Monitoring method and device
By using personalized modeling and multi-dimensional deviation index identification, the problem of misjudgment in vehicle charging metering anomaly monitoring has been solved, and accurate identification and monitoring of vehicle charging metering has been achieved.
Patent Information
- Application Number
- CN202511957854.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-12-23
- Publication Date
- 2026-04-24
AI Technical Summary
Existing technologies ignore the aging of vehicle batteries and differences in charging habits, resulting in a high misjudgment rate in monitoring abnormal charging meters and making it impossible to accurately identify vehicle charging meters.
By acquiring historical and current charging information of the target vehicle, the historical average charging amount and predicted charging amount are determined. Combining baseline weights and charging scenario weights, multi-dimensional deviation indices and verification indices are calculated to achieve personalized modeling and anomaly identification for vehicle charging metering.
It enables accurate identification of vehicle charging metering anomalies, adapts to changes in charging scenarios, avoids misjudgments caused by vehicle battery aging and differences in charging habits, and improves the accuracy and robustness of metering monitoring.
Smart Images
Figure CN121912841A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of electric vehicle charging services, and more particularly to a monitoring method and device. Background Technology
[0002] With the large-scale expansion of electric vehicle charging networks, charging operation platforms have become the core hub connecting charging piles, vehicles, and users. However, the accuracy of charging metering is becoming increasingly prominent. On the one hand, there is a significant discrepancy between the actual charging volume and the reported charging volume; on the other hand, some businesses are even tampering with metering data to gain illicit profits, seriously affecting the industry's reputation and development.
[0003] Existing technologies use group average models for monitoring, such as setting uniform metering thresholds for a particular brand of charging piles. However, this method ignores the differences in battery aging at different lifecycles for the same vehicle and the differences in charging habits among different users, leading to a high false alarm rate. In addition, existing technologies mostly rely on constant current phase data or fixed thresholds at the charging pile for judgment, without combining historical data of individual vehicles to establish a dynamic baseline, and thus cannot capture the personalized metering characteristics of vehicle and charging pile matching.
[0004] Therefore, it is of great significance to achieve accurate identification of abnormal vehicle charging metering. Summary of the Invention
[0005] This application provides a monitoring method and apparatus to solve the problem that the prior art ignores the aging of vehicle batteries and the differences in charging habits, resulting in a high misjudgment rate of abnormal charging metering, and to achieve accurate identification of abnormal vehicle charging metering.
[0006] Firstly, this application provides a monitoring method, the method comprising: Obtain historical and current charging information for the target vehicle; Based on the historical charging information and the current charging information, the historical average charging amount and the predicted charging amount are determined, and the target charging amount baseline is determined based on the baseline weight, the historical average charging amount and the predicted charging amount. Based on the current charging information and the target charging baseline, a multi-dimensional deviation index is determined; Based on the current charging information, determine the charging scenario weight, and based on the current charging information, the multi-dimensional deviation index, and the charging scenario weight, determine the target deviation index; The current multi-dimensional inspection index is determined based on the current charging information, and the current charging metering monitoring result of the target vehicle is determined based on the target deviation index and the current multi-dimensional inspection index.
[0007] Optionally, according to the monitoring method of this application, determining the historical average charging amount and the predicted charging amount based on the historical charging information and the current charging information includes: Based on the current charging information, determine the target state of charge (SOC) range; Based on the historical charging information, determine the historical average charging amount for the target SOC range; Based on the historical charging information, the predicted charging amount for the target SOC range is determined.
[0008] Optionally, according to the monitoring method of this application, determining the target charging baseline based on the baseline weight, the historical average charging amount, and the predicted charging amount includes: Obtain the battery usage time of the target vehicle; If the battery usage time is less than a first duration threshold, the baseline weight is a first weight value; the target charging amount baseline is determined based on the first weight value, the historical average charging amount, and the predicted charging amount. If the battery usage time is greater than or equal to a first duration threshold, the baseline weight is a second weight value; the target charging amount baseline is determined based on the second weight value, the historical average charging amount, and the predicted charging amount. Wherein, the first weight value is greater than the second weight value.
[0009] Optionally, according to the monitoring method of this application, the current charging information includes the current cumulative power, the current SOC change rate, the current output power, and the charging time; The step of determining the multi-dimensional deviation index based on the current charging information and the target charging baseline includes: The power deviation index is determined based on the temperature coefficient, the current cumulative power level, and the target charging baseline. Based on the current SOC change rate and the historical average SOC change rate, determine the SOC change rate deviation index; The power stability deviation index is determined based on the current output power, the charging time, and the SOC power fitting curve. The multi-dimensional deviation index is determined based on the power deviation index, the SOC change rate index, and the power stability deviation index.
[0010] Optionally, according to the monitoring method of this application, the current charging information also includes the current temperature value; The step of determining the charging scenario weight based on the current charging information includes: Based on the current charging information, determine the target SOC range; The charging scenario characteristics are determined based on the abnormal state of the current temperature value, the fluctuation state of the current output power, and the target SOC range; The weight of the charging scenario is determined based on the correlation between the characteristics of historical anomalies and the characteristics of the charging scenario.
