference from the source. This classification method lays an accurate data foundation for subsequent power outage rate range analysis, avoids subsequent judgment bias caused by mixed tag attributes, and significantly improves the quality of early data processing in power outage anomaly monitoring.

[0020] In step S3, the power loss tags are analyzed, and the battery level information in the tags is retrieved. For each battery level information, the corresponding power loss tags are retrieved. These tags with the same battery level information include X non-faulty power loss tags and Y unconfirmed faulty power loss tags. The vehicle battery level of the X non-faulty power loss tags decreases at a rate of {A1, A2, ..., A...}. x ,…,A X}, where A x Let X represent the vehicle battery rate of the xth non-fault power loss tag. Remove impurities from the vehicle battery rate of the X non-fault power loss tags to obtain the undetermined normal range of power loss rate. The upper limit of the undetermined normal range of power loss rate is the maximum value of the vehicle battery rate of the X non-fault power loss tags after removing impurities, and the lower limit is the minimum value of the vehicle battery rate of the X non-fault power loss tags after removing impurities. The battery level of the vehicles with Y pending fault power loss tags decreases at a rate of {B1, B2, ..., B}. y ,…,B Y}, where B yLet represent the vehicle battery depletion rate of the y-th pending fault power loss tag. After removing nodules from the vehicle battery depletion rates of the X pending fault power loss tags, an undetermined abnormal range for power loss rate is obtained. The upper limit of this undetermined abnormal range is the maximum value of the vehicle battery depletion rate of the X pending fault power loss tags after nodules are removed, and the lower limit is the minimum value of the vehicle battery depletion rate of the X pending fault power loss tags after nodules are removed. Therefore, for any given battery level information, a normal undetermined range and an abnormal undetermined range for power loss rate are obtained. For any given battery level information, if the undetermined normal range for power loss rate is... If there is no overlap between the interval and the undetermined abnormal interval of power loss rate, the undetermined normal interval of power loss rate is determined as the normal interval of power loss rate, and the undetermined abnormal interval of power loss rate is determined as the abnormal interval of power loss rate. Otherwise, a secondary analysis is performed on the undetermined interval, which is the intersection of the undetermined normal interval and the undetermined abnormal interval of power loss rate. By classifying power loss tags according to the same power information, and distinguishing between non-fault and unconfirmed fault tags, power loss rate noise removal is performed to accurately construct the undetermined normal and abnormal intervals, effectively avoiding the interval definition deviation caused by data mixing under different power scenarios. The separate processing for intervals with and without overlap enables rapid and accurate determination of non-overlapping scenarios, while reserving space for secondary analysis of fuzzy scenarios with overlap, significantly improving the scientificity and reliability of power loss rate interval division, and providing a solid foundation for accurate interval determination of subsequent abnormal tags.

[0021] In step S4, when there is an intersection between the normal interval and the abnormal interval of the undetermined power loss rate, a secondary analysis is performed on the interval to be determined. In the undetermined interval, for the vehicle battery depletion rate information α, N sets of vehicle features of the power loss tag are called, where the nth set of vehicle features is {C 1_n C 2_n ,…,C q_n ,…,C Q_n}, where C q_n Let Q represent the q-th vehicle feature in the n-th vehicle feature group, where Q represents the number of vehicle features in the group. Substituting each value into n=1,2,…,N, we obtain the q-th vehicle feature in the N-th vehicle feature group. The q-th vehicle feature in the N-th vehicle feature group is {C}. q_1 C q_2 ,…,C q_n ,…,C q_N}, and then obtain the feature deviation D of the q-th vehicle feature when considering the vehicle's battery depletion rate information. q : ; Where E q E represents the q-th standard vehicle feature when considering the vehicle's battery depletion rate information. q Let D be the average value of the q-th vehicle feature in all feature groups within the normal interval.q If the deviation of the qth vehicle feature exceeds the preset deviation threshold, the qth vehicle feature is determined to be abnormal and is considered an abnormal vehicle feature. Otherwise, the qth vehicle feature is determined to be normal. Then, the vehicle battery depletion rate information α is assessed. If an abnormal vehicle feature exists, the vehicle battery depletion rate information α is determined to be within the abnormal power loss rate range; otherwise, the vehicle battery depletion rate information α is determined to be within the normal power loss rate range. This process is repeated for all vehicle battery depletion rate information within the specified range, thus distinguishing between the abnormal and normal power loss rate ranges. After confirming the abnormal power loss rate range, any power loss label within this range is marked as an abnormal label. If the qth vehicle feature is an abnormal vehicle feature, it is marked as a key observation feature. For fuzzy judgment scenarios involving the intersection of power loss rate ranges, feature deviation analysis is innovatively introduced to complete a secondary distinction, overcoming the technical limitation of traditional range judgments that cannot define the attributes of intersecting data. This step relies on the average value of standard features to construct a judgment benchmark. By accurately identifying the characteristics of abnormal vehicles through the degree of feature deviation, it not only achieves precise division of the power loss speed attribute within the undetermined range, but also simultaneously identifies and marks the core features that cause the anomaly as key observation features. This design not only completes the precise definition of anomaly labels, but also provides clear problem directions for subsequent anomaly warnings, avoiding the drawback of merely identifying anomalies without being able to pinpoint the root cause. It significantly improves the accuracy and guidance of power loss anomaly monitoring, laying a solid foundation for users to optimize their subsequent power usage.

