Charging method and system of intelligent charging pile

Abstract

The application relates to the technical field of charging piles, and discloses a charging method and system of an intelligent charging pile, which comprises the following steps: obtaining the identification code of a vehicle to be charged and traversing in a vehicle database, determining the battery life of the vehicle to be charged according to the traversal result, determining the battery life of the vehicle to be charged according to a battery damage model when the identification code does not exist in the vehicle database, verifying the charging type, reacquiring the charging type of the vehicle to be charged by using a clustering algorithm when it is judged that the charging type is incorrect, determining a target charging type, obtaining all historical charging behaviors corresponding to the target charging type, analyzing all the historical charging behaviors, determining a charging adjustment index of the charging power based on the analysis result, and charging the vehicle to be charged according to the adjusted charging power. The application improves the adaptability of charging vehicles of different brands and models and charging batteries.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of charging piles, in particular to a charging method and system of an intelligent charging pile. BACKGROUND

[0002] With the enhancement of global environmental awareness and the increasingly strict emission restrictions on traditional fuel vehicles, the number of new energy vehicles has shown explosive growth in recent years, and higher requirements have been put forward for the charging of new energy vehicles. Different brands and models of new energy vehicles have different rated voltages, currents and other electrical parameters required for charging. At present, the intelligent charging pile lacks adaptability when charging different brands and models of charging vehicles. On the other hand, due to the service life decay of the charging battery of the new energy vehicle, the charging pile cannot adaptively charge according to the use of the charging battery, which accelerates the aging of the battery and shortens the service life of the battery, which may eventually cause the failure of the vehicle charging system.

[0003] Therefore, it is necessary to design a charging method and system of an intelligent charging pile to solve the problems in the prior art. SUMMARY

[0004] In view of this, the present application provides a charging method and system of an intelligent charging pile, aiming to solve the problems of lack of adaptability when charging different brands and models of charging vehicles, and the charging pile cannot adaptively charge according to the use of the charging battery, which accelerates the aging of the battery and shortens the service life of the battery, which may eventually cause the failure of the vehicle charging system.

[0005] In one aspect, the present application provides a charging method of an intelligent charging pile, comprising:

[0006] An identification code of a vehicle to be charged is obtained and traversed in a vehicle database, and the battery life of the vehicle to be charged is determined according to the traversal result. When the identification code does not exist in the vehicle database, the battery life of the vehicle to be charged is determined according to a battery damage model.

[0007] The charging type of the vehicle to be charged is determined based on the battery life, and the charging type is verified. Whether the charging type is correct is judged based on a historical charging data set of the same charging type. When it is determined that the charging type is incorrect, the charging type of the vehicle to be charged is reacquired using a clustering algorithm, and a target charging type is determined.

[0008] The charging power is determined according to the target charging type, all historical charging behaviors corresponding to the target charging type are obtained, and all historical charging behaviors are analyzed. The charging adjustment index of the charging power is determined based on the analysis result.

[0009] adjust the charging power based on the charging adjustment index, and charge the vehicle to be charged according to the adjusted charging power.

[0010] Further, when the identification code of the vehicle to be charged is acquired and traversed in the vehicle database, and the battery life of the vehicle to be charged is determined according to the traversal result, the method comprises:

[0011] when the identification code exists in the vehicle database, determining the battery life of the vehicle to be charged according to the vehicle database;

[0012] when the identification code does not exist in the vehicle database, determining the battery life of the vehicle to be charged according to the battery damage model.

[0013] Further, when the battery life of the vehicle to be charged is determined according to the battery damage model, the method comprises:

[0014] acquiring a battery data set of the vehicle to be charged, and splitting the battery data set into a model training set and a model verification set;

[0015] pre-selecting a neural network model, iteratively training the neural network model according to the model training set, and evaluating the iteratively trained neural network model according to the model verification set;

[0016] if the ternary loss of the current iteratively trained neural network model is greater than or equal to the ternary loss of the neural network model of the previous iteration, adjusting the learning rate using cosine annealing, and continuing to iteratively train until the ternary loss is less than the ternary loss of the neural network model of the previous iteration;

[0017] if the ternary loss of the current iteratively trained neural network model is less than the ternary loss of the neural network model of the previous iteration, stopping the iteration training, and determining the current iteratively trained neural network model as the battery damage model;

[0018] determining the battery life of the vehicle to be charged according to the identification code and the battery damage model.

