Charging plan generation method, electronic device and storage medium

Abstract

The present application discloses a charging plan generation method, an electronic device and a storage medium. The method comprises: acquiring historical charging data; determining charging features on the basis of the historical charging data; determining charging prediction data on the basis of the charging features and a preset charging prediction model; and generating a charging plan on the basis of the historical charging data and the charging prediction data. By mining and analyzing the historical charging data, the charging features representing a charging action of a user is obtained, and a charging demand of the user is predicted on the basis of the charging features. In the process, personal privacy information of the user is not involved, and the blockage problem of difficulty in obtaining the charging demand of the user can be solved. Furthermore, by predicting user preferences, users are divided into groups, and diversified and targeted charging plans are generated for different user groups, solving the problem that existing charging strategies are difficult to meet the charging demands of different users.

Description

Charging plan generation method, electronic device and storage medium

[0001] This application claims priority to the Chinese patent application filed with the China Patent Office on August 16, 2023, with application number 202311038005.9 and application name “Method for generating charging plan, electronic device and storage medium”, the entire contents of which are incorporated by reference into this application. Technical Field

[0002] The present application relates to the field of charging technology, and in particular to a method for generating a charging plan, an electronic device, and a storage medium. Background Art

[0003] As the new energy vehicle industry develops rapidly, the deployment, construction and safe and rational use of charging stations have gradually become the focus of the electric vehicle industry.

[0004] However, the current use of charging stations basically adopts the efficiency-first plug-and-charge method. This charging method is relatively simple and lacks the ability to generate customized charging plans based on user charging needs, which makes it difficult to meet the charging needs of different users in different scenarios.

[0005] Summary of the Invention

[0006] The present application provides a method for generating a charging plan, an electronic device, and a storage medium to solve the problem that the charging method in the prior art is relatively single and lacks the ability to generate charging plans customized to the user's charging needs, which makes it difficult to meet the charging needs of different users in different scenarios.

[0007] An embodiment of the present application provides a method for generating a charging plan, including:

[0008] Get historical charging data;

[0009] determining a charging characteristic based on the historical charging data;

[0010] Determining charging prediction data based on the charging characteristics and a preset charging prediction model;

[0011] A charging plan is generated according to the historical charging data and the charging prediction data.

[0012] Optionally, generating a charging plan based on the historical charging data and the charging prediction data includes:

[0013] Determining a target charging strategy model based on the historical charging data and the charging prediction data;

[0014] The charging plan is generated according to the target charging strategy model.

[0015] Optionally, determining a target charging strategy model based on the historical charging data and the charging prediction data includes:

[0016] Calculating a first charging cost based on a preset cost priority model according to the historical charging data and the charging prediction data;

[0017] Calculating a second charging cost based on a preset efficiency priority model according to the historical charging data and the charging prediction data;

[0018] determining whether a difference between the first charging cost and the second charging cost is greater than a preset threshold;

[0019] If it is greater than, determining that the target charging strategy model is the cost priority model;

[0020] If it is less than, the target charging strategy model is determined to be the efficiency priority model.

[0021] Optionally, the calculating the first charging cost based on a preset cost priority model according to the historical charging data and the charging prediction data includes:

[0022] generating a first charging plan based on a preset cost priority model according to the historical charging data and the charging forecast data;

[0023] The charging cost of the first charging plan is calculated to determine the first charging cost.

[0024] Optionally, generating a first charging plan based on a preset cost priority model according to the historical charging data and the charging prediction data includes:

[0025] Determining constraints of the cost-priority model based on the historical charging data and the charging forecast data;

[0026] determining an objective function of the cost-first model;

[0027] The first charging plan is generated based on an operations research optimization solver according to the objective function and the constraints.

[0028] Optionally, determining the constraint conditions of the cost priority model based on the historical charging data and the charging prediction data includes:

[0029] Determining, based on the charging prediction data, that the constraint condition includes at least one of the following:

[0030] The transmitted power is less than or equal to the predicted maximum receivable power;

[0031] When the charged capacity is lower than the predicted minimum capacity, fast charging is used; and

[0032] During the predicted charging time, the charged capacity is greater than or equal to the predicted charging capacity;

[0033] The charging prediction data includes the predicted maximum receivable power, the predicted minimum power, the predicted charging time, and the predicted charging power.

[0034] Optionally, the calculating the second charging cost based on a preset efficiency priority model according to the historical charging data and the charging prediction data includes:

[0035] generating a second charging plan based on a preset efficiency priority model according to the historical charging data and the charging prediction data;

[0036] The charging cost of the second charging plan is calculated to determine the second charging cost.

[0037] Optionally, generating a second charging plan based on a preset efficiency priority model according to the historical charging data and the charging prediction data includes:

[0038] Determining constraints of the efficiency priority model based on the historical charging data and the charging prediction data;

[0039] According to the constraint conditions, the second charging plan is generated based on a power load balancing allocation algorithm.

[0040] Optionally, determining a target charging strategy model based on the historical charging data and the charging prediction data includes:

[0041] determining a historical charging curve based on the historical charging data;

[0042] Determining the similarity between the historical charging curve and a standard charging curve;

[0043] Determining whether the similarity is greater than a preset threshold;

[0044] When the similarity is greater than the threshold, the target charging strategy model is determined.

[0045] Optionally, generating the charging plan according to the charging strategy model includes:

[0046] Determining a charging curve estimation model according to the target charging strategy model;

[0047] Determining a predicted battery state value according to the charging curve estimation model;

[0048] The charging plan is generated according to the predicted battery state value and the standard charging curve.

[0049] Optionally, determining a target charging strategy model based on the historical charging data and the charging prediction data includes:

[0050] determining a charging time distribution based on the historical charging data;

[0051] Determining a similarity between the charging time distribution and a charging peak time distribution;

[0052] Determining whether the similarity is greater than a preset threshold;

[0053] When the similarity is greater than the threshold, the target charging strategy model is determined.

[0054] Optionally, generating the charging plan according to the target charging strategy model includes:

[0055] Determining whether the charging period is the peak charging period;

[0056] If yes, determining that the charging power during the charging period is the first power;

[0057] If not, determining that the charging power of the charging period is a second power, wherein the first power is less than the second power;

[0058] The charging plan is generated according to the charging period and the charging power.

[0059] Optionally, before determining the charging prediction data according to the charging characteristics and a preset charging prediction model, the method further includes:

[0060] generating the charging prediction model;

[0061] The charging prediction model includes a minimum power prediction model and / or a charging power prediction model.

