Electric vehicle charging load prediction method and device, computer equipment, readable storage medium and program product

Abstract

The invention relates to an electric vehicle charging load prediction method and device, computer equipment, a computer readable storage medium and a computer program product. The method comprises the following steps: acquiring charging data of a to-be-processed electric vehicle in a current time period; performing feature extraction on each piece of charging data to obtain charging features; classifying the to-be-processed electric vehicles based on the charging characteristics, and determining target charging characteristics based on the types of the to-be-processed electric vehicles; and performing prediction based on each target charging feature to obtain the charging load of each type of electric vehicles in the next time period. By adopting the method, the prediction cost can be reduced.

Description

Technical Field

[0001] The present application relates to the field of electrical technology, and in particular to a method, device, computer equipment, computer-readable storage medium and computer program product for predicting charging load of an electric vehicle. Background Art

[0002] With the global focus on improving energy efficiency, electric vehicles (EVs) have become an important development direction in the transportation sector. As a mode of interaction between electric vehicles and the power grid, V2G (Vehicle-to-Grid) technology allows electric vehicles to provide energy to the grid when grid demand is peak, or absorb energy from the grid for charging when electricity prices are low. This two-way energy flow provides additional flexibility and stability for the grid.

[0003] Traditional load estimation methods rely on individual measurements of each EV’s charging behavior, which is not only costly but also difficult to scale to large user groups. Summary of the invention

[0004] Based on this, it is necessary to provide an electric vehicle charging load prediction method, device, computer equipment, computer-readable storage medium and computer program product that can reduce the prediction cost in response to the above technical problems.

[0005] In a first aspect, the present application proposes a method for predicting charging load of an electric vehicle, the method comprising:

[0006] Obtain charging data of the electric vehicle to be processed in the current time period;

[0007] Extracting features from each of the charging data to obtain charging features;

[0008] Classifying each of the electric vehicles to be processed based on each of the charging characteristics, and determining a target charging characteristic based on each category of the electric vehicles to be processed;

[0009] Predictions are made based on the target charging characteristics to obtain the charging load of each category of electric vehicles in the next time period.

[0010] In one embodiment, the classifying each of the electric vehicles to be processed based on each of the charging characteristics, and determining a target charging characteristic for each category of the electric vehicles to be processed, includes:

[0011] Randomly selecting a number of charging characteristics of the electric vehicles to be processed as cluster centers;

[0012] Calculating the distance between the charging characteristics of the other electric vehicles to be processed and the cluster center, and clustering the other electric vehicles to be processed according to the distance to obtain an initial category set;

[0013] The cluster center is updated based on the charging characteristics of each of the electric vehicles to be processed in the initial category set, and the clustering step is re-executed until the iteration condition is met to obtain the target category set;

[0014] The charging features of the electric vehicles to be processed corresponding to each of the cluster centers in each of the target category sets are used as target charging features of each of the target category sets.

[0015] In one embodiment, the charging characteristics include at least one or more of energy receiving characteristics, energy return characteristics, electricity price sensitivity characteristics, charging frequency characteristics, and charging demand characteristics;

[0016] The energy receiving characteristic is used to characterize the total energy received by the electric vehicle to be processed from the power grid;

[0017] The energy return feature is used to characterize the total energy returned to the power grid by the electric vehicle to be processed;

[0018] The electricity price sensitivity feature is used to characterize the sensitivity of the electric vehicle to be processed to the electricity price;

[0019] The charging frequency feature is used to characterize the charging frequency of the electric vehicle to be processed within a preset time period;

[0020] The charging demand feature is used to characterize the demand for electric energy of the electric vehicle to be processed.

[0021] In one embodiment, the calculating the distance from the charging characteristics of the other electric vehicles to be processed to the cluster center includes:

[0022] Respectively calculate the distance components of the energy receiving feature, the energy return feature, the electricity price sensitive feature, the charging frequency feature and the charging demand feature in the charging features of the other electric vehicles to be processed to the energy receiving feature, the energy return feature, the electricity price sensitive feature, the charging frequency feature and the charging demand feature of the electric vehicle to be processed corresponding to the corresponding cluster center;

[0023] The distance components are integrated to obtain the distances from the charging characteristics of other electric vehicles to be processed to the cluster center.

[0024] In one embodiment, the charging data includes charging time, total energy received, total energy returned, charging time, required power and current power;

[0025] The extracting features of each charging data to obtain charging features includes:

[0026] Counting the total amount of energy received by each of the electric vehicles to be processed in each preset time interval to obtain the energy receiving characteristics; the total amount of energy received by the electric vehicle to be processed and the power grid;

[0027] The total energy returned by each of the processed electric vehicles is used as the energy return feature; the total energy returned is the total energy transmitted by the electric vehicle to be processed to the power grid;

[0028] Calculating the ratio of the charging time of each processed electric vehicle to the preset charging time to obtain the electricity price sensitive feature;

[0029] Obtaining the charging frequency characteristic according to the time interval of the charging moment of each processing electric vehicle;

[0030] The charging demand characteristic is obtained according to the difference between the required power and the current power of each processed electric vehicle.

[0031] In one embodiment, the prediction based on each of the target charging characteristics to obtain the charging load of each category of electric vehicles in the next time period is obtained by a pre-trained charging load prediction model; the training process of the charging load prediction model includes:

[0032] Acquire a charging feature sample; the feature carries a charging load tag;

[0033] Inputting the charging characteristic samples into an initial model for training to obtain a predicted charging load;

[0034] A target loss function is obtained based on the predicted charging load and the charging load label, and the parameters of the initial model are adjusted according to the target loss function until the training is completed to obtain the charging load pre-model.

[0035] In a second aspect, the present application proposes a device for predicting charging load of an electric vehicle, the device comprising:

[0036] An acquisition module is used to acquire charging data of the electric vehicle to be processed in the current time period;

[0037] A feature extraction module, used to extract features from the charging data to obtain charging features;

[0038] A classification module is used to classify each of the electric vehicles to be processed based on the charging characteristics, and determine the target charging characteristics for each category of the electric vehicles to be processed; a prediction module is used to predict based on the target charging characteristics to obtain the charging load of each category of electric vehicles in the next time period.