[0019] Further, when the charging type of the vehicle to be charged is determined based on the battery life, the method comprises:

[0020] comparing the battery life with a first preset battery life and a second preset battery life respectively, the first preset battery life being greater than the second preset battery life, and determining the charging type of the vehicle to be charged according to the comparison result;

[0021] determining the charging type of the vehicle to be charged as a fast charging mode when the battery life is greater than or equal to a first preset battery life;

[0022] determining the charging type of the vehicle to be charged as a regular charging mode when the battery life is less than the first preset battery life and greater than a second preset battery life;

[0023] determining the charging type of the vehicle to be charged as a slow charging mode when the battery life is less than or equal to the second preset battery life.

[0024] Further, when judging whether the charging type is correct based on a historical charging data set of the same charging type, the method comprises:

[0025] obtaining historical charging features from the historical charging data set, wherein the historical charging features comprise historical battery parameters and historical battery life;

[0026] determining a charging similarity based on the historical battery parameters and the historical battery life;

[0027] comparing the charging similarity with a charging similarity threshold, and determining whether the charging type is correct according to a comparison result;

[0028] when the charging similarity is greater than or equal to the charging similarity threshold, determining that the charging type is correct, and determining the charging type as a target charging type;

[0029] when the charging similarity is less than the charging similarity threshold, determining that the charging type is incorrect.

[0030] Further, when determining that the charging type is incorrect, reobtaining the charging type of the vehicle to be charged using a clustering algorithm to determine a target charging type, the method comprises:

[0031] obtaining representative data of each charging type from the historical charging data set, and establishing a clustering data set by combining the representative data with the battery life and the identification code of the vehicle to be charged, and extracting a feature vector of each data in the clustering data set;

[0032] determining an expected cluster number k as 3, initializing parameters of Gaussian distribution, calculating a probability of each data in the clustering data set belonging to each Gaussian distribution to obtain a responsibility value, selecting a cluster with the largest responsibility value as the charging type of the battery life, and determining the charging type as the target charging type.

[0033] Further, when obtaining all historical charging behaviors corresponding to the target charging type, analyzing all the historical charging behaviors, and determining a charging adjustment index of the charging power based on an analysis result, the method comprises:

[0034] analyzing all historical charging behaviors to determine power overflow behaviors, normal power behaviors and power deficiency behaviors;

[0035] constructing all power corresponding to all power overflow behaviors as a first power set;

[0036] constructing all power corresponding to all normal power behaviors as a second power set;

[0037] constructing all power corresponding to all power deficiency behaviors as a third power set;

[0038] determining a charging adjustment index of the charging power according to the first power set, the second power set and the third power set.

[0039] Further, in determining the charging adjustment index of the charging power according to the first power set, the second power set and the third power set, comprising:

[0040] obtaining a first average value in the first power set, extracting a data in the second power set as a standard value, and obtaining a third average value in the third power set;

[0041] obtaining a first difference value of the first average value and the standard value, obtaining a third difference value of the standard value and the third average value, and determining a difference ratio n of the first difference value and the third difference value;

[0042] pre-setting a first preset charging adjustment index, a second preset charging adjustment index and a third preset charging adjustment index, the first preset charging adjustment index being greater than the second preset charging adjustment index, and the second preset charging adjustment index being greater than the third preset charging adjustment index;

[0043] when n≤1, taking the first preset charging adjustment index as the charging adjustment index of the charging power;

[0044] when 1

[0045] when 1.5

[0046] Further, in adjusting the charging power based on the charging adjustment index, comprising:

[0047] the charging power and the charging adjustment index are positively correlated.

[0048] Compared with the prior art, the application has the beneficial effects that: by acquiring the identification code of the vehicle to be charged and determining the battery life in the vehicle database or using the battery damage model, the charging type is determined based thereon, the adaptability of the intelligent charging pile to different vehicle brands, models and charging batteries is improved, the accelerated aging of the charging battery due to improper charging method is avoided, the service life of the charging battery is prolonged, thereby reducing the risk of vehicle charging failure, the correctness is judged by using the historical charging data set, when incorrect, the clustering algorithm is used to reacquire, the accuracy of the target charging type is ensured, the charging power output by the intelligent charging pile is matched with the demand of the vehicle battery, the safety of the charging process is ensured, and the safety hazards such as overcharging and overheating caused by improper charging parameters are prevented. After determining the charging power according to the target charging type, the charging adjustment index is determined by analyzing the corresponding historical charging behavior, and the charging power is adjusted according to the charging adjustment index, thereby improving the charging adaptability and flexibility of the intelligent charging pile.