[0011] Optionally, according to the monitoring method of this application, the current charging information includes the current cumulative power consumption, the current charging cost, and the current temperature value; The step of determining the current multi-dimensional inspection index based on the current charging information, and determining the current charging metering monitoring result of the target vehicle based on the target deviation index and the current multi-dimensional inspection index, includes: The physical inspection deviation index is determined based on the ampere-hour integral values of the output voltage and output current and the current cumulative charge. The cost deviation index is determined based on the historical average charging unit price, the current charging cost, and the current cumulative electricity consumption. The temperature test deviation index is determined based on the temperature threshold and the current temperature value; The current multi-dimensional inspection index is determined based on the physical inspection deviation index, the cost inspection deviation index, and the temperature inspection deviation index. Based on the target deviation index and the current multi-dimensional test index, anomaly category labels and anomaly confidence levels are determined, and based on the anomaly category labels and anomaly confidence levels, the current charging metering monitoring results of the target vehicle are determined.
[0012] Secondly, this application also provides a monitoring device, the device comprising: The acquisition module is used to acquire the historical charging information and current charging information of the target vehicle; The first determining module is used to determine the historical average charging amount and the predicted charging amount based on the historical charging information and the current charging information, and to determine the target charging amount baseline based on the baseline weight, the historical average charging amount and the predicted charging amount. The second determining module is used to determine a multi-dimensional deviation index based on the current charging information and the target charging baseline; The third determining module is used to determine the charging scenario weight based on the current charging information, and to determine the target deviation index based on the current charging information, the multi-dimensional deviation index, and the charging scenario weight. The monitoring module is used to determine the current multi-dimensional inspection index based on the current charging information, and to determine the current charging metering monitoring result of the target vehicle based on the target deviation index and the current multi-dimensional inspection index.
[0013] Thirdly, this application provides an electronic device including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the monitoring method described in the first aspect above.
[0014] Fourthly, this application provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the monitoring method described in the first aspect above.
[0015] Fifthly, this application also provides a computer program product, including a computer program that, when executed by a processor, implements the monitoring method described in the first aspect above.
[0016] The monitoring method and apparatus provided in this application determine the target charging baseline based on the target vehicle's historical and current charging information, achieving personalized modeling of the matching characteristics between the vehicle and the charging pile. By using current charging information, multi-dimensional deviation indices, and charging scenario weights, a target deviation index is determined. Based on the target deviation index and the current multi-dimensional verification index, the current charging metering monitoring result of the target vehicle is determined. This adapts to changes in charging scenarios and avoids the problem of high misjudgment rates in charging metering anomaly monitoring caused by vehicle battery aging and differences in charging habits, thereby achieving accurate identification of vehicle charging metering anomalies. Attached Figure Description
[0017] To more clearly illustrate the technical solutions in this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0018] Figure 1 This is a flowchart illustrating the monitoring method provided in this application.
[0019] Figure 2 This is a schematic diagram of the monitoring device provided in this application.
[0020] Figure 3 This is a schematic diagram of the structure of the electronic device provided in this application. Detailed Implementation
[0021] To make the objectives, technical solutions, and advantages of this application clearer, the technical solutions of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.
[0022] The descriptions of "first," "second," etc., appearing in the embodiments of this application are for illustrative purposes and to distinguish the objects being described. They have no order and do not indicate any special limitation on the number of objects in the embodiments of this application. They cannot constitute any limitation on the embodiments of this application.
[0023] Figure 1 This is a flowchart illustrating the monitoring method provided in this application, such as... Figure 1 As shown, the method may include the following steps: Step 110: Obtain the historical charging information and current charging information of the target vehicle; Step 120: Based on historical charging information and current charging information, determine the historical average charging amount and the predicted charging amount, and determine the target charging amount baseline based on the baseline weight, the historical average charging amount and the predicted charging amount. Step 130: Determine the multi-dimensional deviation index based on the current charging information and the target charging baseline; Step 140: Determine the charging scenario weight based on the current charging information, and determine the target deviation index based on the current charging information, multi-dimensional deviation index and charging scenario weight; Step 150: Determine the current multi-dimensional inspection index based on the current charging information, and determine the current charging metering monitoring result of the target vehicle based on the target deviation index and the current multi-dimensional inspection index.
[0024] It should be noted that the aforementioned monitoring method is implemented by computer equipment, which has the capability to collect, process, analyze, and determine anomalies in vehicle charging data. This computer equipment can be integrated into the charging operation platform or can be a distributed processing architecture that includes cloud servers and edge computing devices. This application does not impose specific limitations in this regard.
[0025] Unless otherwise specified, the following embodiments use a charging monitoring device as the execution subject to illustrate the monitoring method of this application.
[0026] Specifically, the target vehicle refers to a car that is connected to a charging station for electrical replenishment. For example, it can be uniquely identified by its VIN (Vehicle Identification Number) to link to the vehicle's historical charging information.
[0027] Historical charging information refers to the charging records of the target vehicle over a past period stored in a database. This historical charging information may include historical SOC (State of Charge) ranges, historical cumulative charge, historical charging duration, historical temperature data, etc. For example, the target vehicle can obtain historical charging information via the CAN (Controller Area Network) bus. This application does not specifically limit this.