[0062] Optionally, generating a charging power prediction model includes:

[0063] Determine the maximum charging power and the average charging power of the charging characteristic as output label values;

[0064] Construct a sample dataset based on charging characteristics;

[0065] According to the sample data set, the charging power prediction model is generated based on a binary classification algorithm.

[0066] Optionally, generating a minimum power consumption prediction model includes:

[0067] Construct a sample dataset based on charging characteristics;

[0068] Determining a plurality of clusters based on a clustering algorithm according to the sample data set;

[0069] The minimum power guarantee of the cluster is determined according to the characteristics of the cluster to generate the minimum power guarantee prediction model.

[0070] An embodiment of the present 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 implements the method described in the first aspect when executing the computer program.

[0071] An embodiment of the present application provides a storage medium, wherein the storage medium stores a computer program, and when the computer program is executed by a processor, the method according to the first aspect is implemented.

[0072] The technical solution of the present application can achieve the following technical effects. This embodiment obtains charging characteristics that characterize the user's charging habits by mining and analyzing historical charging data, predicts the user's preference type based on the charging characteristics, and mines the user's charging needs, and generates a customized charging plan to meet the user's charging needs. BRIEF DESCRIPTION OF THE DRAWINGS

[0073] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments of the present invention. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0074] FIG1 is a schematic diagram of an application environment of a method for generating a charging plan in an embodiment of the present invention;

[0075] FIG2 is a flow chart of a method for generating a charging plan in an embodiment of the present application;

[0076] FIG3 is a schematic diagram of a process for generating a charging plan based on historical charging data and charging prediction data in one embodiment of the present application;

[0077] FIG4 is a time distribution statistical diagram of the charging power of a charging station in one embodiment of the present application;

[0078] FIG5 is a flow chart of a method for generating a first charging plan based on a preset cost priority model according to an embodiment of the present application;

[0079] FIG6 is a schematic diagram of the architecture of a computer device in an embodiment of the present application. DETAILED DESCRIPTION

[0080] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of them. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0081] First, to introduce the method of the present application, some abbreviations and key terms involved below are defined.

[0082] Historical charging data: data contained in the historical charging orders of the device to be charged, including charging power, charging power, charging date, charging time, charging time, charging process curve, etc.

[0083] Charging plan: A charging power schedule assigned to a charging device. Based on this charging power schedule, the charging device charges the battery at a set power at a set time.

[0084] Charging characteristics: Based on historical charging data, feature mining is performed using statistical methods to obtain representative user charging data indirectly.

[0085] Charging strategy model: Based on the different charging needs of different types of users in different scenarios, a model framework structure is generated to meet their charging needs.

[0086] Charging prediction data: This data is generated by the charging prediction model based on charging characteristics to determine the user's charging needs.

[0087] The following describes the application environment of the method for generating a charging plan provided in an embodiment of the present application.

[0088] The method provided in an embodiment of the present invention can be applied in a charging system such as Figure 1, which is a schematic diagram of the application environment of the method for generating a charging plan in one embodiment of the present application. The charging system 10 includes a charging management device 100 and a charging station 200. The charging station 200 includes at least one group 210, which includes at least one charging pile 211. The charging pile 211 includes at least one charging gun 211A. Each charging gun 211A can be paired with one user at a time to charge the user's charging device, such as a new energy electric vehicle 300.

[0089] The charging pile 211 and the charging gun 211A can provide charging outputs of different powers. Specifically, the charging management device 100 controls the power sent by the charging pile 211 and the charging gun 211A, thereby adjusting the charging power of the electric vehicle 300.

[0090] The charging management device 100 is in communication with the charging station 200 and is used to manage the charging station 200. The charging management device 100 can be composed of one or more workstations or servers, and monitors the charging station 200, charging piles 211, and charging guns 211A, collects and queries data, and processes and analyzes the data. The data that the charging management device 100 can obtain includes charging information of charging devices such as the charging piles 211 including the charging guns 211A, device information of devices being charged, such as batteries or new energy electric vehicles 300 electrically connected to the charging devices, and historical charging information corresponding to the devices being charged.

[0091] In one embodiment, a charging station 200 is a facility that provides charging services for electric vehicles 300 in cities, highways, residential areas, or commercial areas. A charging pile 211 under the charging station 200 is connected to the electric vehicle 300 via a charging gun 211A. The user of the electric vehicle 300 establishes a communication connection with the charging management device 100, which then issues a command to the charging pile 211 to charge the user's electric vehicle 300.

[0092] During this process, the charging management device 100 obtains the device identification code of the electric vehicle 300, obtains the electric vehicle 300's past charging data, and predicts the electric vehicle 300's charging preferences and current charging needs based on this historical charging data. It then generates a customized charging plan and sends the charging plan to the charging station 211, which then charges the electric vehicle 300 according to the charging plan. Without requiring access to the user's personal information, data mining is performed to obtain charging preferences and meet the user's charging needs while protecting user privacy.

[0093] It can be understood that the composition of the charging station and charging management equipment shown in Figure 1 above is only an example. In actual situations, equipment or modules can be integrated, adjusted or added as needed. For example, the charging station may also include a power supply system, which mainly provides power for the charging equipment. It can be composed of primary equipment such as switches, transformers, lines, etc. and secondary equipment such as detection, protection, control devices, etc., as well as source filtering devices, etc. This application does not impose any restrictions on this.

[0094] Based on the above-mentioned charging system, the method for generating a charging plan provided in this application can be implemented.

[0095] Please refer to FIG. 2 , which is a flow chart of a method for generating a charging plan provided in an embodiment of the present application. The power adjustment method shown in FIG. 2 may be executed by a charging management device. The method may include:

[0096] S21. Obtain historical charging data.

[0097] The user needs to establish a communication connection with the charging management device to start charging. During this process, the charging management device recognizes the device identification code of the device to be charged and uses the identification code to trace the historical charging data of the device to be charged.

[0098] Specifically in the embodiment, the process of the charging management device obtaining historical charging data can be obtained by querying the corresponding historical charging data in the local database through the device identification code, or by querying the corresponding historical charging data in the cloud server through the device identification code, or the user end can locally store or request the historical charging data from the cloud and then send it directly to the charging management device.

[0099] It should be noted that the device identification code is a unique encrypted number randomly generated by the charging management device. This number serves as the unique identifier of the device to be charged. There is no need to collect the user's personal information. The subsequent process only processes the historical charging data stream and does not involve the user's privacy issues.