[0049] In another aspect, the application also provides a charging system of an intelligent charging pile, which is used for the charging system of the intelligent charging pile described above, and comprises:

[0050] A data acquisition module is configured to acquire the identification code of the vehicle to be charged and perform traversal in the vehicle database, determine the battery life of the vehicle to be charged according to the traversal result, and determine the battery life of the vehicle to be charged according to the battery damage model when the identification code does not exist in the vehicle database;

[0051] A charging judgment module is configured to determine the charging type of the vehicle to be charged based on the battery life, verify the charging type, judge whether the charging type is correct based on the historical charging data set of the same charging type, and determine the target charging type by using the clustering algorithm to reacquire the charging type of the vehicle to be charged when it is determined that the charging type is incorrect;

[0052] A charging processing module is configured to determine the charging power according to the target charging type, acquire all historical charging behaviors corresponding to the target charging type, analyze all historical charging behaviors, and determine the charging adjustment index of the target charging type based on the analysis result;

[0053] A charging adjustment module is configured to adjust the charging power based on the charging adjustment index, and charge the vehicle to be charged according to the adjusted charging power.

[0054] It can be understood that the charging method and system of the intelligent charging pile described above have the same beneficial effects, which will not be described here. BRIEF DESCRIPTION OF DRAWINGS

[0055] Various other advantages and benefits will become apparent to those of ordinary skill in the art, upon reading the following detailed description of the preferred embodiment. The accompanying drawings are included to provide a description of preferred embodiments, and are not intended to limit the scope of the application. Furthermore, the drawings are included to illustrate embodiments of the application, and, to provide a better understanding of the application, the embodiments of which will be described and illustrated in connection with the drawings. In the drawings:

[0056] Figure 1 A flow chart of a charging method of an intelligent charging pile according to an embodiment of the present application is shown in FIG. 1.

[0057] Figure 2 A functional block diagram of a charging system of an intelligent charging pile according to an embodiment of the present application is shown in FIG. 2. DETAILED DESCRIPTION

[0058] Exemplary embodiments of the present disclosure will be described more fully hereinafter with reference to the accompanying drawings; however, they are not intended to limit the scope of the present disclosure, and they are provided to fully convey the scope of the present disclosure to those skilled in the art. It should be understood that the embodiments of the present disclosure and the features thereof can be combined with each other, without conflict, and the embodiments of the present disclosure will be described in detail with reference to the accompanying drawings and the embodiments.

[0059] In some embodiments of the present application, referring to Figure 1 A charging method of an intelligent charging pile, as shown in FIG. 1, includes:

[0060] S100: An identification code of a vehicle to be charged is obtained, and a vehicle database is traversed, the battery life of the vehicle to be charged is determined according to the traversal result, and when the identification code does not exist in the vehicle database, the battery life of the vehicle to be charged is determined according to a battery damage model;

[0061] S200: The charging type of the vehicle to be charged is determined based on the battery life, the charging type is verified, and whether the charging type is correct is determined based on a historical charging data set of the same charging type, when it is determined that the charging type is incorrect, the charging type of the vehicle to be charged is reacquired using a clustering algorithm, and a target charging type is determined;

[0062] S300: The charging power is determined according to the target charging type, all historical charging behaviors corresponding to the target charging type are obtained, and all historical charging behaviors are analyzed, and a charging adjustment index of the charging power is determined based on the analysis result;

[0063] S400: The charging power is adjusted based on the charging adjustment index, and the vehicle to be charged is charged according to the adjusted charging power.