[0022] In step S5, during real-time vehicle operation, the power failure tags are monitored. If a power failure tag is an abnormal tag, the user is prompted to check the vehicle, and key observation features are highlighted. If the vehicle characteristics change within a preset notification period after the notification, a new power failure tag is generated. If the power failure tag is again determined to be an abnormal tag, the user is prompted to check the vehicle again; otherwise, it is determined that the user has completed power optimization. This establishes a closed loop for real-time monitoring and optimization effect verification of vehicle terminal power failure anomalies, achieving timely response to anomalies and accurate judgment of handling effects. When an abnormal tag is detected, this step not only pushes a vehicle inspection prompt to the user but also simultaneously informs them of key observation features, providing clear guidance for the user to troubleshoot the problem and avoiding the drawback of traditional early warnings that only indicate anomalies without specific troubleshooting directions. Meanwhile, by monitoring feature changes and re-judging tags within the alert period, the actual effectiveness of users' electricity optimization operations can be verified. This not only prevents repeated anomalies caused by incomplete optimization, but also allows for timely confirmation of completion status after optimization targets are met, forming a complete link from anomaly warning to handling verification. This significantly improves the effectiveness of users' electricity optimization and the closed-loop nature of the monitoring process.

[0023] In step S6, after a preset update cycle, the historical data in the database is updated.

[0024] A power failure monitoring and power consumption optimization system for vehicle terminals, the system comprising: a vehicle feature database construction module, a power failure tag initial classification module, a power failure interval initial delineation module, a power failure tag re-labeling module, a real-time anomaly early warning optimization module, and a historical database update module; The vehicle feature database construction module is used to obtain real-time battery information as a power loss tag, preprocess and normalize vehicle data to obtain vehicle features, and build a vehicle feature database based on the vehicle features. The initial classification module for power failure tags categorizes them into power failure tags with pending faults and non-fault power failure tags. The power outage range initial delineation module is used to analyze the power outage tags of the fault to be confirmed and the non-fault power outage tags, and to distinguish between the normal range of power outage speed to be determined and the abnormal range of power outage speed to be determined. The power failure tag relabeling module is used to further distinguish between the undetermined normal range and the undetermined abnormal range of power failure speed based on the relationship between the undetermined normal range and the undetermined abnormal range of power failure speed. For power failure tags whose power failure speed is in the abnormal range of power failure speed, the power failure tag is labeled as an abnormal tag. The real-time anomaly warning and optimization module is used to monitor the power loss tags of the vehicle in real time during vehicle operation, prompt the anomaly tags, and finally determine that the user has completed the power optimization. The historical database update module is used to update the historical database based on a preset update cycle.