[0039] In a third aspect, the present application proposes a computer device, comprising a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of the above method when executing the computer program.

[0040] In a fourth aspect, the present application proposes a computer-readable storage medium having a computer program stored thereon, wherein the computer program implements the steps of the above method when executed by a processor.

[0041] In a fifth aspect, the present application proposes a computer program product, comprising a computer program, which implements the steps of the above method when executed by a processor.

[0042] The above-mentioned electric vehicle charging load prediction method, device, computer equipment, computer-readable storage medium and computer program product obtain the charging data of the electric vehicle to be processed in the current time period, perform feature extraction based on the charging data to obtain charging features, and then perform classification based on the charging features of each electric vehicle to be processed to obtain target charging features that can characterize each different category of electric vehicles, and finally perform prediction based on the target charging features to obtain the charging load of each category of electric vehicles in the next time period. In this way, the target charging features of each different category of electric vehicles are obtained by classification, and then only the target charging features are predicted, so that it is not necessary to predict each electric vehicle, which greatly reduces the calculation cost. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] In order to more clearly illustrate the technical solutions in the embodiments of the present application or related technologies, the drawings required for use in the embodiments of the present application or related technical descriptions will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0044] Figure 1 A schematic diagram of a flow chart of a method for predicting charging load of an electric vehicle in one embodiment;

[0045] Figure 2 It is a structural block diagram of an electric vehicle charging load prediction device in one embodiment;

[0046] Figure 3 FIG. 4 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0047] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0048] In one embodiment, Figure 1 As shown, a method for predicting charging load of electric vehicles is provided. This embodiment takes the method applied to a terminal as an example. It can be understood that the method can also be applied to a server, and can also be applied to a system including a terminal and a server, and is implemented through the interaction between the terminal and the server. In this embodiment, the method includes the following steps:

[0049] Step S102, obtaining charging data of the electric vehicle to be processed in the current time period.

[0050] Optionally, the charging data of all the electric vehicles to be processed in the current time period may be obtained through the charging infrastructure of the electric vehicles, such as charging piles or smart grids.

[0051] Optionally, the charging data typically includes the charging start time, charging end time, initial and final values ​​of SOC (battery state), charging power, total charging amount, charging price, and possible V2G (vehicle to grid) energy feedback data for each electric vehicle.

[0052] Optionally, the charging data may be recorded in the form of a time series, reflecting the charging changes of each electric vehicle.

[0053] Optionally, after acquiring the charging data, the integrity and accuracy of the data needs to be verified to avoid distortion of subsequent analysis results due to data distortion or noise.

[0054] For example, it is possible to check whether there are missing values ​​or abnormal values ​​in the data set to verify the integrity of the charging data. For example, it is possible to check whether there are missing values ​​in key fields (such as charging start time, end time, SOC, power, etc.) in the data set.

[0055] For example, the data may be checked for inconsistencies or contradictions. For example, the SOC should increase during charging, and if some records show a decrease in the SOC, this is inconsistent data. For example, preset logic rules are set (such as the SOC value should increase gradually, and the charging power should be within a reasonable range) and checked for any records that violate these rules.

[0056] Step S104, extracting features from each charging data to obtain charging features.

[0057] In this embodiment, key features that can describe charging behavior are extracted from the collected charging data. The goal of feature extraction is to convert complex raw data into feature vectors that can be processed.

[0058] Optionally, the charging characteristics may include charging start time, charging end time, SOC change, charging power and total charging energy, electricity price sensitivity, charging frequency, and V2G feedback energy.

[0059] Among them, the charging start time and end time reflect the user's charging habits and usage patterns; the SOC change records the change in battery power during the charging process; the charging power and total charging amount reflect the energy consumption of each charging; the electricity price sensitivity is the user's sensitivity to price calculated based on the user's charging behavior under different electricity prices; the charging frequency records the frequency of the user's charging, reflecting the user's car usage habits and travel patterns; the V2G feedback energy describes the amount and frequency of energy fed back by the vehicle to the power grid.

[0060] Optionally, feature extraction can also be performed by establishing a charging process model, which defines and describes the behavior of the electric vehicle during the charging process, and then extracts key features from the charging data.

[0061] Step S106, classifying each electric vehicle to be processed based on each charging characteristic, and determining a target charging characteristic based on each category of the electric vehicle to be processed.

[0062] Optionally, a clustering or classification algorithm may be used to classify electric vehicles based on the extracted charging features. The purpose of the classification is to group electric vehicles with similar charging behaviors into the same category in order to simplify data processing during the prediction process.

[0063] Optionally, electric vehicles can be divided into several groups according to charging characteristics. The electric vehicles in each group show similar behaviors in terms of charging time, charging amount, price sensitivity, etc.

[0064] Alternatively, supervised learning algorithms such as decision trees and random forests can be used to classify electric vehicles.

[0065] In this way, one or more target charging features can be determined for each category through classification. These features represent the charging behavior patterns of electric vehicles in this category, thereby reducing the complexity of data processing and improving the efficiency and accuracy of the prediction model.

[0066] Step S108, predicting based on each target charging characteristic to obtain the charging load of each category of electric vehicles in the next time period.

[0067] In this embodiment, the charging load of each type of electric vehicles in the following time period is predicted based on the target charging characteristics determined in step S106.

[0068] Optionally, the charging load of various types of electric vehicles in the next time period can be predicted by a pre-trained prediction model.

[0069] In this way, through prediction, it is possible to understand in advance the charging demand of electric vehicles in the next time period, thereby helping the power grid to reasonably allocate resources, balance the load, and avoid grid overload or resource waste.

[0070] This kind of prediction is particularly important for electric vehicles with V2G function, because such electric vehicles can not only obtain electric energy from the power grid, but also feed back electric energy to the power grid.