[0064] Specifically, the Vehicle Identifier (VIN) is a unique code that identifies the vehicle to be charged. This VIN is determined by the new energy vehicle manufacturer, and the vehicle database is built based on a big data model. It includes information such as the manufacturer, model, production date, battery type, and battery specifications for various vehicles. By traversing the VIN in the vehicle database, if a matching record is found, the battery life can be determined directly based on the current usage of the vehicle. For example, the battery manufacturer's instruction manual may state that the expected lifespan of the rechargeable battery is 5 years, and that the battery capacity is expected to decrease by 10%-15% after each production date. If the vehicle database has no relevant record, it may be due to missing data, the vehicle being charged being a new model / niche brand, or the battery information not being entered into the database in a timely manner. In this case, the battery life is inferred based on a battery damage model, laying the foundation for subsequent charging strategy development. The charging type for the vehicle to be charged is determined based on the battery life and then verified. By analyzing historical charging datasets of the same charging type—representing charging datasets for vehicles of the same type—the charging type is compared with historical datasets to determine its accuracy. If incorrect, a clustering algorithm is used to re-analyze and obtain the accurate charging type, thus determining the target charging type. This ensures the reliability and adaptability of the target charging type, enabling the development of charging plans based on the actual condition of the battery. The charging power is determined according to the target charging type, specifically for the charging pile. For example, in fast charging mode, the charging power can reach 50-150kW. By collecting and analyzing all historical charging behaviors corresponding to the target charging type, this data is analyzed to determine a charging power adjustment index. This index comprehensively considers various factors in the past charging process and reflects the charging adjustment strategy for that charging power. Based on the charging adjustment index, the charging power is adjusted, and the charging pile charges the vehicle according to the adjusted charging power. This achieves adaptive charging for different vehicles using intelligent charging piles.

[0065] Understandably, by determining the battery life of the vehicle to be charged through identification codes and battery damage models, and then matching the charging type, the adaptability of smart charging piles to different brands and models of new energy vehicles is ensured. Furthermore, by verifying the charging type based on historical charging datasets, the reliability and stability of the target charging type are ensured. This can maximize the lifespan of the charging battery while maintaining charging efficiency. Secondly, by continuously optimizing the charging strategy through data-driven methods, the adaptability of smart charging piles is continuously improved with the accumulation of data and the optimization of algorithms, thereby reducing the probability of vehicle charging system failures.

[0066] In some embodiments of this application, when obtaining the identification code of the vehicle to be charged and traversing the vehicle database, and determining the battery life of the vehicle to be charged based on the traversal results, the process includes: when the identification code exists in the vehicle database, determining the battery life of the vehicle to be charged based on the vehicle database; when the identification code does not exist in the vehicle database, determining the battery life of the vehicle to be charged based on a battery damage model.

[0067] Specifically, in determining the battery life of a vehicle to be charged, when the vehicle's identification code exists in the vehicle database, the system can directly retrieve data from the database to accurately obtain key information such as battery usage time and health status. Based on this data, the battery life is determined, providing strong support for the formulation of subsequent charging solutions. Conversely, when the vehicle database lacks a corresponding identification code, the battery damage model analyzes the characteristic parameters of the charging battery to predict its lifespan. This effectively avoids blind spots caused by a lack of data, ensuring that every vehicle to be charged receives an effective battery lifespan. This enhances the universality of smart charging piles in managing charging for vehicles, laying the foundation for optimizing charging strategies, ensuring battery safety, and extending battery lifespan.

[0068] In some embodiments of this application, determining the battery life of a vehicle to be charged based on a battery damage model includes: acquiring a battery dataset of the vehicle to be charged; splitting the battery dataset into a model training set and a model validation set; pre-selecting a neural network model; iteratively training the neural network model based on the model training set; evaluating the iteratively trained neural network model based on the model validation set; if the ternary loss of the neural network model after the current iterative training is greater than or equal to the ternary loss of the neural network model after the previous iterative training, using cosine annealing to adjust the learning rate and continuing iterative training until the ternary loss is less than the ternary loss of the neural network model after the previous iterative training; if the ternary loss of the neural network model after the current iterative training is less than the ternary loss of the neural network model after the previous iterative training, stopping the iterative training, and determining the currently iteratively trained neural network model as the battery damage model; and determining the battery life of the vehicle to be charged based on the identification code and the battery damage model.

[0069] Specifically, the battery dataset includes data such as battery type (e.g., lithium-ion battery, lithium iron phosphate battery), rated capacity, rated voltage, initial health state, real-time voltage change data, depth of charge / discharge (DOD), and cumulative charge / discharge cycle count. The battery dataset is split into a model training set and a model validation set, with a preferred split ratio of 3:2. This ensures that both the model training and validation sets contain a variety of data to improve the model's generalization ability. The model training set is used to train the neural network model, while the model validation set is used to evaluate the performance of the trained model. The neural network model contains multiple layers, each responsible for extracting features at different levels. In the neural network model, activation functions are used to perform nonlinear transformations, thereby helping the neural network model capture complex patterns in the data.