[0028] Current charging information refers to information collected in real time during the vehicle's current charging process. This information may include current cumulative battery level, current State of Charge (SOC) change rate, current output power, current temperature, and charging duration.
[0029] The baseline weight is used to adjust the ratio of historical average charging amount to predicted charging amount, thereby generating a target charging amount baseline that better reflects the current actual state of the vehicle.
[0030] The target charging baseline refers to the benchmark value used to measure whether the vehicle's current charging is abnormal.
[0031] Charging scenario weight refers to a coefficient generated based on current charging information, used to adjust the importance of the multi-dimensional deviation index in the current charging metering and monitoring results.
[0032] Current charging metering and monitoring results can include whether there is an anomaly, the type of anomaly, and the confidence level of the anomaly.
[0033] Specifically, the charging monitoring equipment obtains the target vehicle's historical and current charging information via the charging pile's CAN bus. The charging pile is compatible with mainstream charging pile communication protocols, such as supporting OCPP (Open Charge Point Protocol) 1.6J.
[0034] Specifically, in step 110, the charging monitoring device indexes the target vehicle's historical charging information based on its VIN code; simultaneously, it collects the settlement data and process data at the end of this charging session as current charging information. For example, current charging information may include current cumulative battery level, current SOC change rate, current output power, current temperature value, and charging duration.
[0035] In step 120, the charging monitoring device determines the historical average charging amount based on historical charging information, and at the same time, inputs the historical charging information into the artificial intelligence model to determine the predicted charging amount.
[0036] Subsequently, the charging monitoring equipment determines the target charging baseline based on baseline weights, historical average charging volume, and predicted charging volume.
[0037] In step 130, the charging monitoring device determines a multi-dimensional deviation index based on the current cumulative power, current SOC change rate, current output power, charging time, and target charging baseline in the current charging information.
[0038] In step 140, the charging monitoring device analyzes the current charging information and identifies charging scenario characteristics. For example, it identifies whether the current charging scenario is a normal scenario with suitable temperature or a special scenario with extreme temperature. Based on the charging scenario characteristics, the system determines the corresponding charging scenario weight. This weight indicates which type of deviation feature in the multi-dimensional deviation index should be given more attention in the current charging scenario. For example, in some scenarios, more attention may be paid to the deviation in the charging quantity dimension, while in other scenarios, more attention may be paid to the deviation in the power dimension.
[0039] Next, the charging monitoring device determines the target deviation index based on the charging duration, multi-dimensional deviation index, and charging scenario weight of the current charging information. In this way, the charging monitoring device can use scenario weights to correct the multi-dimensional deviation index, filter out non-substantial deviations caused by the charging scenario, and calculate a target deviation index that more accurately reflects metering anomalies.
[0040] In step 150, the charging monitoring device determines the current multi-dimensional inspection index based on the current cumulative power, current charging cost, and current temperature value in the current charging information.
[0041] Subsequently, the charging monitoring equipment inputs the target deviation index and the current multi-dimensional test index into the XGBoost (eXtreme Gradient Boosting) classification model, and the XGBoost classification model outputs the current charging metering monitoring results.
[0042] The monitoring method provided in this application determines the target charging baseline based on the target vehicle's historical and current charging information, achieving personalized modeling of the matching characteristics between the vehicle and the charging station. By using current charging information, multi-dimensional deviation indices, and charging scenario weights, a target deviation index is determined. Based on this target deviation index and the current multi-dimensional verification index, the current charging metering monitoring result for the target vehicle is determined. This method adapts to changes in charging scenarios and avoids the problem of high misjudgment rates in charging metering anomaly monitoring caused by vehicle battery aging and differences in charging habits, thereby achieving accurate identification of vehicle charging metering anomalies.
[0043] In one embodiment, determining the historical average charge amount and the predicted charge amount based on historical charging information and current charging information may include: Determine the target SOC range based on the current charging information; Based on historical charging information, determine the historical average charging amount for the target SOC range; Based on historical charging information, the predicted charging amount for the target SOC range is determined.
[0044] Specifically, the charging monitoring device obtains the starting SOC and ending SOC of this charge based on the current charging information, and determines the target SOC range based on the starting SOC and ending SOC.
[0045] The charging monitoring equipment reads vehicle parameters via the CAN bus, receives historical charging information via HTTP (Hypertext Transfer Protocol) or HTTPS (Hypertext Transfer Protocol Secure), and uses 3D charging technology. The criteria filter out abrupt changes in historical charging information and align data by timestamp.
[0046] Subsequently, charging records matching the target SOC range are searched within the pre-processed historical charging information. To ensure the timeliness and reference value of the data, the charging monitoring equipment can employ a sliding window algorithm, which can be deployed at the edge of the charging pile. For example, the 10 most recent valid charging records within the same SOC range can be selected.
[0047] Furthermore, the charging monitoring equipment calculates the average of the cumulative electricity consumption over these historical records as the historical average charging amount. The historical average charging amount can be expressed by the following formula: in, This represents the historical average charging amount, and N represents the number of historical samples. This indicates the historical cumulative electricity consumption.