[0071] In the above-mentioned electric vehicle charging load prediction method, the charging data of the electric vehicles to be processed in the current time period is obtained, feature extraction is performed based on the charging data to obtain charging features, and then the charging features of each electric vehicle to be processed are classified separately to obtain target charging features that can represent different categories of electric vehicles. Finally, prediction is performed based on the target charging features to obtain the charging loads of various categories of electric vehicles in the next time period. In this way, by classifying to obtain the target charging features of different categories of electric vehicles and then only predicting based on the target charging features, it is not necessary to predict each electric vehicle, greatly reducing the calculation cost.

[0072] In one embodiment, classifying each electric vehicle to be processed based on each charging feature and determining the target charging features for each category of the electric vehicles to be processed includes: randomly selecting the charging features of several electric vehicles to be processed as cluster centers; calculating the distances from the charging features of other electric vehicles to be processed to the cluster centers, and clustering the other electric vehicles to be processed according to the distances to obtain an initial category set; updating the cluster centers based on the charging features of the electric vehicles to be processed in the initial category set, and re-executing the clustering step until the iteration condition is reached to obtain a target category set; using the charging features of the electric vehicles to be processed corresponding to each cluster center in each target category set as the target charging features of each target category set.

[0073] Optionally, several charging features are selected from the charging features of the electric vehicles to be processed as cluster centers first, and then the distances from the charging features of other electric vehicles to be processed to the cluster centers are calculated, and this distance will be used as the classification criterion for other electric vehicles to be processed.

[0074] Exemplarily, the distance from the charging feature of the electric vehicle to be processed 1 to cluster center A is x, and the distance to cluster center B is y, where x < y, then the electric vehicle to be processed 1 is classified into category A.

[0075] Exemplarily, the distance from the charging feature of each electric vehicle to be processed to cluster center A can be calculated in combination with formula (1).

[0076] Formula (1)

[0077] Optionally, after all other electric vehicles to be processed are classified, several initial category sets are obtained, wherein the number of the several initial category sets is consistent with the number of the several cluster centers.

[0078] Optionally, after obtaining several initial category sets, the cluster centers may change, so the cluster centers are recalculated based on each initial category set.

[0079] Optionally, the mean of the charging characteristics of each electric vehicle to be processed in each initial category set is calculated as the new cluster center, and then the above classification process is repeated until the iteration condition is met, and the initial category set that meets the iteration condition is used as the target category set.

[0080] Optionally, the charging features of the electric vehicles to be processed corresponding to the cluster centers of each target category set are used as the target charging features of the target category set.

[0081] In this embodiment, each electric vehicle to be processed is classified, with the goal of aggregating electric vehicles with similar charging behaviors.

[0082] In the above embodiment, the charging characteristics of electric vehicles can be effectively classified, and a representative charging characteristic is selected for each category.

[0083] In one embodiment, the charging characteristics include at least one or more of an energy receiving characteristic, an energy returning characteristic, an electricity price sensitive characteristic, a charging frequency characteristic, and a charging demand characteristic.

[0084] Among them, the energy reception feature is used to characterize the total energy received by the electric vehicle to be processed from the power grid; the energy return feature is used to characterize the total energy returned to the power grid by the electric vehicle to be processed; the electricity price sensitivity feature is used to characterize the sensitivity of the electric vehicle to be processed to the electricity price; the charging frequency feature is used to characterize the charging frequency of the electric vehicle to be processed within a preset time period; and the charging demand feature is used to characterize the demand for electricity by the electric vehicle to be processed.

[0085] Furthermore, in one embodiment, the distances from the charging characteristics of other electric vehicles to be processed to the cluster center are calculated, including: respectively calculating the distance components of the energy receiving characteristics, energy return characteristics, electricity price sensitivity characteristics, charging frequency characteristics and charging demand characteristics of the charging characteristics of other electric vehicles to be processed to the corresponding cluster center; and combining the distance components to obtain the distances from the charging characteristics of other electric vehicles to be processed to the cluster center.

[0086] Among them, the distance component is the component from each feature in the charging feature to the corresponding feature of the cluster center.

[0087] In this embodiment, the distance components of the energy reception characteristics of other electric vehicles to be processed to the energy reception characteristics of the electric vehicle to be processed corresponding to the cluster center, the distance components of the energy return characteristics of other electric vehicles to be processed to the energy return characteristics of the electric vehicle to be processed corresponding to the cluster center, the distance components of the electricity price sensitive characteristics of other electric vehicles to be processed to the electricity price sensitive characteristics of the electric vehicle to be processed corresponding to the cluster center, the distance components of the charging frequency characteristics of other electric vehicles to be processed to the charging frequency characteristics of the electric vehicle to be processed corresponding to the cluster center, and the distance components of the charging demand characteristics of other electric vehicles to be processed to the charging demand characteristics of the electric vehicle to be processed corresponding to the cluster center are calculated respectively.

[0088] Then, the distance components are integrated to obtain the distances from the charging characteristics of other electric vehicles to be processed to the cluster center.

[0089] Optionally, the sum of the squares of the distance components may be used to obtain the distances from the charging features of other electric vehicles to be processed to the cluster center.

[0090] Optionally, if some features contribute more to the overall distance calculation than other features, weights may be assigned to each feature component to obtain the distances from the charging features of other electric vehicles to be processed to the cluster center.

[0091] Optionally, the electric vehicles to be processed are assigned to the cluster corresponding to the nearest cluster center according to the calculated distance. This will determine the classification of the electric vehicles and thus perform corresponding load forecasting.

[0092] In the above embodiment, it is helpful to allocate electric vehicles to the most suitable cluster, thereby improving the accuracy of prediction and analysis.