[0070] Understandably, the ternary loss mechanism allows the neural network to bring similar data in the training set closer together and push dissimilar data further apart. If the ternary loss of the neural network model after the current iteration is less than that after the previous iteration, cosine annealing is used to adjust the model's learning rate. The process of adjusting the learning rate using cosine annealing is lengthy and well-established, and will not be described in detail here. By adjusting the model's learning rate and continuing iterative training, in each iteration, the model attempts to learn patterns and relationships in the input data to improve its prediction or validation capabilities until the ternary loss is less than that after the previous iteration. If the ternary loss of the neural network model after the current iteration is less than that after the previous iteration, iterative training stops. At this point, the neural network model has stabilized and is close to the global optimum. This model is identified as the battery damage model, and the battery life of the vehicle to be charged is determined by substituting the identification code into the battery damage model, thus improving the reliability of battery life.

[0071] In some embodiments of this application, determining the charging type of a vehicle to be charged based on battery life includes: comparing the battery life with a first preset battery life and a second preset battery life respectively, wherein the first preset battery life is greater than the second preset battery life; determining the charging type of the vehicle to be charged based on the comparison result; when the battery life is greater than or equal to the first preset battery life, determining the charging type of the vehicle to be charged as fast charging mode; when the battery life is less than the first preset battery life but greater than the second preset battery life, determining the charging type of the vehicle to be charged as normal mode; and when the battery life is less than or equal to the second preset battery life, determining the charging type of the vehicle to be charged as slow charging mode.

[0072] Specifically, when the battery life is in good condition (greater than or equal to the first preset battery life), the charging type is determined to be fast charging mode, which shortens the charging waiting time, ensuring charging efficiency without causing additional damage to the battery. If the battery life is at a medium level (less than the first preset battery life but greater than the second preset battery life), the charging type is determined to be normal charging mode, which completes the charging task within a reasonable charging time, avoiding accelerated battery aging due to excessive pursuit of charging speed, thereby extending battery life. When the battery life is low (less than or equal to the second preset battery life), the charging type is determined to be slow charging mode. Slow charging mode reduces the degree of chemical reaction inside the battery, effectively protecting the battery's lifespan. Dynamically determining the charging type of the vehicle to be charged based on battery life effectively balances charging efficiency and battery protection, thereby improving the adaptability and stability of the smart charging station.

[0073] In some embodiments of this application, when determining whether the charging type is correct based on a historical charging dataset of the same charging type, the method includes: obtaining historical charging features based on the historical charging dataset, the historical charging features including historical battery parameters and historical battery life; determining charging similarity based on historical battery parameters and historical battery life; comparing the charging similarity with a charging similarity threshold; determining whether the charging type is correct based on the comparison result; when the charging similarity is greater than or equal to the charging similarity threshold, determining that the charging type is correct and identifying the charging type as the target charging type; when the charging similarity is less than the charging similarity threshold, determining that the charging type is incorrect.

[0074] Specifically, the historical charging dataset represents a charging dataset for vehicles of the same type awaiting charging. Based on this dataset, the correctness of the charging type is determined. Data-driven analysis allows for in-depth analysis of the operational patterns and characteristics of charging batteries under similar charging scenarios. Historical battery parameters include historical discharge current and voltage. The charging similarity is determined by comparing historical battery parameters and lifespan with current battery parameters and lifespan. Charging similarity can be determined using methods such as cosine similarity and Euclidean distance, with a preferred threshold of 0.8. When the charging similarity is less than the threshold, the charging type is directly determined to be incorrect. When the charging similarity is greater than or equal to the threshold, the current charging type is determined to be correct and identified as the target charging type. This ensures the adaptability of the charging strategy to the battery state. Furthermore, continuous accumulation and utilization of historical charging datasets helps to continuously optimize the charging type judgment logic, thus forming a virtuous cycle, improving the adaptability of smart charging piles, and simultaneously increasing the utilization efficiency of charging resources.

[0075] In some embodiments of this application, when the charging type is determined to be incorrect, a clustering algorithm is used to re-obtain the charging type of the vehicle to be charged. When determining the target charging type, the following steps are taken: obtaining representative data for each charging type based on historical charging datasets, combining the representative data with the battery life and identification code of the vehicle to be charged to establish a clustering dataset, extracting the feature vector of each data in the clustering dataset, determining the expected number of clusters k to be 3, initializing the parameters of the Gaussian distribution, calculating the probability that each data in the clustering dataset belongs to each Gaussian distribution, obtaining the responsibility value, selecting the cluster with the largest responsibility value as the charging type of the battery life, and determining this charging type as the target charging type.