[0048] Meanwhile, charging monitoring equipment can utilize artificial intelligence models to determine the predicted charging amount within the target SOC range. For example, in the process of building the artificial intelligence model, in order to enable the model to have accurate predictive capabilities, a basic dataset containing 100,000 vehicles and 12 months of charging records (including VIN codes and anomaly labeling) is used for model training; to address the problem of scarce abnormal samples, data augmentation is achieved through time series flipping, temperature parameter perturbation, and other methods.
[0049] For example, in the actual monitoring and prediction process, the charging monitoring equipment inputs the historical charging data of nearly 3 months into a bidirectional LSTM (Long Short-Term Memory) network in time series. The historical charging data can include feature sequences such as historical SOC range, historical temperature data, and historical output power data.
[0050] Optionally, to reduce latency, in terms of network structure, the number of hidden layer nodes in the bidirectional LSTM can be reduced from 128 to 64, and after TensorRT quantization acceleration, the inference latency can be controlled to around 25ms. At the same time, the attention layer can be simplified to a single-head structure, reducing the parameter size by 60%. In terms of distributed architecture, the bidirectional LSTM can be deployed in the cloud to process nationwide batch data. Meanwhile, a lightweight attention layer and an MLP (Multi-Layer Perceptron) priority model can be deployed at the edge, such as charging piles or vehicle-mounted T-BOX (Telematics BOX), and the edge only uploads abnormal data to the cloud, thereby effectively reducing bandwidth pressure.
[0051] The monitoring method provided in this application determines the historical average charging amount of the target SOC range through historical charging information; then, it determines the predicted charging amount of the target SOC range through historical charging information, thereby effectively capturing the charging amount change pattern of the same vehicle in the same SOC range. This avoids the baseline inaccuracy problem caused by traditional methods ignoring battery aging and differences in charging habits, thus providing a reliable data foundation for the subsequent generation of a high-precision target charging amount baseline.
[0052] In one embodiment, determining the target charging baseline based on baseline weights, historical average charging amount, and predicted charging amount may include: Obtain the battery usage time of the target vehicle; If the battery usage time is less than the first duration threshold, the baseline weight is the first weight value; the target charging amount baseline is determined based on the first weight value, the historical average charging amount, and the predicted charging amount. If the battery usage time is greater than or equal to the first duration threshold, the baseline weight is the second weight value; the target charging amount baseline is determined based on the second weight value, the historical average charging amount, and the predicted charging amount. Among them, the first weight value is greater than the second weight value.
[0053] Specifically, battery usage time is used to characterize the degree of battery aging.
[0054] The first duration threshold is used as a reference point for dividing the battery life cycle stages. For example, the first duration threshold could be 2 years. This application does not impose a specific limitation on this.
[0055] Specifically, the charging monitoring equipment obtains the battery usage time of the target vehicle.
[0056] If the battery usage time is detected to be less than a first duration threshold, the charging monitoring device determines that the battery is in a stable performance period. At this time, the historical average charging amount has high reference value, and the charging monitoring device sets the baseline weight to a high first weight value. Subsequently, the charging monitoring device determines the target charging amount baseline based on the first weight value, the historical average charging amount, and the predicted charging amount.
[0057] If the battery usage time is detected to be greater than or equal to a first duration threshold, the charging monitoring device determines that the battery is in its performance aging period. At this time, the battery performance degradation exhibits a non-linear trend, and the historical average charging amount is prone to lag bias and has low reference value. Therefore, the charging monitoring device sets the baseline weight to a lower second weight value. Subsequently, the charging monitoring device determines the target charging amount baseline based on the second weight value, the historical average charging amount, and the predicted charging amount.
[0058] The target charging baseline can be determined using the following formula: in, Indicates the target charging baseline. This is the baseline weight, initially set at 0.6, and adjusted downwards as battery usage time increases. This indicates the predicted charging amount.
[0059] The monitoring method provided in this application uses a first weight value for the baseline when the battery usage time is less than a first time threshold, and a second weight value when the battery usage time is greater than or equal to the first time threshold. This allows the method to adaptively adjust the baseline weight according to different stages of the battery life cycle, avoiding the problem that a single fixed weight cannot cope with the historical data lag and baseline inaccuracy caused by battery aging, thereby significantly improving the robustness and accuracy of charging metering anomaly monitoring.
[0060] In one embodiment, the current charging information includes the current cumulative battery level, the current SOC change rate, the current output power, and the charging duration. Based on current charging information and the target charging baseline, a multi-dimensional deviation index is determined, including: The power deviation index is determined based on the temperature coefficient, the current cumulative power consumption, and the target charging baseline. The SOC change rate deviation index is determined based on the current SOC change rate and the historical average SOC change rate. The power stability deviation index is determined based on the current output power, charging time, and SOC power fitting curve. A multi-dimensional deviation index is determined based on the power deviation index, the SOC change rate index, and the power stability deviation index.
[0061] Specifically, the temperature coefficient can be output through a random forest regression model. For example, by inputting the historical highest temperature, battery cycle count, and output power into the random forest regression model, the model outputs a real-time dynamic temperature coefficient. For instance, it might output a temperature coefficient of 0.02 at a low temperature of -10°C and 0.01 at a normal temperature of 25°C. This application does not impose any specific limitations on this.