[0100] S22. Determine charging characteristics based on historical charging data.

[0101] Charging characteristics are obtained by processing historical charging data using statistical methods. They include two categories: charging preference characteristics and intermediate auxiliary characteristics. Charging preference characteristics are used to characterize the charging performance parameters of the device being charged and charging parameters related to user preferences. Intermediate auxiliary characteristics, on the other hand, characterize other non-charging parameters related to user preferences. Although these non-charging parameters do not directly affect the charging process, they are influenced by user preferences, affecting the user's charging parameters and thus indirectly influencing the charging process.

[0102] In one embodiment, the charging preference parameters include average charging time, and the intermediate auxiliary features include whether the charging date is a holiday, the device's charging frequency, and the time since the last charge. When predicting the current charging time, it is necessary to consider not only the user's recent average charging time, but also whether it is a holiday, whether the user has long-distance travel needs, the time since the last charge, and whether the user needs to charge a large amount this time.

[0103] It should be noted that the specific composition of the charging preference parameters and the intermediate auxiliary features, as well as the method of obtaining charging features by mining historical charging data, will be described in detail later and will not be presented here.

[0104] S23. Determine charging prediction data based on charging characteristics and a preset charging prediction model.

[0105] After mining historical charging data to obtain charging characteristics, combined with a preset charging prediction model, the charging characteristics are used as model parameters for input to obtain output charging prediction data.

[0106] In some embodiments, the charging prediction data includes a predicted minimum charge and a predicted charging charge. The corresponding charging prediction model includes a minimum charge prediction model and a charging power prediction model. The charging prediction model is generated by training a machine learning algorithm with a training dataset. The specific training method is described in detail below and is not discussed here.

[0107] It should be noted that in some other embodiments, the charging prediction data may also include other data, such as the predicted upper power limit and predicted charging time of the device to be charged. Some charging prediction data may not use a charging prediction model, but may be directly obtained by processing the charging characteristics through statistical methods. For example, the predicted upper power limit value of the device to be charged can directly use the value of the maximum receivable power in the charging characteristics. The predicted charging time of the device to be charged can be obtained by taking a weighted average of parameters such as the gun withdrawal time, average dwell time, and average charging time in the charging characteristics.

[0108] S24. Generate a charging plan based on historical charging data and charging forecast data.

[0109] The generation of a charging plan specifically includes: determining the preference type based on historical charging data and charging forecast data, then determining the most appropriate target charging strategy model from multiple preset charging strategy models based on the preference type, and finally generating a matching charging plan based on the target charging strategy model.

[0110] In some embodiments, generating a charging plan specifically includes determining a target charging strategy model based on historical charging data and charging prediction data; and generating a charging plan based on the charging strategy model.

[0111] In the step of determining the target charging strategy model, the preference type of the user corresponding to the device to be charged is determined based on the historical charging data and the charging prediction data, and the corresponding charging strategy model is determined based on the user's preference type.

[0112] Specifically, user preferences include safety, energy conservation, price sensitivity, and efficiency sensitivity. Safety users prioritize maintaining battery health to maximize the lifespan of their devices. Energy conservation users prioritize charging during off-peak hours to maintain charging stability. Price-sensitive users prioritize charging when electricity prices are relatively low to reduce total charging costs. Efficiency-sensitive users prioritize maintaining fast charging mode to fully charge their devices, thereby improving the efficiency of their devices.

[0113] It is understandable that in some other embodiments, the user's preference type may also include other types, such as comprehensive fast charging type, balanced charging type, etc.

[0114] In some embodiments, safety-oriented users correspond to matching safety-oriented charging strategy models, energy-saving users correspond to matching energy-saving charging strategy models, price-sensitive users correspond to matching cost-priority charging strategy models, and efficiency-sensitive users correspond to matching efficiency-priority charging strategy models.

[0115] The step of generating a charging plan based on the charging strategy model includes determining corresponding constraints and objective functions based on the charging strategy model, calculating the optimal solution to achieve the extreme value of the objective function under the constraints, and generating a charging plan based on the calculation results.

[0116] In some embodiments, the safety-ensured charging strategy model, under constraints, calculates the optimal solution that brings the charging curve closer to the standard charging curve as a charging plan. The energy-saving charging strategy model, under constraints, calculates the optimal solution that avoids peak charging times as a charging plan. The cost-priority charging strategy model, under constraints, calculates the optimal solution that minimizes the total charging cost as a charging plan. The efficiency-priority charging strategy model, under constraints, calculates the optimal solution that minimizes the charging time as a charging plan.

[0117] The composition and determination method of the charging characteristics in the embodiments of the present application are described in detail below.

[0118] Specifically, the charging characteristics include a charging preference characteristic and an intermediate auxiliary characteristic.

[0119] In one embodiment, the charging preference characteristics include maximum charging capacity, average charging capacity, acceptable maximum charging power, average charging time, mode of charging unplugging time, and average dwell time.

[0120] More specifically, the maximum charging power is based on the charged power of each order, and the maximum value of the charged power in all historical orders is taken as the maximum charging power.

[0121] The average charging capacity is also based on the charged capacity of each order. Three time windows are introduced: the last 7 days, the last 30 days, and all historical charging time. The average charged capacity of the charging orders in the three time windows is taken as the average to obtain the average charging capacity.

[0122] It can be understood that the most representative characteristic of the device to be charged in the three time windows is the average charging power of the last 7 days, followed by the average charging power of the last 30 days, and finally the average charging power of all historical charging times. When using the average charging power as a charging feature, if there is charging data for the last 7 days, the average charging power of the last 7 days is used first. If there is no charging data for the last 7 days, the average charging power of the last 30 days is used as the charging feature, and so on.

[0123] In some other embodiments, before averaging the charging orders within a time window to obtain the average charged capacity, data cleaning is required to ensure data reliability. For example, the charged capacity of all charging orders within the time window is sorted by capacity, and the charged capacities before the 5th percentile and after the 95th percentile in the sequence are removed as abnormal values. The remaining charged capacities after removing the abnormal values ​​are then averaged.

[0124] The maximum acceptable charging power is based on the maximum charging power of each order. Specifically, because the maximum charging power of AC and DC charging piles is completely different, it is necessary to group them into groups according to the maximum charging power of AC and DC charging piles, and use the maximum charging power as the maximum charging power of AC and DC charging piles, respectively.