[0076] Specifically, representative data is extracted from historical charging datasets, and clustering datasets are constructed by combining battery life and identification codes, fully utilizing the correlation and features of the data. By setting the expected number of clusters to 3 and optimizing the Gaussian distribution parameters, a responsibility value is obtained by calculating the probability that data belongs to each Gaussian distribution. Based on this, the cluster with the highest responsibility value is selected to determine the target charging type. This achieves dynamic and accurate identification of charging types, avoiding errors caused by relying solely on preliminary judgments and improving the accuracy of charging type identification. The Gaussian mixture model allows for automatic adjustment of classification parameters based on the natural distribution of data, thus adapting to different charging types and data patterns, further enhancing the adaptability of smart charging piles to charging different brands and models of vehicles and rechargeable batteries.

[0077] In some embodiments of this application, when acquiring all historical charging behaviors corresponding to the target charging type, parsing all historical charging behaviors, and determining the charging adjustment index of the charging power based on the parsing results, the process includes: parsing all historical charging behaviors to determine power overflow behaviors, normal power behaviors, and power insufficiency behaviors; constructing all power corresponding to all power overflow behaviors into a first power set; constructing all power corresponding to all normal power behaviors into a second power set; constructing all power corresponding to all power insufficiency behaviors into a third power set; and determining the charging adjustment index of the charging power based on the first power set, the second power set, and the third power set.

[0078] In some embodiments of this application, determining the charging adjustment index of charging power based on a first power set, a second power set, and a third power set includes: obtaining a first average value in the first power set; extracting a data point from the second power set as a standard value; obtaining a third average value in the third power set; obtaining a first difference between the first average value and the standard value; obtaining a third difference between the standard value and the third average value; determining the ratio n of the difference between the first difference and the third difference; and presetting a first preset charging adjustment index, a second preset charging adjustment index, and a third preset charging adjustment index, wherein the first preset charging adjustment index is greater than the second preset charging adjustment index, the second preset charging adjustment index is greater than the third preset charging adjustment index, and when n ≤ 1, the first preset charging adjustment index is used as the charging adjustment index of charging power; when 1 < n ≤ 1.5, the second preset charging adjustment index is used as the charging adjustment index of charging power; and when 1.5 < n, the third preset charging adjustment index is used as the charging adjustment index of charging power.

[0079] In some embodiments of this application, when adjusting the charging power based on the charging adjustment index, the charging power and the charging adjustment index are positively correlated.

[0080] Specifically, after determining the charging power for the target charging type, due to the influence of the environment and electromagnetic fields, the charging power of smart charging piles will fluctuate even when charging according to the specified power. Power overflow behavior, normal power behavior, and insufficient power behavior are determined based on the smart charging pile's performance for the target charging type. For example, in fast charging mode, the charging power can reach 50-150kW. Therefore, normal power behavior is 50-150kW, power overflow behavior is greater than 150kW, and insufficient power behavior is less than 50kW. By analyzing all historical charging behaviors and dividing them into corresponding power sets, the reliability of obtaining the charging adjustment index is ensured. When n≤1, it indicates that insufficient power behavior is more frequent, requiring a larger charging adjustment index. When 1 < n ≤ 1.5, it indicates that power overflow behavior is gradually increasing, requiring a smaller charging adjustment index. When 1.5 < n, it indicates that power overflow behavior is also increasing, requiring an even smaller charging adjustment index. The charging adjustment index is dynamically determined based on the difference ratio, realizing intelligent adjustment of charging power. Since charging power and charging adjustment index are positively correlated, when the charging adjustment index is selected due to power overflow behavior and its value is small, the charging power decreases accordingly, avoiding irreversible damage to the rechargeable battery caused by excessive charging power, thereby effectively extending the battery's lifespan. When the charging adjustment index is selected due to insufficient power behavior and its value is large, the charging power increases accordingly, quickly compensating for insufficient power, thereby reducing charging waiting time and improving charging efficiency.

[0081] Understandably, the positive correlation mechanism enables smart charging piles to intelligently and accurately adjust the charging power according to real-time conditions, which not only ensures the safety and stability of the charging process, but also ensures adaptability to charging different brands and models of vehicles and rechargeable batteries.