[0062] Specifically, the charging monitoring equipment determines the power deviation index based on the temperature coefficient, the current cumulative power level, and the target charging baseline. The power deviation index can be expressed by the following formula: in, This indicates the deviation of the power consumption index. This indicates the current cumulative battery level. This represents the temperature coefficient.
[0063] The charging monitoring equipment monitors the current SOC change rate in real time during the charging process. Simultaneously, the equipment retrieves the historical average SOC change rate for the vehicle under identical conditions from a database and calculates the relative deviation between the two values to determine the SOC change rate deviation index. The SOC change rate deviation index can be expressed by the following formula: in, This indicates that the rate of change of SOC deviates from the exponent. This represents the current rate of change of SOC. This represents the historical average rate of change of SOC.
[0064] Establish SOC power fitting curve The SOC power fitting curve represents the ideal trajectory of a vehicle's charging power as a function of SOC under normal conditions. Subsequently, the charging monitoring equipment determines the power stability deviation index based on the current output power, charging duration, and the SOC power fitting curve. The power stability deviation index can be expressed by the following formula: in, This represents the power stability deviation index, where T represents the charging time, and φ represents the time variable during the charging process, representing any sampling time point. This represents the current output power collected at time point A. This represents the SOC corresponding to time point 𝑡. This represents the power value at SoC(t).
[0065] Subsequently, a multi-dimensional deviation index was determined based on the power deviation index, the SOC change rate index, and the power stability deviation index.
[0066] The monitoring method provided in this application determines a multi-dimensional deviation index through the power deviation index, the SOC change rate index, and the power stability deviation index. This avoids the shortcomings of being unable to distinguish between natural battery degradation and actual metering faults due to a single verification dimension, thereby achieving comprehensive and accurate monitoring of charging metering anomalies.
[0067] In one embodiment, the current charging information also includes the current temperature value; Based on the current charging information, determine the weight of the charging scenario, including: Determine the target SOC range based on the current charging information; Based on the abnormal state of the current temperature value, the fluctuation state of the current output power, and the target SOC range, the characteristics of the charging scenario are determined; The weight of a charging scenario is determined based on the correlation between historical anomaly information features and charging scenario features.
[0068] Specifically, the abnormal state of the current temperature value is used to characterize whether the current battery temperature is in an extreme range that may affect charging performance or cause malfunctions. For example, if the current highest temperature exceeds a preset normal standard range, the current temperature value is determined to be in an abnormal state.
[0069] The fluctuation state of the current output power is used to characterize the instability of power output during charging. For example, the current output power may exhibit severe oscillations, jumps, or other non-stationary conditions. This application does not impose specific limitations on this.
[0070] Historical anomaly information features refer to features extracted from a large number of historical anomaly cases. For example, electricity parameters are identified as key discriminant indicators in 90% of anomaly cases.
[0071] Specifically, firstly, the charging monitoring equipment determines the target SOC range based on the current charging information.
[0072] Subsequently, the charging monitoring equipment determines the characteristics of the charging scenario based on the abnormal state of the current temperature, the fluctuation state of the current output power, and the target SOC range. For example, the characteristics of the charging scenario include whether the current temperature is abnormal, whether the current output power is stable, and whether the target SOC range is a critical range for fast charging.
[0073] Furthermore, the charging monitoring device utilizes the attention layer of the Transformer to calculate the correlation between charging scene features and historical anomaly information features, and then determines the charging scene weight based on this correlation. For example, in a low-temperature scenario, the output charging scene weight might be... Prioritize the power stability deviation index; for example, in high-temperature scenarios, the weight of the output charging scenario may be [missing information]. Prioritize monitoring the power deviation index.
[0074] At this point, the target deviation index can be expressed by the following formula: in, This indicates a deviation from the target index. This indicates the weight of the charging scenario.
[0075] The monitoring method provided in this application determines the characteristics of the charging scenario by considering the abnormal state of the current temperature value, the fluctuation state of the current output power, and the target SOC range. Then, based on the correlation between the historical abnormal information characteristics and the charging scenario characteristics, the charging scenario weight is determined. This method can adjust the charging scenario weight according to the charging scenario characteristics, thereby paying attention to the deviation index in different dimensions and avoiding misjudgment by the fixed weight strategy in complex environments. This significantly improves the accuracy and adaptability of charging metering anomaly monitoring.
[0076] In one embodiment, the current charging information includes the current cumulative battery level, the current charging cost, and the current temperature. Based on the current charging information, a current multi-dimensional inspection index is determined, and based on the target deviation index and the current multi-dimensional inspection index, the current charging metering monitoring result for the target vehicle is determined, including: The physical inspection deviation index is determined based on the ampere-hour integral values of the output voltage and output current and the current cumulative charge. The cost deviation index is determined based on the historical average charging unit price, current charging cost, and current cumulative electricity consumption. The temperature inspection deviation index is determined based on the temperature threshold and the current temperature value; The current multi-dimensional inspection index is determined based on the physical inspection deviation index, the cost inspection deviation index, and the temperature inspection deviation index. Based on the target deviation index and the current multi-dimensional test index, determine the anomaly category label and anomaly confidence level, and based on the anomaly category label and anomaly confidence level, determine the current charging metering monitoring result of the target vehicle.