[0025] Example 1: In step S1, the vehicle terminal first sends a data collection authorization request to the vehicle's central control system. After the vehicle owner confirms the authorization, the power collection module is activated to obtain real-time power information. This information includes the vehicle's remaining power status and the rate of power loss. The rate of power loss is determined based on the ratio of power change to total power within a preset time period, and the power information for this period is encapsulated as a power loss tag. Simultaneously, the terminal collects vehicle internal operating data within the period, including real-time battery temperature, vehicle electrical operating power, motor load status, etc., as well as vehicle environmental data, including ambient temperature, road bumpiness, altitude, etc. The collected multi-source data is preprocessed to remove abnormal noise data, and then normalization is performed to eliminate the dimensional differences between different data to form standardized vehicle features. Finally, the features are associated with the corresponding power loss tags, stored in the constructed vehicle feature database, and classified and archived according to the tags.

[0026] In step S2, for a certain type of power failure tag in the database, all corresponding vehicle feature groups are retrieved and analyzed in conjunction with the fault records of the on-board diagnostic system: if a vehicle is diagnosed with a battery fault within a preset period after a certain feature group is collected, it is marked as a fault manifestation feature group; after verification one by one, the proportion of fault manifestation feature groups is counted, and based on the comparison result of this proportion with the preset benchmark, the power failure tags are divided into two categories: power failure tags to be confirmed faults and non-fault power failure tags.

[0027] In step S3, the system categorizes power loss tags according to the same power information, extracts the power loss rate data of non-fault and unconfirmed fault tags, removes extreme outliers to complete the noise removal process, and constructs undetermined normal range and undetermined abnormal range for power loss rate. If the two ranges do not overlap, they are directly identified as the final normal and abnormal ranges. If there is an overlap, the overlapping part is designated as an undetermined range for secondary analysis.

[0028] For the undetermined interval, the secondary analysis process is initiated in step S4: extract the vehicle feature groups corresponding to the battery depletion rate within the interval, compare each feature group with the average level of the same type of feature in the normal interval, and determine whether the feature is abnormal by the degree of feature deviation; if there are abnormal features, the speed data is determined to belong to the abnormal interval, and the abnormal features are marked as key observation features, and finally the abnormal interval is accurately defined and the corresponding abnormal label is marked.

[0029] In step S5, the terminal monitors and generates a power failure tag in real time during vehicle operation. If an abnormal tag is detected, the terminal immediately pushes a vehicle inspection prompt to the owner through the vehicle display screen and simultaneously informs the owner of the key observation features. After the owner adjusts the use of the vehicle's electrical appliances or completes the initial inspection, the terminal re-collects data and generates a tag within a specified period. If the new tag is normal, the power optimization is determined to be completed; otherwise, the warning is repeated.

[0030] In step S6, the system integrates newly added vehicle feature and power failure tag data according to a preset cycle, synchronously cleans up outdated historical data, and completes dynamic database updates to ensure that the monitoring model always adapts to the actual state of vehicle battery aging and operating condition changes.

[0031] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary sensing device embodiments described above, and that the invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the invention. Therefore, the embodiments should be considered in all respects as exemplary and non-limiting, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of equivalents of the claims are intended to be included within the present invention. No reference numerals in the claims should be construed as limiting the scope of the claims.

Claims

1. A method for power failure monitoring and power consumption optimization applied to vehicle-mounted terminals, characterized in that: The method includes the following steps: S1. Obtain real-time battery information as a power loss tag, preprocess and normalize vehicle data to obtain vehicle features, and establish a vehicle feature database based on the vehicle features. S2. For power failure tags, divide power failure tags into power failure tags for pending faults and power failure tags for non-faults; S3. Analyze the power outage tags for unconfirmed faults and non-fault power outage tags, and distinguish between the normal range of undetermined power outage speed and the abnormal range of undetermined power outage speed. S4. Based on the relationship between the undetermined normal range and the undetermined abnormal range of power loss speed, a second distinction is made between the undetermined normal range and the undetermined abnormal range of power loss speed. For power loss tags whose power loss speed is in the abnormal range of power loss speed, the power loss tag is marked as an abnormal tag. S5. During real-time vehicle operation, monitor the power failure tags of the vehicle, prompt abnormal tags, and finally determine that the user has completed the power optimization. S6. Update the historical database based on the preset update cycle.