[0082] In summary, the beneficial effects of this invention are as follows: By obtaining the identification code of the vehicle to be charged and determining the battery life through a vehicle database or by using a battery damage model, the charging type is determined accordingly. This improves the adaptability of smart charging piles to charging different vehicle brands, models, and batteries, avoiding accelerated battery aging due to inappropriate charging methods, extending battery life, and reducing the risk of vehicle charging failures. The accuracy of the historical charging dataset is verified; if incorrect, a clustering algorithm is used to re-acquire the data, ensuring the accuracy of the target charging type. This allows the charging power output of the smart charging pile to match the needs of the vehicle battery, ensuring the safety of the charging process and preventing safety hazards such as overcharging and overheating caused by improper charging parameters. After determining the charging power based on the target charging type, a charging adjustment index is determined by analyzing corresponding historical charging behavior, and the charging power is adjusted accordingly, improving the charging adaptability and flexibility of the smart charging pile.

[0083] In another preferred embodiment based on the above embodiments, see [reference] Figure 2 As shown, this embodiment provides a charging system for a smart charging pile, used to apply the charging method of the aforementioned smart charging pile, including:

[0084] The data acquisition module is configured to acquire the identification code of the vehicle to be charged and traverse the vehicle database. Based on the traversal results, the battery life of the vehicle to be charged is determined. When the identification code is not found in the vehicle database, the battery life of the vehicle to be charged is determined based on the battery damage model.

[0085] The charging judgment module is configured to determine the charging type of the vehicle to be charged based on battery life, verify the charging type, and judge whether the charging type is correct based on the historical charging dataset of the same charging type. When the charging type is determined to be incorrect, a clustering algorithm is used to re-obtain the charging type of the vehicle to be charged and determine the target charging type.

[0086] The charging processing module is configured to determine the charging power according to the target charging type, acquire all historical charging behaviors corresponding to the target charging type, analyze all historical charging behaviors, and determine the charging adjustment index of the charging power based on the analysis results.

[0087] The charging adjustment module is configured to adjust the charging power based on the charging adjustment index, and charge the vehicle to be charged according to the adjusted charging power.

[0088] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program goods. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program goods embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0089] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program goods according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0090] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0091] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0092] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the specific implementation of the present invention. Any modifications or equivalent substitutions that do not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. A charging method for a smart charging pile, characterized in that, include: The identification code of the vehicle to be charged is obtained and traversed in the vehicle database. The battery life of the vehicle to be charged is determined based on the traversal results. When the identification code is not found in the vehicle database, the battery life of the vehicle to be charged is determined based on the battery damage model. The charging type of the vehicle to be charged is determined based on the battery life. The charging type is then verified. The correctness of the charging type is determined based on a historical charging dataset of the same charging type. If the charging type is determined to be incorrect, a clustering algorithm is used to re-obtain the charging type of the vehicle to be charged and determine the target charging type. The charging power is determined according to the target charging type, all historical charging behaviors corresponding to the target charging type are obtained, and all historical charging behaviors are analyzed. Based on the analysis results, the charging adjustment index of the charging power is determined. The charging power is adjusted based on the charging adjustment index, and the vehicle to be charged is charged according to the adjusted charging power. When acquiring all historical charging behaviors corresponding to the target charging type, analyzing all historical charging behaviors, and determining the charging power adjustment index based on the analysis results, the process includes: Analyze all historical charging behaviors to identify power overflow behavior, normal power behavior, and power under-power behavior. Construct the first power set by all the power corresponding to all power overflow behaviors; Construct a second power set by representing all the power corresponding to all normal power behaviors; Construct a third power set from all the power corresponding to all power-deficient behaviors; The charging adjustment index of the charging power is determined based on the first power set, the second power set, and the third power set; When determining the charging adjustment index of the charging power based on the first power set, the second power set, and the third power set, the following steps are included: Obtain the first average value from the first power set, extract one data point from the second power set as a standard value, and obtain the third average value from the third power set; Obtain the first difference between the first average value and the standard value, obtain the third difference between the standard value and the third average value, and determine the difference ratio n between the first difference and the third difference; A first preset charging adjustment index, a second preset charging adjustment index, and a third preset charging adjustment index are preset, wherein the first preset charging adjustment index is greater than the second preset charging adjustment index, and the second preset charging adjustment index is greater than the third preset charging adjustment index. When n≤1, the first preset charging adjustment index is used as the charging adjustment index of the charging power; When 1 < n ≤ 1.5, the second preset charging adjustment index is used as the charging adjustment index of the charging power; When 1.5 < n, the third preset charging adjustment index is used as the charging adjustment index of the charging power; When adjusting the charging power based on the charging adjustment index, the following is included: The charging power and the charging adjustment index are positively correlated.