[0093] Furthermore, the above charging data includes charging time, total energy received, total energy returned, charging time, required power and current power; the above-mentioned feature extraction of each charging data obtains charging characteristics, including: counting the total amount received by each electric vehicle to be processed in each preset time interval to obtain energy reception characteristics; the total amount received is the total energy received by the electric vehicle to be processed and sent to the power grid; taking the total energy returned by each processed electric vehicle as the energy return characteristic; the total energy returned is the total energy transmitted by the electric vehicle to be processed to the power grid; calculating the ratio of the charging time of each processed electric vehicle to the preset charging time to obtain electricity price sensitivity characteristics; according to the time interval of the charging time of each processed electric vehicle, obtaining the charging frequency characteristics; according to the difference between the required power and the current power of each processed electric vehicle, obtaining the charging demand characteristics.

[0094] The energy receiving characteristic represents the total energy received by the electric vehicle from the power grid in each preset time interval. The preset time interval is preset and can be set per hour or per day.

[0095] Optionally, for each electric vehicle to be processed, the total energy received from the power grid is counted within each preset time interval, and the statistical value is used as the energy reception feature.

[0096] Among them, the energy return characteristic represents the total energy returned by electric vehicles to the power grid.

[0097] Optionally, the total energy returned to the power grid by each electric vehicle to be processed within a preset time period is counted, that is, the electric energy is returned to the power grid through V2G technology, and the total returned energy is the energy return feature.

[0098] It should be noted that V2G technology provides some opportunities for power system operators, such as using V2G capabilities for peak load regulation. Charging piles inject power into the grid when the grid needs it. Since the capacity that electric vehicles can discharge depends on the user's charging behavior, it is very necessary to use the power sent back to the grid as a feature. In addition, in this embodiment, it is set that electric vehicles can only transmit power to the grid at peak electricity prices.

[0099] Among them, the electricity price sensitivity feature is used to measure the sensitivity of electric vehicle charging behavior to electricity prices.

[0100] Optionally, the actual charging time of each electric vehicle may be calculated, and then the preset charging time of the electric vehicle may be calculated, and the actual charging time may be divided by the preset charging time to obtain the electricity price sensitive feature, wherein the preset charging time is calculated based on the battery capacity and charging power of the electric vehicle.

[0101] For example, depending on the travel behavior of electric vehicle owners, the user's sensitivity to electricity prices is also different. If the stay time is equal to the minimum required charging time, then the user's sensitivity to electricity prices is very low. However, if the stay time is much longer than the charging time, then charging can be scheduled according to the electricity price. The formula for the user's sensitivity to electricity prices is shown in formula (2):

[0102] Formula (2)

[0103] In formula (2), Indicates the sensitivity to electricity price during the charging process s, which is greater than or equal to 1. Equal to 1 means that the user's charging event is 0 sensitive to electricity prices. The larger the value of , the higher the sensitivity to electricity prices. Indicates the length of stay, Indicates the minimum charging time.

[0104] in, The calculation formula is shown in formula (3).

[0105] Formula (3)

[0106] Formula (3) represents the residence time in charging event s. s Indicates the charging start time, PO S Indicates the charging end time.

[0107] in, The calculation formula of is shown in formula (4).

[0108] Formula (4)

[0109] Formula (4) represents the shortest charging time required to charge to the desired SOC. Indicates the SOC at the end of the charging time, Indicates the SOC at the beginning of the charging time. is the efficiency factor, which is used to adjust the relationship between different charging parameters. It is the product of maximum power and efficiency, indicating the maximum energy that an electric vehicle can obtain from the power grid.

[0110] Optionally, the charging frequency feature is used to characterize the charging frequency of the electric vehicle in different time intervals, reflecting the regularity of the charging behavior.

[0111] Optionally, the number of times the electric vehicle is charged within a preset time period, such as a week or a month, can be counted, the time intervals between these charging events can be calculated, and these time intervals can be used to define the charging frequency characteristics.

[0112] Optionally, the interval time between two consecutive charging behaviors depends on the travel behavior of the user, so the interval time between two consecutive charging behaviors can be calculated as the charging frequency feature.

[0113] Exemplarily, the charging frequency characteristic may be calculated according to formula (5).

[0114] Formula (5)

[0115] in, yes The end time of the charging behavior, It is The start time of the first charging behavior, The larger the value is, the longer the interval between two charging behaviors is.

[0116] Among them, the charging demand characteristics reflect the charging needs of electric vehicles.

[0117] Optionally, the current power of each electric vehicle is calculated, as well as the demand of the electric vehicle, and finally the difference between the demand power and the current power is calculated to obtain the charging demand characteristics.

[0118] Exemplarily, the calculation process of the charging demand characteristic is shown in formula (6).

[0119] Formula (6)

[0120] In the above embodiment, the charging behavior of the electric vehicle can be fully described by extracting features such as energy reception, energy return, electricity price sensitivity, charging frequency, and charging demand. These features can reflect the vehicle's energy usage pattern, response to electricity prices, charging frequency, and demand for electricity.

[0121] In one embodiment, the above prediction based on each target charging feature to obtain the charging load of each category of electric vehicles in the next time period is obtained through a pre-trained charging load prediction model; the training process of the charging load prediction model includes: obtaining charging feature samples; the features carry charging load labels; inputting the charging feature samples into the initial model for training to obtain the predicted charging load; obtaining the target loss function based on the predicted charging load and the charging load label, and adjusting the parameters of the initial model according to the target loss function until the training is completed to obtain the charging load pre-model.

[0122] First, feature samples are extracted from the historical charging data of electric vehicles. These samples include various charging features, such as energy receiving features, energy return features, electricity price sensitivity features, charging frequency features, and charging demand features.

[0123] Each charging feature sample carries a charging load tag, which indicates the actual charging load corresponding to the feature. The charging load tag is obtained from historical data and represents the actual charging demand of the electric vehicle group under specific conditions.

[0124] Then, these charging feature samples are input into an initial prediction model for model training. The model will output the predicted charging load based on the input feature samples, compare the predicted charging load with the actual charging load label, and calculate the target loss function. The target loss function measures the difference between the predicted value and the true value. According to the value of the target loss function, the parameters of the initial model are adjusted using an optimization algorithm (such as gradient descent). This process will continue until the loss function reaches the minimum value or converges, thus obtaining an optimized charging load prediction model.