[0125] It should be noted that when you subsequently use the maximum charging power to generate a charging plan, you need to first determine whether the charging equipment is an AC charging pile or a DC charging pile, and select the corresponding maximum charging power based on the type of charging pile.

[0126] It should also be noted that in some other embodiments, the maximum charging current can also be used as a charging feature for mining and processing. The specific process is similar to the maximum charging power. Those skilled in the art can know the feature mining process of the maximum charging current based on the above-mentioned mining and processing process of the maximum charging power, so it will not be repeated here.

[0127] The average charging time is calculated based on the start and end time of the charging of the order, and the charging time of the order is averaged to obtain the average charging time.

[0128] It can be understood that in some other embodiments, the average charging time is also set with multiple time windows, and the charging time of orders in the corresponding time windows is averaged according to different time windows to serve as the average charging time, which is not limited here.

[0129] It can also be understood that in some other embodiments, before taking the average of the charging time of the order, data cleaning is required, such as eliminating some data outliers, etc., which is also not limited here.

[0130] The charging gun removal time is different from the charging time. After the charging device is charged, the gun is often not removed and disconnected immediately. Instead, it is removed and disconnected when the user needs to use the device. Therefore, the charging gun removal time can be used to represent the time the device stays in the charging station, thereby helping to predict the charging time.

[0131] Specifically, the charging gun unplugging time is obtained by rounding the unplugging time of each historical charging order to 30 minutes as the minimum unit, and then taking the mode as the charging gun unplugging time.

[0132] The average stay time is calculated based on the start time and draw time of each historical order. The average stay time of all historical orders is taken as the average.

[0133] It can be understood that, in some embodiments, the average residence time includes calculation in time windows and data cleaning before calculating the average value.

[0134] In one embodiment, the intermediate auxiliary features include parameters such as whether it is a holiday, the average number of temporary gun draws per day, the time since the last charge, the charging frequency, the last charged power, and whether the last charge was fully charged.

[0135] Specifically, whether it is a holiday includes whether the day before charging is a holiday, whether the day is a holiday, and whether tomorrow is a holiday, which is used as a parameter of the minimum power prediction model and the charging power prediction model.

[0136] The average daily number of temporary gun removals was obtained by statistically analyzing the gun insertion and removal times of daily charging orders in history.

[0137] Specifically in the embodiment, if the charging time of a single order in a day is less than 30 minutes, and the unplugging time of the order is less than 10 minutes from the plugging time of the next order on the same day, it is judged that a temporary unplugging behavior has occurred in the order, and the average daily number of temporary unplugging times is obtained by dividing the number of orders with temporary unplugging behavior by the total number of historical charging days.

[0138] The charging frequency is calculated by dividing the number of historical charging orders by the historical charging time. In the embodiment, the charging frequency is also set with a time window, such as the last 7 days, the last 30 days, and the total historical charging time, and the charging frequency is calculated according to the corresponding time window.

[0139] The time since the last charge is in days, which counts the time since the last charge of the device to be charged.

[0140] The last charging power statistics include the charging power of the most recent charging order of the device to be charged.

[0141] Whether the last charging reached a full charge state Statistics whether the device to be charged reached a full charge state in the most recent charging order.

[0142] Specifically, since the charging status of the device to be charged cannot be obtained in real time, the reported power of the most recent charging order is analyzed to determine whether it has reached a full charge state.

[0143] Specifically in the embodiment, if the reported power continuously decays to 0 before the end of the charging time, it is determined that the device to be charged has reached a fully charged state.

[0144] In some other embodiments, the judgment can also be made by analyzing the user's behavior. For example, if the charging end time of the device to be charged is less than 1 hour from the time the gun is drawn, then it is determined that the device to be charged has reached a fully charged state.

[0145] The preset charging prediction model mentioned in the above embodiment is introduced in detail below.

[0146] In some embodiments, the charging prediction model includes a minimum power prediction model and / or a charging power prediction model.

[0147] The method for generating the charging power prediction model includes the following steps:

[0148] Determine the maximum charging power and average charging power of the charging characteristics as output tags;

[0149] Construct a sample dataset based on charging characteristics;

[0150] According to the sample data set, a charging capacity prediction model is generated based on the binary classification algorithm.

[0151] Specifically, in the step of determining the maximum charging power and average charging power of the charging characteristics as output labels, since the charging power prediction model uses a classification algorithm rather than a clustering algorithm, the output value is a preset label value, and the classification label value needs to be set before training the model.

[0152] Specifically in this embodiment, the maximum charge capacity in the charging feature is determined to be a label value of 1, and the average charge capacity is determined to be a label value of 0. When the classification result is 1, the maximum charge capacity is determined as the predicted charge capacity; when the classification result is 0, the average charge capacity is determined as the predicted charge capacity.

[0153] It should be noted that when determining the average charging capacity as the predicted charging capacity, the average charging capacity is determined according to the priority of the aforementioned time window. For example, if the average charging capacity of the last 7 days is available, the average charging capacity of the last 7 days is used as the predicted charging capacity. Otherwise, the average charging capacity of the last 30 days is used as the predicted charging capacity, and so on. Finally, the average charging capacity of all time periods in the historical orders is used as the predicted charging capacity.

[0154] The step of constructing a sample data set according to charging characteristics specifically includes the following steps:

[0155] Calculate the order deviation value, determine the sample type based on the order deviation value, determine the parameters of the sample data, and construct a sample data set based on the sample type and parameters.

[0156] In the step of calculating the order deviation value, for each historical order, a first deviation value between the order charging power of the order and the maximum charging power is calculated, and a second deviation value between the order charging power of the order and the average charging power is calculated.

[0157] In the step of determining the sample type based on the order deviation value, it is determined whether the absolute value of the first deviation value is less than the absolute value of the second deviation value. If so, the order is determined to be a positive sample; if so, the order is determined to be a negative sample.

[0158] In the step of determining the parameters of the sample data, since some charging characteristics are highly correlated with the charging power, such as whether the aforementioned charging date is a holiday, whether the last charge was fully charged, etc., some charging characteristics are selected to construct the sample data to predict the charging power through the inherent correlation of the data.

[0159] Specifically in the embodiment, whether the current day is a holiday, whether the previous day is a holiday, whether tomorrow is a holiday, the time since the last charge, the charging frequency, the last charged power, and whether the last charge reached a full charge state are selected as parameters to construct sample data.

[0160] In the step of constructing a sample dataset based on sample type and parameters, the sample dataset includes a training sample set, a validation sample set, and a test sample set. Specifically, a large number of positive and negative samples are extracted from all historical orders as the training sample set, and a portion of positive and negative samples are simultaneously extracted as the validation sample set and the test sample set.