2. The charging method for the intelligent charging pile according to claim 1, characterized in that, When obtaining the identification code of the vehicle to be charged and traversing the vehicle database, and determining the battery life of the vehicle to be charged based on the traversal results, the process includes: When the identification code exists in the vehicle database, the battery life of the vehicle to be charged is determined according to the vehicle database. When the identification code is not found in the vehicle database, the battery life of the vehicle to be charged is determined according to the battery damage model.

3. The charging method for the intelligent charging pile according to claim 2, characterized in that, Determining the battery life of the vehicle to be charged based on a battery damage model includes: Obtain the battery dataset of the vehicle to be charged, and split the battery dataset into a model training set and a model validation set; A neural network model is pre-selected, and the neural network model is iteratively trained based on the model training set. The iteratively trained neural network model is then evaluated based on the model validation set. If the ternary loss of the neural network model after the current iteration is greater than or equal to the ternary loss of the neural network model after the previous iteration, cosine annealing is used to adjust the learning rate, and iterative training continues until the ternary loss is less than the ternary loss of the neural network model after the previous iteration. If the ternary loss of the neural network model after the current iteration is less than the ternary loss of the neural network model after the previous iteration, then stop the iterative training and determine the neural network model after the current iteration as the battery damage model. The battery life of the vehicle to be charged is determined based on the identification code and the battery damage model.

4. The charging method for the intelligent charging pile according to claim 3, characterized in that, When determining the charging type of the vehicle to be charged based on the battery life, the following steps are included: The battery life is compared with a first preset battery life and a second preset battery life, respectively. The first preset battery life is greater than the second preset battery life. The charging type of the vehicle to be charged is determined based on the comparison results. When the battery life is greater than or equal to the first preset battery life, the charging type of the vehicle to be charged is determined to be fast charging mode. When the battery life is less than the first preset battery life but greater than the second preset battery life, the charging type of the vehicle to be charged is determined to be the normal mode. When the battery life is less than or equal to the second preset battery life, the charging type of the vehicle to be charged is determined to be slow charging mode.

5. The charging method for the intelligent charging pile according to claim 4, characterized in that, When determining whether the charging type is correct based on a historical charging dataset of the same charging type, the following is included: Historical charging features are obtained based on the historical charging dataset, including historical battery parameters and historical battery life. The charging similarity is determined based on the historical battery parameters and historical battery life. The charging similarity is compared with the charging similarity threshold, and the charging type is determined to be correct based on the comparison result. When the charging similarity is greater than or equal to the charging similarity threshold, the charging type is determined to be correct, and the charging type is identified as the target charging type. When the charging similarity is less than the charging similarity threshold, the charging type is determined to be incorrect.

6. The charging method for the intelligent charging pile according to claim 5, characterized in that, When the charging type is determined to be incorrect, a clustering algorithm is used to re-obtain the charging type of the vehicle to be charged. Determining the target charging type includes: Based on the historical charging dataset, representative data for each charging type is obtained, and the representative data is combined with the battery life and identification code of the vehicle to be charged to establish a clustering dataset. The feature vector of each data in the clustering dataset is then extracted. The desired number of clusters k is determined to be 3, and the parameters of the Gaussian distribution are initialized. The probability of each data point in the clustered dataset belonging to each Gaussian distribution is calculated to obtain the responsibility value. The cluster with the largest responsibility value is selected as the charging type for the battery life again, and this charging type is determined as the target charging type.

7. A charging system for a smart charging pile, used for applying the charging method of the smart charging pile as described in any one of claims 1-6, characterized in that, include: The data acquisition module is configured to acquire the identification code of the vehicle to be charged and traverse the vehicle database, determine the battery life of the vehicle to be charged based on the traversal results, and determine the battery life of the vehicle to be charged based on the battery damage model when the identification code is not found in the vehicle database. The charging determination module is configured to determine the charging type of the vehicle to be charged based on the battery life, verify the charging type, determine whether the charging type is correct based on a historical charging dataset of the same charging type, and when the charging type is determined to be incorrect, use a clustering algorithm to re-obtain the charging type of the vehicle to be charged and determine the target charging type. The charging processing module is configured to determine the charging power according to the target charging type, acquire all historical charging behaviors corresponding to the target charging type, analyze all historical charging behaviors, and determine the charging adjustment index of the charging power based on the analysis results. The charging adjustment module is configured to adjust the charging power based on the charging adjustment index, and charge the vehicle to be charged according to the adjusted charging power.

Images