[0125] Optionally, the initial prediction model can be LR (linear regression) and MLP (multi-layer perceptron), two classic feed-forward artificial neural networks.

[0126] In the above embodiment, by training the charging load prediction model, learning and optimization are performed based on actual historical data, so that the future charging load of electric vehicles can be predicted more accurately.

[0127] In one embodiment, the charging data can be obtained through the established charging process model. This is because the SOC at the start of charging, the SOC at the end of charging, the charging start time and the charging end time can be directly extracted through the charging data. Since the data of the charging process cannot be obtained. In this embodiment, price incentives are usually used to guide users to charge in an orderly manner, thereby bringing economic benefits to electric vehicle owners. On this basis, a charging process model is established to calculate the energy exchanged between the electric vehicle and the power grid during the charging process.

[0128] Formula (7)

[0129] Formula (8)

[0130] Formula (9)

[0131] Formula (10)

[0132] Formula (11)

[0133] Formula (12)

[0134] Formula (13)

[0135] In formula (7) Indicates the charging start time. Represents the charging stop time, They represent the power delivered by the grid to the electric vehicle and the power received by the electric vehicle from the grid. Representatives in The purchase price of electricity at the time, Representatives in The charging price at the time, is the time step. An indicator representing the current direction of energy flow, Represents the battery level, represents the energy transfer efficiency, represents the battery capacity. The objective function is to minimize the cost of purchasing electricity from the grid and maximize the income from selling energy to the grid during charging, thereby obtaining the maximum net profit (or minimizing the charging cost). Formulas (8)-(10) are the constraints for energy transmission between electric vehicles and the grid. Formulas (8) and (9) are the maximum power constraints for charging / transmission, and formula (10) is the constraint to ensure that charging and transmission points do not occur at the same time. Formula (11) is calculated in Event, time The battery SOC at the time, formula (12) is the battery power constraint, and formula (13) ensures that the SOC at the end of charging is equal to the expected SOC. This is because the user will have an expected power when charging, and here it is to ensure that the user's charging needs are met. The above objective function and constraints constitute the optimization problem in this embodiment. In most charging data, the charging start time, charging end time, and battery SOC at the start and end of charging are all available.

[0136] Furthermore, in this embodiment, the core idea is to reduce the dimension of the input data to estimate the charging load of all electric vehicles. In a time span, the charging data of all electric car owners can be obtained. Then, the entire available data set can be expressed as:

[0137] Formula (14)

[0138] In equation 14, the power is positive when the electric vehicle is charging, and the power is negative when the electric vehicle is discharging.

[0139] Formula (15)

[0140] in, It can be calculated by equations (7)-(13) or directly obtained from charging pile data. After clustering electric vehicle owners, the representative of each cluster can be found. In this patent, the electric vehicle owner with the smallest distance from the centroid is introduced as the cluster representative. Therefore, after grouping all electric vehicle owners, it is reasonable to estimate The simplified information set of an electric vehicle load can be expressed as:

[0141] Formula (16)

[0142] Exemplarily, further, two function approximation methods can be used to estimate the charging load of all electric vehicles using the charging data represented by J clusters. To this end, the parameters of the input-output mapping function are estimated to minimize the difference between the estimated charging load and the actual value. The actual load of the electric vehicle within time t can be expressed as:

[0143] Formula (17)

[0144] The mapping function can be expressed as:

[0145] Formula (18)

[0146] in and are the estimated model and the error respectively.

[0147] In the above embodiment, less electric vehicle data can be used to obtain a better generalized electric vehicle charging demand prediction model.

[0148] It should be understood that, although the various steps in the flowcharts involved in the above-mentioned embodiments are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-mentioned embodiments can include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.

[0149] Based on the same inventive concept, the embodiment of the present application also provides an electric vehicle charging load prediction device for implementing the electric vehicle charging load prediction method involved above. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the above method, so the specific limitations in one or more electric vehicle charging load prediction device embodiments provided below can refer to the limitations of the electric vehicle charging load prediction method above, and will not be repeated here.

[0150] In an exemplary embodiment, Figure 2 As shown, a device for predicting charging load of an electric vehicle is provided, comprising: an acquisition module 210, a feature extraction module 220, a classification module 230 and a prediction module 240, wherein:

[0151] The acquisition module 210 is used to acquire the charging data of the electric vehicle to be processed in the current time period.

[0152] The feature extraction module 220 is used to extract features from the charging data to obtain charging features.

[0153] The classification module 230 is used to classify each electric vehicle to be processed based on each charging feature, and determine a target charging feature for each category of the electric vehicle to be processed.

[0154] The prediction module 240 is used to make predictions based on each target charging feature to obtain the charging load of each category of electric vehicles in the next time period.

[0155] In one embodiment, the classification module 230 includes:

[0156] The cluster center selection unit is used to randomly select the charging characteristics of several electric vehicles to be processed as cluster centers.

[0157] The distance calculation unit is used to calculate the distance from the charging characteristics of other electric vehicles to be processed to the cluster center, and cluster the other electric vehicles to be processed according to the distance to obtain an initial category set.

[0158] The updating unit is used to update the cluster center based on the charging characteristics of each electric vehicle to be processed in the initial category set, and re-execute the clustering step until the iteration condition is met to obtain the target category set.

[0159] The feature acquisition unit is used to use the charging features of the electric vehicles to be processed corresponding to each cluster center in each target category set as the target charging features of each target category set.

[0160] In one embodiment, the feature extraction module 220 includes at least one or more of an energy receiving feature, an energy return feature, an electricity price sensitive feature, a charging frequency feature, and a charging demand feature; the energy receiving feature is used to characterize the total energy received by the electric vehicle to be processed from the power grid; the energy return feature is used to characterize the total energy returned to the power grid by the electric vehicle to be processed; the electricity price sensitive feature is used to characterize the sensitivity of the electric vehicle to be processed to the electricity price; the charging frequency feature is used to characterize the charging frequency of the electric vehicle to be processed within a preset time period; and the charging demand feature is used to characterize the demand for electricity by the electric vehicle to be processed.