[0077] Specifically, output voltage and output current refer to the sequence of voltage and current values output by the charging pile in real time during the charging process.
[0078] The ampere-hour integral value refers to the physical energy value obtained by integrating the output voltage and output current over time.
[0079] The physical deviation index is used to measure whether the current cumulative electricity reported by the charging pile conforms to the basic physical law of conservation of energy.
[0080] The cost deviation index is used to measure whether there are any anomalies in the current billing logic. For example, a sudden increase in unit price.
[0081] The temperature deviation index is used to determine whether the current anomaly is caused by performance degradation due to extreme environments.
[0082] Anomaly category labels refer to the specific anomaly classifications output by the artificial intelligence model. For example, analysis using the XGBoost classification model yields three classification results: power tampering, equipment malfunction, or environmental interference.
[0083] Anomaly confidence refers to the degree of confidence that an artificial intelligence model has in an anomaly category label.
[0084] Specifically, the charging monitoring equipment calculates the ampere-hour integral value. The ampere-hour integral value can be expressed by the following formula: in, Indicates the output voltage. Indicates at a point in time The output current.
[0085] Subsequently, the charging monitoring equipment confirms the consistency between the ampere-hour integral value and the current cumulative power consumption reported by the charging pile. For example, if the deviation between the ampere-hour integral value and the current cumulative power consumption exceeds a preset threshold, a physical discrepancy is marked, thereby determining the physical inspection deviation index. The preset threshold can be 3%, and this application does not specify a particular threshold.
[0086] The charging monitoring equipment calculates the actual unit price of the current charging by dividing the current charging cost by the current cumulative electricity consumption. The equipment then compares this actual unit price with the historical average charging unit price. For example, if the deviation between the actual unit price and the historical average charging unit price exceeds a preset threshold, and no official price adjustment record is found, the equipment marks a mismatch between cost and electricity consumption, thereby determining a cost deviation index. The preset threshold can be 10%, and this application does not specify a particular threshold.
[0087] The charging monitoring equipment determines whether the current temperature exceeds a temperature threshold. For example, if the temperature is extreme, it indicates that the current measurement deviation may be due to environmental factors causing battery performance degradation. In this case, a temperature deviation index is determined to lower the warning level in subsequent processes.
[0088] The charging monitoring equipment determines the current multi-dimensional inspection index based on the physical inspection deviation index, the cost inspection deviation index, and the temperature inspection deviation index; Subsequently, the charging monitoring device inputs the current multi-dimensional inspection index and the target deviation index into a pre-trained XGBoost classification model. This XGBoost classification model is trained based on a large number of historical labeled cases, such as power tampering and equipment failure. The XGBoost classification model performs classification reasoning on the input features and outputs anomaly category labels and corresponding anomaly confidence scores. For example, for power tampering, the confidence score is 92%.
[0089] Optionally, the charging monitoring device can also perform auxiliary priority ranking. The device inputs the anomaly category label, anomaly confidence score, and the number of historical violations of the charging pile into the MLP, and the MLP model outputs the warning priority. For example, a high-priority warning is output for a charging pile with tampered power and 3 historical violations; a low-priority warning is output for a charging pile with environmental interference and 0 historical violations.
[0090] Optionally, for every 100 newly labeled cases, the charging monitoring device updates the parameters of the XGBoost classification model through incremental learning to maintain a classification accuracy of over 95%. Simultaneously, the charging monitoring device monitors the XGBoost classification accuracy in real time, and automatically triggers full-data retraining if the accuracy falls below 90%.
[0091] Optionally, the charging monitoring device pushes JSON format warning information to the operation platform. The JSON format warning information includes the anomaly level, type, confidence level, and suggested measures, and supports integration with the merchant management backend and user APP.
[0092] Optionally, the charging monitoring device reserves a data field for market regulators. This data field includes traditional verification results, AI model version, and AI feature weights, thereby providing traceable evidence of measurement anomalies.
[0093] The monitoring method provided in this application determines the current multi-dimensional inspection index by using physical inspection deviation index, cost inspection deviation index, and temperature inspection deviation index; then, by using the target deviation index and the current multi-dimensional inspection index, it determines the anomaly category label and anomaly confidence level, and determines the current charging metering monitoring result of the target vehicle based on the anomaly category label and anomaly confidence level. This method can accurately identify the specific category and confidence level of the anomaly, avoiding the shortcomings of traditional methods that cannot distinguish between natural battery degradation and actual metering faults due to a single verification dimension, thereby significantly improving the accuracy of identifying vehicle charging metering anomalies.
[0094] The monitoring device provided in this application will be described below. The monitoring device described below can be referred to in correspondence with the monitoring method described above and can achieve the same technical effect. It will not be described again here.
[0095] Figure 2 This is a schematic diagram of the monitoring device provided in this application, as shown below. Figure 2 As shown, the device may include: The acquisition module 210 is used to acquire the historical charging information and current charging information of the target vehicle; The first determining module 220 is used to determine the historical average charging amount and the predicted charging amount based on the historical charging information and the current charging information, and to determine the target charging amount baseline based on the baseline weight, the historical average charging amount and the predicted charging amount. The second determining module 230 is used to determine a multi-dimensional deviation index based on the current charging information and the target charging amount baseline; The third determining module 240 is used to determine the charging scenario weight based on the current charging information, and to determine the target deviation index based on the current charging information, the multi-dimensional deviation index and the charging scenario weight; The monitoring module 250 is used to determine the current multi-dimensional inspection index based on the current charging information, and to determine the current charging metering monitoring result of the target vehicle based on the target deviation index and the current multi-dimensional inspection index.