[0161] Specifically in the embodiment, 10,000 positive samples and 10,000 negative samples are selected in the training sample set, and 5,000 positive samples and 5,000 negative samples are selected in the validation sample set and the test sample set.

[0162] In the step of generating a charging power prediction model based on a binary classification algorithm according to a sample data set, a preliminary charging power prediction model is generated according to the binary classification algorithm, and the charging power prediction model is trained using a training sample set, a validation set is used to calibrate the model parameters, and then a test set is used to test the accuracy.

[0163] Specifically in the embodiment, the binary classification algorithm is based on the algorithm function directly called by the xgboost tool, and the recall rate of the test set reaches 90% and the precision rate reaches 86%.

[0164] It should be noted that the above model generation steps are just a brief description. The actual model training and calibration steps are more complicated and will not be explained in detail here.

[0165] It should also be noted that the algorithm can be an algorithm tool directly called from an open source database, such as xgboost, or a self-developed classification algorithm, which is not limited here.

[0166] In some other embodiments, the method for generating the minimum power forecast model includes the following steps:

[0167] Construct a sample dataset based on charging characteristics;

[0168] According to the sample data set, multiple clusters are determined based on the clustering algorithm;

[0169] The minimum power consumption of the cluster is determined according to the characteristics of the cluster to generate a minimum power consumption prediction model.

[0170] Specifically, constructing a sample dataset based on charging characteristics includes the following steps:

[0171] The parameters of the sample data are determined according to the charging characteristics, the sample data is standardized, and a sample data set is constructed.

[0172] In the step of determining sample data parameters based on charging characteristics, it is necessary to determine charging characteristics associated with the minimum charge as parameters of the minimum charge prediction model. Specifically, the charging characteristics used in the minimum charge prediction model include the predicted charge, the average number of times the gun is unplugged per day, and the average dwell time. Sample data is constructed based on these charging characteristics.

[0173] In some embodiments, the sample data is standardized for use in subsequent clustering algorithms.

[0174] An appropriate amount of sample data is selected to construct the sample data set to avoid insufficient accuracy of the model due to too little sample data or complicated processing due to too much data. Therefore, in this embodiment, the sample data set preferably includes a total of 20,000 sample data.

[0175] In the step of determining the minimum power of the cluster according to the characteristics of the cluster to generate the minimum power prediction model, the clustering algorithm divides the sample data into multiple clusters according to the characteristics of the sample data set, and the number of clusters can be pre-set by the algorithm.

[0176] In some embodiments, the clustering algorithm is DBSCAN (Density-Based Spatial Clustering of Applications with Noise). The DBSCAN algorithm divides the sample data set into five categories and calculates the average value of the parameters of each category. The calculation results are shown in the following table:

[0177] Specifically in this embodiment, based on the characteristics and categories of the clusters, the cluster of category 0 is identified as having high charging power, high temporary unloading time, and low dwell time, and a high minimum power is assigned to the cluster.

[0178] The cluster of category 1 is identified as having low charging power, low temporary unloading power, and low stay time, and is given a medium minimum guaranteed power.

[0179] The cluster of category 2 is identified as having low charging power, low temporary unplugging, and high dwell time, and a high minimum power is assigned to this cluster.

[0180] The cluster of category 3 is identified as having high charging power, high temporary unloading power, and high dwell time, and is given a high minimum power guarantee.

[0181] The cluster of category 4 is identified as having high charging power, low temporary unplugging, and high dwell time, and a low minimum power is assigned to this cluster.

[0182] It should be noted that the specific values ​​of the high minimum power, medium minimum power, and low minimum power can be adjusted according to actual conditions. In this embodiment, the high minimum power is assigned to 5kwh, the medium minimum power is assigned to 3kwh, and the low minimum power is assigned to 1kwh.

[0183] It should also be noted that the algorithm used for clustering and the number of clusters set can be adaptively modified by those skilled in the art according to actual needs, and the above embodiments are for illustrative purposes only.

[0184] Please refer to FIG3 , which is a flowchart of generating a charging plan based on historical charging data and charging prediction data provided in one embodiment of the present application, specifically including the following steps:

[0185] S310, determining whether the charging strategy model is a safety assurance strategy model;

[0186] S320: Generate a charging plan based on the safety assurance strategy model;

[0187] S330, determining whether the charging strategy model is an energy-saving strategy model;

[0188] S340, generating a charging plan according to the energy-saving strategy model;

[0189] S350: Determine whether the charging strategy model is a cost-priority charging strategy model;

[0190] S360: Determine the charging plan as the first charging plan based on the cost priority strategy model.

[0191] S370: Determine the charging plan as the second charging plan based on the efficiency priority strategy model.

[0192] The following is a detailed introduction to the various steps in generating a charging plan based on historical charging data and charging forecast data.

[0193] S310: Determining whether the charging strategy model is a safety-guaranteed strategy model includes the following steps:

[0194] S311. Determine a historical charging curve based on historical charging data;

[0195] S312. Determine a first similarity between the historical charging curve and the standard charging curve;

[0196] S313: Determine whether the first similarity is greater than a preset first threshold;

[0197] S314: When the first similarity is greater than the first threshold, determine that the target charging strategy model is a safety assurance strategy model, and proceed to step S320.

[0198] S315: When the first similarity is less than the first threshold, proceed to step S330.

[0199] Each step is described in detail below with reference to the embodiments.

[0200] In the step of determining a historical charging curve according to historical charging data at S311 , the charging process of each order is queried according to the historical charging data of the device to be charged to determine the historical charging curve.

[0201] Specifically in the embodiment, the historical charging curve is a charging curve of SOC (State of charge)-Power, that is, a graph showing the relationship between the battery state and the charging power during the charging process.

[0202] Among them, SOC can also be replaced by units such as Ah (ampere-hour) and kWh (kilowatt-hour), and Power can be replaced by units such as V (voltage).

[0203] In step S312, determining the first similarity between the historical charging curve and the standard charging curve, different devices to be charged have different performance parameters and therefore different and unique standard charging curves. By comparing the historical charging curve obtained for each order with the standard charging curve, the proportion of the overlap between the two curves in the historical charging curve can be determined.

[0204] In some embodiments, the first similarity may be determined by averaging the overlap ratios of the historical charging curve and the standard charging curve within a set time window.