[0161] In one embodiment, the distance calculation unit includes:

[0162] The component calculation subunit is used to respectively calculate the distance components of the energy receiving characteristics, energy return characteristics, electricity price sensitivity characteristics, charging frequency characteristics and charging demand characteristics of the charging characteristics of other electric vehicles to be processed to the energy receiving characteristics, energy return characteristics, electricity price sensitivity characteristics, charging frequency characteristics and charging demand characteristics of the electric vehicle to be processed corresponding to the corresponding cluster center.

[0163] The comprehensive calculation subunit is used to integrate various distance components to obtain the distance from the charging characteristics of other electric vehicles to be processed to the cluster center.

[0164] In one embodiment, the feature extraction module 220 includes:

[0165] The first extraction unit is used to count the total amount of energy received by each electric vehicle to be processed in each preset time interval to obtain energy reception characteristics; the total amount of energy received by the electric vehicle to be processed and sent to the power grid.

[0166] The second extraction unit is used to use the total returned energy of each processed electric vehicle as the energy return feature; the total returned energy is the total energy transmitted from the electric vehicle to be processed to the power grid.

[0167] The third extraction unit is used to calculate the ratio of the charging time of each processed electric vehicle to the preset charging time to obtain the electricity price sensitive feature.

[0168] The fourth extraction unit is used to obtain the charging frequency characteristics according to the time interval of the charging moments of each processed electric vehicle.

[0169] The fifth extraction unit is used to obtain the charging demand characteristics according to the difference between the required power and the current power of each processed electric vehicle.

[0170] In one embodiment, the apparatus comprises a training module, the training module comprising:

[0171] The sample acquisition unit is used to acquire charging feature samples; the features carry charging load tags.

[0172] The prediction unit is used to input the charging feature samples into the initial model for training to obtain the predicted charging load.

[0173] The training unit is used to obtain a target loss function based on the predicted charging load and the charging load label, and adjust the parameters of the initial model according to the target loss function until the training is completed to obtain a charging load pre-model.

[0174] Each module in the above-mentioned electric vehicle charging load prediction device can be implemented in whole or in part by software, hardware and a combination thereof. Each of the above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory in the computer device in the form of software, so that the processor can call and execute the corresponding operations of each of the above modules.

[0175] In an exemplary embodiment, a computer device is provided. The computer device may be a server, and its internal structure diagram may be as shown in FIG. Figure 3As shown. The computer device includes a processor, a memory, an input / output interface (Input / Output, referred to as I / O) and a communication interface. Among them, the processor, the memory and the input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. Among them, 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 charging data. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a method for predicting the charging load of an electric vehicle is implemented.

[0176] Those skilled in the art will understand that Figure 3 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0177] In an exemplary embodiment, a computer device is provided, including a memory and a processor, wherein a computer program is stored in the memory, and the processor implements the following steps when executing the computer program: obtaining charging data of an electric vehicle to be processed in a current time period; performing feature extraction on each charging data to obtain a charging feature; classifying each electric vehicle to be processed based on each charging feature, and determining a target charging feature based on each category of the electric vehicle to be processed; and performing prediction based on each target charging feature to obtain a charging load of each category of electric vehicles in the next time period.

[0178] In one embodiment, the processor further implements the following steps when executing the computer program: randomly selecting the charging features of several electric vehicles to be processed as cluster centers; calculating the distance from the charging features of other electric vehicles to be processed to the cluster centers, and clustering the other electric vehicles to be processed according to the distance to obtain an initial category set; updating the cluster centers based on the charging features of each electric vehicle to be processed in the initial category set, and re-executing the clustering steps until the iteration condition is met to obtain a target category set; and using the charging features of the electric vehicles to be processed corresponding to each cluster center in each target category set as the target charging features of each target category set.

[0179] In one embodiment, the processor further implements the following steps when executing the computer program: the charging characteristics include at least one or more of energy receiving characteristics, energy return characteristics, electricity price sensitivity characteristics, charging frequency characteristics and charging demand characteristics; the energy receiving characteristics are used to characterize the total energy received by the electric vehicle to be processed from the power grid; the energy return characteristics are used to characterize the total energy returned to the power grid by the electric vehicle to be processed; the electricity price sensitivity characteristics are used to characterize the sensitivity of the electric vehicle to be processed to the electricity price; the charging frequency characteristics are used to characterize the charging frequency of the electric vehicle to be processed within a preset time period; and the charging demand characteristics are used to characterize the demand for electricity by the electric vehicle to be processed.

[0180] In one embodiment, when the processor executes the computer program, the following steps are also implemented: respectively calculating the distance components of the energy receiving characteristics, energy return characteristics, electricity price sensitivity characteristics, charging frequency characteristics and charging demand characteristics of the charging characteristics of other electric vehicles to be processed to the energy receiving characteristics, energy return characteristics, electricity price sensitivity characteristics, charging frequency characteristics and charging demand characteristics of the electric vehicle to be processed corresponding to the corresponding cluster center; and combining the distance components to obtain the distance from the charging characteristics of other electric vehicles to be processed to the cluster center.

[0181] In one embodiment, the processor also implements the following steps when executing the computer program: counting the total amount received by each to-be-processed electric vehicle within each preset time interval to obtain energy reception characteristics; the total amount received is the total energy received by the to-be-processed electric vehicle to the power grid; taking the total returned energy of each processed electric vehicle as the energy return characteristic; the total returned energy is the total energy transmitted by the to-be-processed electric vehicle to the power grid; calculating the ratio of the charging time of each processed electric vehicle to the preset charging time to obtain the electricity price sensitivity characteristic; obtaining the charging frequency characteristic based on the time interval between the charging moments of each processed electric vehicle; and obtaining the charging demand characteristic based on the difference between the required power of each processed electric vehicle and the current power.