[0096] In one embodiment, the first determining module 220 is specifically used for: Based on the current charging information, determine the target state of charge (SOC) range; Based on the historical charging information, determine the historical average charging amount for the target SOC range; Based on the historical charging information, the predicted charging amount for the target SOC range is determined.
[0097] In one embodiment, the first determining module 220 is specifically used for: Obtain the battery usage time of the target vehicle; If the battery usage time is less than a first duration threshold, the baseline weight is a first weight value; the target charging amount baseline is determined based on the first weight value, the historical average charging amount, and the predicted charging amount. If the battery usage time is greater than or equal to a first duration threshold, the baseline weight is a second weight value; the target charging amount baseline is determined based on the second weight value, the historical average charging amount, and the predicted charging amount. Wherein, the first weight value is greater than the second weight value.
[0098] In one embodiment, the current charging information includes the current cumulative battery level, the current SOC change rate, the current output power, and the charging duration; The second determining module 230 is specifically used for: The power deviation index is determined based on the temperature coefficient, the current cumulative power level, and the target charging baseline. Based on the current SOC change rate and the historical average SOC change rate, determine the SOC change rate deviation index; The power stability deviation index is determined based on the current output power, the charging time, and the SOC power fitting curve. The multi-dimensional deviation index is determined based on the power deviation index, the SOC change rate index, and the power stability deviation index.
[0099] In one embodiment, the current charging information further includes the current temperature value; The third determining module 240 is specifically used for: Based on the current charging information, determine the target SOC range; The charging scenario characteristics are determined based on the abnormal state of the current temperature value, the fluctuation state of the current output power, and the target SOC range; The weight of the charging scenario is determined based on the correlation between the characteristics of historical anomalies and the characteristics of the charging scenario.
[0100] In one embodiment, the current charging information includes the current cumulative power consumption, the current charging cost, and the current temperature value; The monitoring module 250 is specifically used for: The physical inspection deviation index is determined based on the ampere-hour integral values of the output voltage and output current and the current cumulative charge. The cost deviation index is determined based on the historical average charging unit price, the current charging cost, and the current cumulative electricity consumption. The temperature test deviation index is determined based on the temperature threshold and the current temperature value; The current multi-dimensional inspection index is determined based on the physical inspection deviation index, the cost inspection deviation index, and the temperature inspection deviation index. Based on the target deviation index and the current multi-dimensional test index, anomaly category labels and anomaly confidence levels are determined, and based on the anomaly category labels and anomaly confidence levels, the current charging metering monitoring results of the target vehicle are determined.
[0101] Figure 3 An example is a schematic diagram of the physical structure of an electronic device, such as... Figure 3 As shown, the electronic device may include: a processor 310, a communications interface 320, a memory 330, and a communication bus 340, wherein the processor 310, the communications interface 320, and the memory 330 communicate with each other via the communication bus 340. The processor 310 can call logical instructions in the memory 330 to execute the monitoring method described in any of the above embodiments, such as including: Obtain the target vehicle's historical and current charging information; Based on the historical charging information and the current charging information, the historical average charging amount and the predicted charging amount are determined, and the target charging amount baseline is determined based on the baseline weight, the historical average charging amount and the predicted charging amount. Based on the current charging information and the target charging baseline, a multi-dimensional deviation index is determined; Based on the current charging information, determine the charging scenario weight, and based on the current charging information, the multi-dimensional deviation index, and the charging scenario weight, determine the target deviation index; The current multi-dimensional inspection index is determined based on the current charging information, and the current charging metering monitoring result of the target vehicle is determined based on the target deviation index and the current multi-dimensional inspection index.
[0102] Furthermore, the logical instructions in the aforementioned memory 330 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.
[0103] On the other hand, this application also provides a computer program product, which includes a computer program that can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer is able to perform the monitoring provided by the methods described above, the method including: Obtain the target vehicle's historical and current charging information; Based on the historical charging information and the current charging information, the historical average charging amount and the predicted charging amount are determined, and the target charging amount baseline is determined based on the baseline weight, the historical average charging amount and the predicted charging amount. Based on the current charging information and the target charging baseline, a multi-dimensional deviation index is determined; Based on the current charging information, determine the charging scenario weight, and based on the current charging information, the multi-dimensional deviation index, and the charging scenario weight, determine the target deviation index; The current multi-dimensional inspection index is determined based on the current charging information, and the current charging metering monitoring result of the target vehicle is determined based on the target deviation index and the current multi-dimensional inspection index.