[0205] In S313, in the step of determining whether the first similarity is greater than a preset first threshold, the first threshold is set to 90%. When the first similarity between the historical charging curve and the standard charging curve is greater than 90%, it can be determined that the user is a safety-guaranteed user, and the charging strategy model can be further determined to be a safety-guaranteed strategy model, and the process proceeds to step S320.

[0206] S320: Generate a charging plan based on the safety assurance strategy model, including the following steps:

[0207] S321. Determine a charging curve estimation model based on the charging strategy model;

[0208] S322. Determine a predicted battery state value based on a charging curve estimation model;

[0209] S323: Generate a charging plan based on the predicted battery status value and the standard charging curve.

[0210] In the step of determining a charging curve estimation model according to the charging strategy model in S321, when the charging strategy model is determined to be a safety-guaranteed strategy model, a charging curve estimation model corresponding to the device to be charged is determined.

[0211] In step S322, determining the predicted battery state value according to the charging curve estimation model, the charging curve estimation model is combined with the obtained initial SOC value of the device to be charged to predict the state value of the device to be charged at any time, that is, the SOC value.

[0212] In S323, in the step of generating a charging plan based on the predicted battery status value and the standard charging curve, the power value is determined for each predicted moment based on the standard charging curve, so that the device to be charged is charged according to the standard charging curve, thereby reducing the loss of the device to be charged during the charging process.

[0213] It should be noted that the standard charging curve and the charging curve estimation model in the above embodiment are obtained by fitting the test data of each device to be charged.

[0214] S330: Determining whether the charging strategy model is an energy-saving strategy model includes the following steps:

[0215] S331. Determine charging time distribution based on historical charging data;

[0216] S332. Determine a second similarity between the charging time distribution and the charging peak time distribution;

[0217] S333: Determine whether the second similarity is greater than a preset second threshold;

[0218] S334: When the similarity is less than the threshold, determine that the target charging strategy model is an energy-saving strategy model.

[0219] S335: When the similarity is less than the threshold, proceed to step S350.

[0220] Specifically, in the step of determining the charging time distribution according to the historical charging data in S331, the charging time of each charging order is determined according to the historical charging data of the device to be charged, thereby determining the charging time distribution.

[0221] In some embodiments, the charging time of a user in a certain charging order is from 19:00 to 6:00 the next day, that is, the charging time is distributed as 19:00-0:00 and 0:00-6:00.

[0222] In S332, the second similarity step of determining the charging time distribution and the charging peak time distribution, the charging time distribution is compared with the charging peak time distribution of the corresponding charging station to obtain the time overlap ratio between the charging time distribution of the order and the charging peak time distribution. The time overlap ratio within the set time window is averaged to determine the similarity.

[0223] It should be noted that the peak charging time distribution of the charging station is determined by performing a statistical analysis of the charging power distribution of the charging station at each moment.

[0224] Please refer to Figure 4, which is a time distribution statistical diagram of the charging power of the charging station provided in an embodiment of the present application. As can be seen from the figure, the peak charging times are distributed in three time periods: 7:00 to 9:00, 12:00 to 14:00, and 17:00 to 19:00.

[0225] In S334, when the second similarity is less than the second threshold, the charging strategy model is determined to be an energy-saving strategy model. Specifically, if the similarity between the user's charging time distribution and the charging peak time distribution is less than 10%, it can be determined that the user is deliberately avoiding charging peak time, and the user is determined to be an energy-saving user. Further, the charging strategy model can be determined to be an energy-saving strategy model, and the process proceeds to step S340.

[0226] It should be noted that in some other embodiments, the order of steps S310 and S320 is reversed from that of steps S330 and S340. That is, the charging strategy model is first determined to be an energy-saving strategy model, and then the charging strategy model is determined to be a safety-enhancing strategy model. In other words, the order of steps S310 and S320 and the order of steps S330 and S340 can be swapped and are not limited to this embodiment.

[0227] S340: Generate a charging plan based on the energy-saving strategy model, including the following steps:

[0228] S341, determining whether the charging period is a peak charging period;

[0229] S342: If yes, determine that the charging power during the charging period is the first power;

[0230] S343: If not, determine that the charging power during the charging period is the second power;

[0231] S344: Generate a charging plan based on the charging period and charging power.

[0232] In the steps of S342, "If yes, then determine the charging power of the charging period to be the first power" and S343, "If no, then determine the charging power of the charging period to be the second power", when the charging period is determined to be a peak charging period, charging at the relatively low first power is preferentially adopted. When the charging period is determined to be a non-peak charging period, charging at the relatively high second power is preferentially adopted.

[0233] In some embodiments, when the charging period is a peak charging period, the first power is set to 0, that is, the device to be charged is not charged during the peak charging period.

[0234] In S344, in the step of generating a charging plan based on the charging period and the charging power, the charging plan includes a continuous charging time, wherein part of the charging time is a peak charging period and is charged at a first power, and part of the charging time is a non-peak charging period and is charged at a second power.

[0235] S350: Determining whether the charging strategy model is a cost-priority charging strategy model includes the following steps:

[0236] S351. Calculate a first charging cost based on a preset cost priority model according to historical charging data and charging forecast data;

[0237] S352: Calculate a second charging cost based on a preset efficiency priority model according to historical charging data and charging forecast data;

[0238] S353: Determine whether the difference between the first charging cost and the second charging cost is greater than a preset threshold.

[0239] S354: If it is greater than, determine that the charging strategy model is a cost-priority model, and proceed to step S360.

[0240] S355: If it is less than, determine that the charging strategy model is an efficiency priority model, and go to step S370.

[0241] S351, calculating the first charging cost based on the preset cost priority model according to the historical charging data and the charging prediction data, specifically includes the following steps:

[0242] According to the historical charging data and the charging forecast data, a first charging plan is generated based on a preset cost priority model; the charging cost of the first charging plan is calculated to determine the first charging cost.

[0243] The steps of generating the first charging plan based on the preset cost priority model according to the historical charging data and the charging prediction data will be described in detail later and will not be shown here.

[0244] In the step of calculating the charging cost of the first charging plan and determining the first charging cost, after the first charging plan is generated, the total charging cost is calculated according to the first charging plan in combination with the real-time charging electricity price of the charging station, and is determined as the first charging cost.

[0245] In S352, the step of calculating the second charging cost based on the preset efficiency priority model according to the historical charging data and the charging prediction data includes: generating a second charging plan based on the preset efficiency priority model according to the historical charging data and the charging prediction data; and calculating the charging cost of the second charging plan to determine the second charging cost.