2. The power failure monitoring and power consumption optimization method for vehicle-mounted terminals according to claim 1, characterized in that: In step S1, after authorization, real-time battery information is obtained. The battery information includes vehicle battery information and vehicle battery depletion rate information. The vehicle battery depletion rate is the ratio of the difference between the vehicle battery level before a power outage monitoring cycle and the current vehicle battery level to the vehicle's total battery level. The power outage monitoring cycle is a preset time length. The battery information is used as a power outage tag for the power outage monitoring cycle. Vehicle data within the power outage monitoring cycle is collected. The vehicle data includes vehicle internal operating data and vehicle environmental data. The vehicle data is preprocessed and normalized to obtain a set of vehicle features. A vehicle feature database is established based on the vehicle features. In the vehicle feature database, the vehicle features are classified based on the tag of the power outage monitoring cycle.

3. The power failure monitoring and power consumption optimization method for an in-vehicle terminal according to claim 2, characterized in that: In step S2, for any type of power failure tag, there are M groups of vehicle features corresponding to the power failure tag. The m-th group of vehicle features is analyzed. If the vehicle is judged to have a battery failure in a preset fault cycle after the m-th group of vehicle features is collected, the m-th group of vehicle features is judged to be a fault manifestation feature group. Otherwise, determine that the m-th group of vehicle features is not a fault display feature group, substitute each of m=1,2,…,M to obtain the number of fault display feature groups as M1, and then obtain the battery failure rate A=M1 / M of the power failure tag. If A is greater than A0, determine that the power failure tag is a power failure tag to be confirmed. A0 is the preset battery failure rate threshold. Otherwise, the power-off tag is determined not to be a non-fault power-off tag, and then the power-off tags are classified into unconfirmed fault power-off tags and non-fault power-off tags.

4. The power failure monitoring and power consumption optimization method for an in-vehicle terminal according to claim 3, characterized in that: In step S3, the power loss tags are analyzed, and the battery level information in the tags is retrieved. For each battery level information, the corresponding power loss tags are retrieved. These tags with the same battery level information include X non-faulty power loss tags and Y unconfirmed faulty power loss tags. The vehicle battery level of the X non-faulty power loss tags decreases at a rate of {A1, A2, ..., A...}. x ,…,A X }, where A x Let X represent the vehicle battery rate of the xth non-fault power loss tag. Remove impurities from the vehicle battery rate of the X non-fault power loss tags to obtain the undetermined normal range of power loss rate. The upper limit of the undetermined normal range of power loss rate is the maximum value of the vehicle battery rate of the X non-fault power loss tags after removing impurities, and the lower limit is the minimum value of the vehicle battery rate of the X non-fault power loss tags after removing impurities.

5. The power failure monitoring and power consumption optimization method for an in-vehicle terminal according to claim 4, characterized in that: The battery level of the vehicles with Y pending fault power loss tags decreases at a rate of {B1, B2, ..., B}. y ,…,B Y }, where B y Let represent the vehicle battery depletion rate of the y-th pending fault power loss tag. Remove nodules from the vehicle battery depletion rates of the X pending fault power loss tags to obtain the undetermined abnormal range of power loss rate. The upper limit of the undetermined abnormal range of power loss rate is the maximum value of the vehicle battery depletion rate of the X pending fault power loss tags after removing nodules, and the lower limit is the minimum value of the vehicle battery depletion rate of the X pending fault power loss tags after removing nodules. Then, for any battery information, obtain the undetermined normal range of power loss rate and the undetermined abnormal range of power loss rate. For any battery information, if the undetermined normal range of power loss rate and the undetermined abnormal range of power loss rate do not intersect, determine that the undetermined normal range of power loss rate is the normal range of power loss rate, and determine that the undetermined abnormal range of power loss rate is the abnormal range of power loss rate. Otherwise, a secondary analysis is performed on the undetermined interval, which is the intersection of the undetermined normal interval of power loss rate and the undetermined abnormal interval of power loss rate.