[0104] In another aspect, this application also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, is implemented to perform the monitoring methods provided by the methods described above, the method comprising: Obtain the target vehicle's historical and current charging information; Based on the historical charging information and the current charging information, the historical average charging amount and the predicted charging amount are determined, and the target charging amount baseline is determined based on the baseline weight, the historical average charging amount and the predicted charging amount. Based on the current charging information and the target charging baseline, a multi-dimensional deviation index is determined; Based on the current charging information, determine the charging scenario weight, and based on the current charging information, the multi-dimensional deviation index, and the charging scenario weight, determine the target deviation index; The current multi-dimensional inspection index is determined based on the current charging information, and the current charging metering monitoring result of the target vehicle is determined based on the target deviation index and the current multi-dimensional inspection index.
[0105] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.
[0106] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.
[0107] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application.
Claims
1. A monitoring method, characterized in that, The method includes: Obtain the target vehicle's historical and current charging information; Based on the historical charging information and the current charging information, the historical average charging amount and the predicted charging amount are determined, and the target charging amount baseline is determined based on the baseline weight, the historical average charging amount and the predicted charging amount. Based on the current charging information and the target charging baseline, a multi-dimensional deviation index is determined; Based on the current charging information, determine the charging scenario weight, and based on the current charging information, the multi-dimensional deviation index, and the charging scenario weight, determine the target deviation index; The current multi-dimensional inspection index is determined based on the current charging information, and the current charging metering monitoring result of the target vehicle is determined based on the target deviation index and the current multi-dimensional inspection index.
2. The monitoring method according to claim 1, characterized in that, The step of determining the historical average charging amount and the predicted charging amount based on the historical charging information and the current charging information includes: Based on the current charging information, determine the target state of charge (SOC) range; Based on the historical charging information, determine the historical average charging amount for the target SOC range; Based on the historical charging information, the predicted charging amount for the target SOC range is determined.
3. The monitoring method according to claim 1, characterized in that, The step of determining the target charging baseline based on the baseline weight, the historical average charging amount, and the predicted charging amount includes: Obtain the battery usage time of the target vehicle; If the battery usage time is less than a first duration threshold, the baseline weight is a first weight value; the target charging amount baseline is determined based on the first weight value, the historical average charging amount, and the predicted charging amount. If the battery usage time is greater than or equal to a first duration threshold, the baseline weight is a second weight value; the target charging amount baseline is determined based on the second weight value, the historical average charging amount, and the predicted charging amount. Wherein, the first weight value is greater than the second weight value.
4. The monitoring method according to any one of claims 1 to 3, characterized in that, The current charging information includes the current cumulative battery level, the current SOC change rate, the current output power, and the charging time. The step of determining the multi-dimensional deviation index based on the current charging information and the target charging baseline includes: The power deviation index is determined based on the temperature coefficient, the current cumulative power level, and the target charging baseline. Based on the current SOC change rate and the historical average SOC change rate, determine the SOC change rate deviation index; The power stability deviation index is determined based on the current output power, the charging time, and the SOC power fitting curve. The multi-dimensional deviation index is determined based on the power deviation index, the SOC change rate index, and the power stability deviation index.
5. The monitoring method according to claim 4, characterized in that, The current charging information also includes the current temperature value; The step of determining the charging scenario weight based on the current charging information includes: Based on the current charging information, determine the target SOC range; The charging scenario characteristics are determined based on the abnormal state of the current temperature value, the fluctuation state of the current output power, and the target SOC range; The weight of the charging scenario is determined based on the correlation between the characteristics of historical anomalies and the characteristics of the charging scenario.
6. The monitoring method according to any one of claims 1 to 3, characterized in that, The current charging information includes the current cumulative battery level, current charging cost, and current temperature value; The step of determining the current multi-dimensional inspection index based on the current charging information, and determining the current charging metering monitoring result of the target vehicle based on the target deviation index and the current multi-dimensional inspection index, includes: The physical inspection deviation index is determined based on the ampere-hour integral values of the output voltage and output current and the current cumulative charge. The cost deviation index is determined based on the historical average charging unit price, the current charging cost, and the current cumulative electricity consumption. The temperature test deviation index is determined based on the temperature threshold and the current temperature value; The current multi-dimensional inspection index is determined based on the physical inspection deviation index, the cost inspection deviation index, and the temperature inspection deviation index. Based on the target deviation index and the current multi-dimensional test index, anomaly category labels and anomaly confidence levels are determined, and based on the anomaly category labels and anomaly confidence levels, the current charging metering monitoring results of the target vehicle are determined.
7. A monitoring device, characterized in that, include: The acquisition module is used to acquire the historical charging information and current charging information of the target vehicle; The first determining module is used to determine the historical average charging amount and the predicted charging amount based on the historical charging information and the current charging information, and to determine the target charging amount baseline based on the baseline weight, the historical average charging amount and the predicted charging amount. The second determining module is used to determine a multi-dimensional deviation index based on the current charging information and the target charging baseline; The third determining module is used to determine the charging scenario weight based on the current charging information, and to determine the target deviation index based on the current charging information, the multi-dimensional deviation index, and the charging scenario weight. The monitoring module is used to determine the current multi-dimensional inspection index based on the current charging information, and to determine the current charging metering monitoring result of the target vehicle based on the target deviation index and the current multi-dimensional inspection index.
8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the steps of the monitoring method as described in any one of claims 1-6.
9. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the monitoring method as described in any one of claims 1-6.
10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the monitoring method as described in any one of claims 1-6.