[0246] The step of generating a second charging plan based on a preset efficiency-first model based on the historical charging data and the charging forecast data includes: determining constraints of the efficiency-first model based on the historical charging data and the charging forecast data; and generating the second charging plan based on the constraints and a power load balancing algorithm.

[0247] Specifically, the objective function of the efficiency-first model is determined to minimize the total charging time, so the solution is relatively simple. It only needs to consider load balancing the power transmitted by charging stations, groups, charging piles and charging guns, and then set the power transmitted in fast charging mode after distribution.

[0248] Specifically in the embodiment, the power load balancing distribution algorithm includes any one of an average distribution algorithm, a weighted average algorithm, and a first-come, first-served algorithm.

[0249] In the step of calculating the charging cost of the second charging plan and determining the second charging cost, after the second charging plan is generated, the charging cost is calculated according to the second charging plan in combination with the real-time charging electricity price of the charging station and is determined as the second charging cost.

[0250] In step S354, "If greater than, then determining the charging strategy model as a cost-priority model," since the first charging cost is calculated using the cost-priority model, the first charging cost is less than the second charging cost. When the difference between the first and second charging costs is greater than a certain threshold, it can be determined that the first charging cost calculated based on the cost-priority model is significantly less than the second charging cost calculated based on the efficiency-priority model. Therefore, it is determined that the user of the device to be charged is a cost-sensitive customer, and the charging strategy model used is the cost-priority model, and the process proceeds to step S360.

[0251] In S355, if it is less than, then determine that the charging strategy model is an efficiency priority model step, when the difference between the first charging cost and the second charging cost is less than a certain threshold, it can be judged that the first charging cost calculated based on the cost priority model is not significantly less than the second charging cost calculated based on the efficiency priority model. Therefore, it is determined that the user of the device to be charged is not sensitive to the charging cost, and is classified as an efficiency-sensitive customer. The charging strategy model adopted is the efficiency priority model, and the process proceeds to step S370.

[0252] In some embodiments, the threshold is set to 5%, that is, when the cost reduction rate of the first charging cost compared to the second charging cost is greater than 5%, the cost priority model is used, and when it is less than 5%, the efficiency priority model is used.

[0253] In S360, in the step of determining the charging plan as the first charging plan based on the cost priority strategy model, since the first charging plan and the second charging plan are generated first and then the charging costs of the two are compared, the first charging plan has been generated in advance. Therefore, in this step, only the first charging plan needs to be determined.

[0254] It can be understood that the same principle applies to the step of determining the charging plan as the second charging plan in S370 according to the efficiency priority strategy model, and therefore will not be described in detail.

[0255] The method for generating the first charging plan is described in detail below.

[0256] Please refer to FIG5 , which is a flowchart of a method for generating a first charging plan based on a preset cost priority model according to an embodiment of the present application, specifically including:

[0257] S510: Determine the constraints of the cost priority model based on historical charging data and charging prediction data.

[0258] S520: Determine the objective function of the cost priority model.

[0259] S530: Generate a first charging plan based on an operations research optimization solver according to the objective function and the constraints.

[0260] In S510, the step of determining the constraints of the cost-priority model based on historical charging data and charging forecast data, the parameters used in the cost-priority model include charging cost electricity price, starting charging time, charging time, predicted charging power, predicted minimum power, predicted maximum received power, charging station power limit, group power limit, and charging pile power limit. Each interval of time, such as 5 minutes, from the starting charging time to the charging time is split into a time segment, and the corresponding power transmitted in each time segment is set as an unknown variable.

[0261] Specifically in the embodiment, based on the above parameters, determining the constraint conditions includes:

[0262] Constraint 1: The total power of the charging station in each time segment does not exceed the upper limit of the charging station power.

[0263] Constraint 2: The total power of the group in each time segment does not exceed the group upper limit.

[0264] Constraint 3: The total power of the charging guns of the charging pile in each time segment does not exceed the pile upper limit.

[0265] Constraint 4: The transmitted power in each time segment does not exceed the predicted power upper limit.

[0266] Constraint 5: Charge in fast charging mode until the charged capacity reaches the predicted minimum capacity.

[0267] Constraint 6: Before the charging time reaches the draw time, the charged capacity is greater than or equal to the predicted charging capacity.

[0268] It should be noted that, in some embodiments, those skilled in the art may adaptively adjust and modify the number or content of the constraints according to actual needs.

[0269] It should also be noted that, in some other embodiments, the constraint conditions are not only applied to the cost-first model, but can also be applied to the security-guarantee policy model, the energy-saving policy model, and the efficiency-first policy model.

[0270] In the step of determining the objective function of the cost priority model at S520, according to the policy model corresponding to the device to be charged being a cost priority policy model, the objective function of the cost priority model is determined to minimize the total charging cost.

[0271] In S530, based on the objective function and constraints, a first charging plan is generated based on an operations research optimization solver. Based on the above objective function and constraints, the open source or-tools is called to build a model and calculate the optimal solution. The optimal solution of the model is the first charging plan.

[0272] Specifically in the embodiment, the first charging plan is output in the following format:

[0273] [{"beginTime":"12:00","chargingUp":16.0},

[0274] {"beginTime":"14:00","chargingUp":0.0},

[0275] {"beginTime":"15:00","chargingUp":16.0}]

[0276] BeginTime refers to the starting time of power delivery, and chargingUp refers to the power delivery value corresponding to this starting time. The actual charging power is controlled by scheduling the power value and delivery time to achieve safe and stable charging without exceeding the power load, meet the customer's charging power needs, and achieve the best cost effect.

[0277] More specifically, in this embodiment, the period from 12:00 to 15:00 is a high-cost electricity price period. However, because the initial charging capacity does not reach the predicted minimum capacity, fast charging mode is continued at a high current of 16kW from 12:00 to 14:00. At 14:00, the charged capacity reaches the predicted minimum capacity. At this time, to meet the objective function of minimizing the total charging cost, charging is stopped to reduce the charging cost. After 15:00, the low-cost electricity price period begins, and charging is resumed at a distributed power of 16kW until the charged capacity reaches the predicted charging capacity.

[0278] The technical solution of this application can achieve the following technical effects: by mining and analyzing historical charging data, charging characteristics that characterize user charging behaviors are obtained, and based on these charging characteristics, the user's charging needs are predicted. This process does not involve user privacy information and can solve the problem of difficulty in obtaining user charging needs. Furthermore, by predicting user preferences, users are divided into groups and diversified, targeted charging plans are generated for different user groups, thus solving the problem that existing charging strategies cannot meet the charging needs of different users.