[0182] In one embodiment, when the processor executes the computer program, the following steps are also implemented: obtaining charging feature samples; the features carry charging load labels; inputting the charging feature samples into the initial model for training to obtain predicted charging load; obtaining the target loss function based on the predicted charging load and the charging load label, and adjusting the parameters of the initial model according to the target loss function until the training is completed to obtain a charging load pre-model.

[0183] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the following steps are implemented: obtaining charging data of an electric vehicle to be processed in a current time period; performing feature extraction on each charging data to obtain a charging feature; classifying each electric vehicle to be processed based on each charging feature, and determining a target charging feature based on each category of the electric vehicle to be processed; and performing prediction based on each target charging feature to obtain a charging load of each category of electric vehicles in the next time period.

[0184] In one embodiment, when the computer program is executed by the processor, the following steps are also implemented: randomly selecting the charging characteristics of several electric vehicles to be processed as cluster centers; calculating the distance from the charging characteristics of other electric vehicles to be processed to the cluster centers, and clustering the other electric vehicles to be processed according to the distance to obtain an initial category set; updating the cluster centers based on the charging characteristics of each electric vehicle to be processed in the initial category set, and re-executing the clustering step until the iteration condition is met to obtain a target category set; and using the charging characteristics of the electric vehicles to be processed corresponding to each cluster center in each target category set as the target charging characteristics of each target category set.

[0185] In one embodiment, when the computer program is executed by the processor, the following steps are also implemented: the charging characteristics include at least one or more of energy reception characteristics, energy return characteristics, electricity price sensitivity characteristics, charging frequency characteristics and charging demand characteristics; the energy reception characteristics are used to characterize the total energy received by the electric vehicle to be processed from the power grid; the energy return characteristics are used to characterize the total energy returned to the power grid by the electric vehicle to be processed; the electricity price sensitivity characteristics are used to characterize the sensitivity of the electric vehicle to be processed to the electricity price; the charging frequency characteristics are used to characterize the charging frequency of the electric vehicle to be processed within a preset time period; the charging demand characteristics are used to characterize the demand for electricity by the electric vehicle to be processed.

[0186] In one embodiment, when the computer program is executed by the processor, the following steps are also implemented: respectively calculating the distance components of the energy receiving characteristics, energy return characteristics, electricity price sensitivity characteristics, charging frequency characteristics and charging demand characteristics of the charging characteristics of other electric vehicles to be processed to the energy receiving characteristics, energy return characteristics, electricity price sensitivity characteristics, charging frequency characteristics and charging demand characteristics of the electric vehicle to be processed corresponding to the corresponding cluster center; combining the distance components to obtain the distance from the charging characteristics of other electric vehicles to be processed to the cluster center.

[0187] In one embodiment, when the computer program is executed by the processor, the following steps are also implemented: the total amount of energy received by each electric vehicle to be processed in each preset time interval is counted to obtain energy reception characteristics; the total amount of energy received is the total energy received by the electric vehicle to be processed and sent to the power grid; the total energy returned by each processed electric vehicle is used as the energy return characteristic; the total energy returned is the total energy transmitted by the electric vehicle to be processed and sent to the power grid; the ratio of the charging time of each processed electric vehicle and the preset charging time is calculated to obtain the electricity price sensitivity characteristic; the charging frequency characteristic is obtained according to the time interval between the charging moments of each processed electric vehicle; the charging demand characteristic is obtained according to the difference between the required power of each processed electric vehicle and the current power.

[0188] In one embodiment, when the computer program is executed by the processor, the following steps are also implemented: obtaining charging feature samples; the features carry charging load labels; inputting the charging feature samples into the initial model for training to obtain predicted charging load; obtaining the target loss function based on the predicted charging load and the charging load label, and adjusting the parameters of the initial model according to the target loss function until the training is completed to obtain a charging load pre-model.

[0189] In one embodiment, a computer program product is provided, comprising a computer program, which implements the following steps when executed by a processor: obtaining charging data of an electric vehicle to be processed in a current time period; performing feature extraction on each charging data to obtain a charging feature; classifying each electric vehicle to be processed based on each charging feature, and determining a target charging feature based on each category of the electric vehicle to be processed; and performing prediction based on each target charging feature to obtain a charging load of each category of electric vehicles in the next time period.

[0190] In one embodiment, when the computer program is executed by the processor, the following steps are also implemented: randomly selecting the charging characteristics of several electric vehicles to be processed as cluster centers; calculating the distance from the charging characteristics of other electric vehicles to be processed to the cluster centers, and clustering the other electric vehicles to be processed according to the distance to obtain an initial category set; updating the cluster centers based on the charging characteristics of each electric vehicle to be processed in the initial category set, and re-executing the clustering step until the iteration condition is met to obtain a target category set; and using the charging characteristics of the electric vehicles to be processed corresponding to each cluster center in each target category set as the target charging characteristics of each target category set.

[0191] In one embodiment, when the computer program is executed by the processor, the following steps are also implemented: the charging characteristics include at least one or more of energy reception characteristics, energy return characteristics, electricity price sensitivity characteristics, charging frequency characteristics and charging demand characteristics; the energy reception characteristics are used to characterize the total energy received by the electric vehicle to be processed from the power grid; the energy return characteristics are used to characterize the total energy returned to the power grid by the electric vehicle to be processed; the electricity price sensitivity characteristics are used to characterize the sensitivity of the electric vehicle to be processed to the electricity price; the charging frequency characteristics are used to characterize the charging frequency of the electric vehicle to be processed within a preset time period; the charging demand characteristics are used to characterize the demand for electricity by the electric vehicle to be processed.

[0192] In one embodiment, when the computer program is executed by the processor, the following steps are also implemented: respectively calculating the distance components of the energy receiving characteristics, energy return characteristics, electricity price sensitivity characteristics, charging frequency characteristics and charging demand characteristics of the charging characteristics of other electric vehicles to be processed to the energy receiving characteristics, energy return characteristics, electricity price sensitivity characteristics, charging frequency characteristics and charging demand characteristics of the electric vehicle to be processed corresponding to the corresponding cluster center; combining the distance components to obtain the distance from the charging characteristics of other electric vehicles to be processed to the cluster center.