6. The power failure monitoring and power consumption optimization method for an in-vehicle terminal according to claim 5, characterized in that: In step S4, when there is an intersection between the normal interval and the abnormal interval of the undetermined power loss rate, a secondary analysis is performed on the interval to be determined. In the undetermined interval, for the vehicle battery depletion rate information α, N sets of vehicle features of the power loss tag are called, where the nth set of vehicle features is {C 1_n C 2_n ,…,C q_n ,…,C Q_n }, where C q_n Let Q represent the q-th vehicle feature in the n-th vehicle feature group, where Q represents the number of vehicle features in the group. Substituting each value into n=1,2,…,N, we obtain the q-th vehicle feature in the N-th vehicle feature group. The q-th vehicle feature in the N-th vehicle feature group is {C}. q_1 C q_2 ,…,C q_n ,…,C q_N }, and then obtain the feature deviation D of the q-th vehicle feature when considering the vehicle's battery depletion rate information. q : ； Where E q E represents the q-th standard vehicle feature when considering the vehicle's battery depletion rate information. q Let D be the average value of the q-th vehicle feature in all feature groups within the normal interval. q If the deviation of the qth vehicle feature is greater than the preset deviation threshold, the qth vehicle feature is determined to be abnormal and is considered an abnormal vehicle feature; otherwise, the qth vehicle feature is determined to be normal, and then the vehicle battery depletion rate information α is judged. If there is an abnormal vehicle feature, the vehicle battery depletion rate information α is determined to be in the abnormal power loss rate range. Otherwise, if the vehicle battery depletion rate information α is determined to be within the normal range of power loss rate, then all vehicle battery depletion rate information in the specified range will be judged to complete the distinction between the abnormal range and the normal range of power loss rate. After confirming the abnormal range of power loss rate, any power loss tag within the abnormal range of power loss rate will be marked as an abnormal tag. If the q-th vehicle feature is an abnormal vehicle feature, the q-th vehicle feature will be marked as a key observation feature.

7. The power failure monitoring and power consumption optimization method for an in-vehicle terminal according to claim 6, characterized in that: In step S5, during real-time vehicle operation, the power failure tag is monitored in real time. If the power failure tag is an abnormal tag, the user is prompted to check the vehicle and is advised to pay attention to key features. If the vehicle features change within a preset prompting period after the prompt, the power failure tag is regenerated. If the power failure tag is again determined to be an abnormal tag, the user is prompted to check the vehicle again; otherwise, it is determined that the user has completed the power optimization.

8. The power failure monitoring and power consumption optimization method for an in-vehicle terminal according to claim 6, characterized in that: In step S6, after a preset update cycle, the historical data in the database is updated.

9. A power failure monitoring and power consumption optimization system for an in-vehicle terminal, wherein the system is applied to the power failure monitoring and power consumption optimization method for an in-vehicle terminal as described in any one of claims 1-8, characterized in that: The system includes: a vehicle feature library construction module, a power failure tag initial classification module, a power failure interval initial delineation module, a power failure tag re-labeling module, a real-time anomaly warning optimization module, and a historical database update module; The vehicle feature database construction module is used to obtain real-time power information as a power loss tag, preprocess and normalize vehicle data to obtain vehicle features, and establish a vehicle feature database based on the vehicle features. The power failure tag initial classification module classifies power failure tags into power failure tags with pending faults and non-fault power failure tags. The power outage interval initial delineation module is used to analyze the power outage tags to be confirmed and the non-fault power outage tags, and to distinguish between the normal range of undetermined power outage speed and the abnormal range of undetermined power outage speed. The power failure tag relabeling module is used to perform a second distinction between the undetermined normal range and the undetermined abnormal range of power failure speed based on the relationship between the undetermined normal range and the undetermined abnormal range of power failure speed. For power failure tags whose power failure speed is in the abnormal range of power failure speed, the power failure tag is labeled as an abnormal tag. The real-time anomaly warning and optimization module is used to monitor the power loss tag of the vehicle in real time during vehicle operation, prompt the anomaly tag, and finally determine that the user has completed the power optimization. The historical database update module is used to update the historical database based on a preset update cycle.

Images