[0279] Predict user preference types based on charging characteristics and explore user charging needs, and generate customized charging plans to meet user charging needs.

[0280] An embodiment of the present application further provides a computer-readable storage medium, which stores a computer program. The computer program includes program instructions. When the program instructions are executed by a computer, the computer executes the method of the aforementioned embodiment.

[0281] An embodiment of the present application also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the method of the aforementioned embodiment is implemented when the processor executes the computer program.

[0282] In one embodiment, a computer device is provided, which may be a server, and its internal structure diagram may be as shown in Figure 6. The computer device includes a processor, a memory, a network interface, and a database connected via a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store data to be saved in the power adjustment method based on the charging pile. The network interface of the computer device is used to communicate with an external terminal via a network connection. When the computer program is executed by the processor, it implements the method as described in the aforementioned embodiment.

[0283] Those skilled in the art will appreciate that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0284] Those skilled in the art will clearly understand that for the sake of convenience and brevity of description, only the division of the above-mentioned functional units and modules is used as an example. In actual applications, the above-mentioned functions can be distributed and completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above.

[0285] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. These modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present invention, and should all be included in the scope of protection of the present invention.

Claims

1. A method for generating a charging plan, characterized in that: include: Get historical charging data; determining charging characteristics according to the historical charging data; Determining charging prediction data according to the charging characteristics and a preset charging prediction model; A charging plan is generated according to the historical charging data and the charging prediction data.

2. The method according to claim 1, characterized in that The generating a charging plan according to the historical charging data and the charging prediction data comprises: Determining a target charging strategy model according to the historical charging data and the charging prediction data; The charging plan is generated according to the target charging strategy model.

3. The method according to claim 2, characterized in that The determining of the target charging strategy model according to the historical charging data and the charging prediction data comprises: Calculating a first charging cost based on a preset cost priority model according to the historical charging data and the charging prediction data; Calculating a second charging cost based on a preset efficiency priority model according to the historical charging data and the charging prediction data; Determining whether a difference between the first charging cost and the second charging cost is greater than a preset threshold; If it is greater than, determining that the target charging strategy model is the cost priority model; If it is less than, the target charging strategy model is determined to be the efficiency priority model.

4. The method according to claim 3, characterized in that The calculating the first charging cost based on the preset cost priority model according to the historical charging data and the charging prediction data includes: Generate a first charging plan based on a preset cost priority model according to the historical charging data and the charging prediction data; The charging cost of the first charging plan is calculated to determine the first charging cost.

5. The method according to claim 4, characterized in that The generating a first charging plan based on a preset cost priority model according to the historical charging data and the charging prediction data includes: Determining the constraint conditions of the cost priority model according to the historical charging data and the charging prediction data; Determining an objective function of the cost priority model; According to the objective function and the constraint conditions, the first charging plan is generated based on an operations research optimization solver.

6. The method according to claim 5, characterized in that Determining the constraint conditions of the cost priority model according to the historical charging data and the charging prediction data includes: Determining, according to the charging prediction data, that the constraint condition includes at least one of the following: The transmitted power is less than or equal to the predicted maximum receivable power; When the charged capacity is lower than the predicted minimum capacity, fast charging is used; and Within the predicted charging time, the charged power is greater than or equal to the predicted charging power; The charging prediction data includes the predicted maximum receivable power, the predicted minimum guaranteed power, the predicted charging time and the predicted charging power.

7. The method according to claim 3, characterized in that The calculating the second charging cost based on the preset efficiency priority model according to the historical charging data and the charging prediction data includes: Generate a second charging plan based on a preset efficiency priority model according to the historical charging data and the charging prediction data; The charging cost of the second charging plan is calculated to determine the second charging cost.

8. The method according to claim 7, characterized in that The generating a second charging plan based on a preset efficiency priority model according to the historical charging data and the charging prediction data includes: Determining the constraint conditions of the efficiency priority model according to the historical charging data and the charging prediction data; According to the constraint conditions, the second charging plan is generated based on a power load balancing allocation algorithm.

9. The method according to claim 2, characterized in that The determining of the target charging strategy model according to the historical charging data and the charging prediction data comprises: Determine a historical charging curve according to the historical charging data; Determining a first similarity between the historical charging curve and a standard charging curve; Determining whether the first similarity is greater than a preset first threshold; When the first similarity is greater than the first threshold, it is determined that the target charging strategy model is a safety assurance strategy model.

10. The method according to claim 9, characterized in that The generating the charging plan according to the target charging strategy model includes: Determining a charging curve estimation model according to the charging strategy model; Determining a predicted battery state value according to the charging curve estimation model; The charging plan is generated according to the predicted battery state value and the standard charging curve.

11. The method according to claim 2, characterized in that The determining of the target charging strategy model according to the historical charging data and the charging prediction data comprises: determining a charging time distribution according to the historical charging data; determining a second similarity between the charging time distribution and the charging peak time distribution; Determining whether the second similarity is greater than a preset second threshold; When the second similarity is greater than the second threshold, it is determined that the target charging strategy model is an energy-saving strategy model.

12. The method according to claim 11, characterized in that Generating the charging plan according to the target charging strategy model includes: Determining whether the charging period is the peak charging period; If yes, determining that the charging power during the charging period is the first power; If not, determining that the charging power of the charging period is a second power, wherein the first power is less than the second power; The charging plan is generated according to the charging period and the charging power.

13. The method according to any one of claims 1 to 12, characterized in that: Before determining the charging prediction data according to the charging characteristics and the preset charging prediction model, the method further includes: generating the charging prediction model; The charging prediction model includes a minimum power prediction model and / or a charging power prediction model.

14. The method according to claim 13, characterized in that Generating a charging prediction model comprises: Determine the maximum charging power and the average charging power of the charging characteristic as output label values; Construct a sample data set based on charging characteristics; According to the sample data set, the charging power prediction model is generated based on a binary classification algorithm.

15. The method according to claim 13, characterized in that Generating a charging prediction model comprises: Construct a sample data set based on charging characteristics; According to the sample data set, determining a plurality of clusters based on a clustering algorithm; The guaranteed minimum electricity quantity of the cluster is determined according to the characteristics of the cluster to generate the guaranteed minimum electricity quantity prediction model.

16. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the method according to any one of claims 1 to 15 is implemented.

17. A storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 15 is implemented.

Images