[0193] In one embodiment, when the computer program is executed by the processor, the following steps are also implemented: the total amount of energy received by each electric vehicle to be processed in each preset time interval is counted to obtain energy reception characteristics; the total amount of energy received is the total energy received by the electric vehicle to be processed and sent to the power grid; the total energy returned by each processed electric vehicle is used as the energy return characteristic; the total energy returned is the total energy transmitted by the electric vehicle to be processed and sent to the power grid; the ratio of the charging time of each processed electric vehicle and the preset charging time is calculated to obtain the electricity price sensitivity characteristic; the charging frequency characteristic is obtained according to the time interval between the charging moments of each processed electric vehicle; the charging demand characteristic is obtained according to the difference between the required power of each processed electric vehicle and the current power.

[0194] In one embodiment, when the computer program is executed by the processor, the following steps are also implemented: obtaining charging feature samples; the features carry charging load labels; inputting the charging feature samples into the initial model for training to obtain predicted charging load; obtaining the target loss function based on the predicted charging load and the charging load label, and adjusting the parameters of the initial model according to the target loss function until the training is completed to obtain a charging load pre-model.

[0195] A person of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment method can be completed by instructing the relevant hardware through a computer program, and 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 the memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. Non-relational databases may include distributed databases based on blockchains, etc., but are not limited to this. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, an artificial intelligence (AI) processor, etc., but are not limited to this.

[0196] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0197] The above-described embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. A method for predicting electric vehicle charging load, characterized in that: The method comprises: Obtain charging data of the electric vehicle to be processed in the current time period; Extracting features from each of the charging data to obtain charging features; Classifying each of the electric vehicles to be processed based on each of the charging characteristics, and determining a target charging characteristic based on each category of the electric vehicles to be processed; Predictions are made based on the target charging characteristics to obtain the charging load of each category of electric vehicles in the next time period.

2. The method according to claim 1, characterized in that The step of classifying each of the electric vehicles to be processed based on the charging characteristics and determining a target charging characteristic for each category of the electric vehicles to be processed includes: Randomly selecting a number of charging characteristics of the electric vehicles to be processed as cluster centers; Calculating the distance between the charging characteristics of the other electric vehicles to be processed and the cluster center, and clustering the other electric vehicles to be processed according to the distance to obtain an initial category set; The cluster center is updated based on the charging characteristics of each of the electric vehicles to be processed in the initial category set, and the clustering step is re-executed until the iteration condition is met to obtain the target category set; The charging features of the electric vehicles to be processed corresponding to each of the cluster centers in each of the target category sets are used as target charging features of each of the target category sets.

3. The method according to claim 1 or 2, characterized in that: The charging characteristics include at least one or more of energy receiving characteristics, energy return characteristics, electricity price sensitivity characteristics, charging frequency characteristics, and charging demand characteristics; The energy receiving characteristic is used to characterize the total energy received by the electric vehicle to be processed from the power grid; The energy return feature is used to characterize the total energy returned to the power grid by the electric vehicle to be processed; The electricity price sensitivity feature is used to characterize the sensitivity of the electric vehicle to be processed to the electricity price; The charging frequency feature is used to characterize the charging frequency of the electric vehicle to be processed within a preset time period; The charging demand feature is used to characterize the demand for electric energy of the electric vehicle to be processed.

4. The method according to claim 3, characterized in that The calculating the distance from the charging characteristics of the other electric vehicles to be processed to the cluster center includes: Respectively calculate the distance components of the energy receiving feature, the energy return feature, the electricity price sensitive feature, the charging frequency feature and the charging demand feature in the charging features of the other electric vehicles to be processed to the energy receiving feature, the energy return feature, the electricity price sensitive feature, the charging frequency feature and the charging demand feature of the electric vehicle to be processed corresponding to the corresponding cluster center; The distance components are integrated to obtain the distances from the charging characteristics of other electric vehicles to be processed to the cluster center.

5. The method according to claim 3, characterized in that: The charging data includes charging time, total energy received, total energy returned, charging time, required power and current power; The extracting features of each charging data to obtain charging features includes: Counting the total amount of energy received by each of the electric vehicles to be processed in each preset time interval to obtain the energy receiving characteristics; the total amount of energy received by the electric vehicle to be processed and the power grid; The total energy returned by each of the processed electric vehicles is used as the energy return feature; the total energy returned is the total energy transmitted by the electric vehicle to be processed to the power grid; Calculating the ratio of the charging time of each processed electric vehicle to the preset charging time to obtain the electricity price sensitive feature; Obtaining the charging frequency characteristic according to the time interval of the charging moment of each processing electric vehicle; The charging demand characteristic is obtained according to the difference between the required power and the current power of each processed electric vehicle.

6. The method according to claim 1, characterized in that The prediction based on each of the target charging characteristics is performed to obtain the charging load of each category of electric vehicles in the next time period through a pre-trained charging load prediction model; The training process of the charging load prediction model includes: Acquire a charging feature sample; the feature carries a charging load tag; Inputting the charging characteristic samples into an initial model for training to obtain a predicted charging load; A target loss function is obtained based on the predicted charging load and the charging load label, and the parameters of the initial model are adjusted according to the target loss function until the training is completed to obtain the charging load pre-model.

7. An electric vehicle charging load prediction device, characterized in that: The device comprises: An acquisition module is used to acquire charging data of the electric vehicle to be processed in the current time period; A feature extraction module, used to extract features from the charging data to obtain charging features; A classification module, used for classifying each of the electric vehicles to be processed based on each of the charging characteristics, and determining a target charging characteristic for each category of the electric vehicles to be processed; The prediction module is used to make predictions based on the target charging characteristics to obtain the charging load of each category of electric vehicles in the next time period.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 6 are implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images