Charging quantity prediction method and device, electronic equipment and storage medium

Abstract

The embodiment of the invention discloses a charging quantity prediction method and device, electronic equipment and a storage medium. The method comprises the steps that a target prediction area, a target prediction time period and n vehicle types input by a user are acquired; acquiring historical vehicle energy consumption data, historical environment temperature data and charging pile data of the target prediction area in a historical time period; calculating a temperature correction coefficient of the target prediction area in the target prediction time period according to the historical vehicle energy consumption data and the historical environment temperature data; calculating a charging loss coefficient of the target prediction area in the target prediction time period according to the historical charging pile data; predicting the number of vehicles of each vehicle type in the target prediction region within the target prediction time period to obtain n vehicle numbers; and according to the temperature correction coefficient, the charging loss coefficient and the number of the n vehicles, predicting the target charging amount of each vehicle type in the target prediction region within the target prediction time period. According to the embodiment of the invention, the prediction precision of the charging amount can be improved.

Description

Technical Field

[0001] This invention relates to the field of power technology, and in particular to a method, apparatus, electronic device, and storage medium for predicting charging capacity. Background Technology

[0002] In recent years, with the rapid growth in the number of electric vehicles, the demand for charging has exploded, posing a huge challenge to power grid planning and charging infrastructure construction. In order to accurately match the supply of charging resources with actual demand and avoid problems such as regional power grid overload, idle and wasted charging facilities, or supply shortages caused by supply-demand imbalances, it is urgent to predict the charging demand of electric vehicles.

[0003] However, traditional forecasting methods often use only one parameter, which leads to a large deviation between the forecast results and actual needs, resulting in low forecast accuracy. Summary of the Invention

[0004] To address the aforementioned problems, embodiments of the present invention provide a charging quantity prediction method, apparatus, electronic device, and storage medium, which can improve the accuracy of charging quantity prediction when predicting the charging quantity required by a vehicle.

[0005] In a first aspect, embodiments of the present invention provide a method for predicting charging capacity, including: Obtain the target prediction area, target prediction time period, and n vehicle types input by the user; Acquire historical vehicle energy consumption data, historical ambient temperature data, and charging pile data for the target prediction area within a historical time period; Based on the historical vehicle energy consumption data and the historical ambient temperature data, calculate the temperature correction coefficient for the target prediction area during the target prediction time period; Based on the historical charging pile data, calculate the charging loss coefficient of the target prediction area during the target prediction time period; Predict the number of vehicles of each vehicle type in the target prediction area within the target prediction time period to obtain n vehicle counts; Based on the temperature correction coefficient, the charging loss coefficient, and the number of n vehicles, the target charging amount for each vehicle type in the target prediction area during the target prediction time period is predicted.

[0006] Secondly, embodiments of the present invention provide a charging quantity prediction device, the device comprising an acquisition unit and a processing unit; The acquisition unit is used to acquire the target prediction area, target prediction time period, and n types of vehicles input by the user. Acquire historical vehicle energy consumption data, historical ambient temperature data, and charging pile data for the target prediction area within a historical time period; The processing unit is used to calculate the temperature correction coefficient of the target prediction area during the target prediction time period based on the historical vehicle energy consumption data and the historical ambient temperature data; the temperature correction coefficient is used to correct the deviation between the actual energy consumption and the standard energy consumption of the vehicle under different temperature environments. Based on the historical charging pile data, the charging loss coefficient of the target prediction area during the target prediction time period is calculated; the charging loss coefficient is used to quantify the proportion of energy loss during the charging process. Predict the number of vehicles of each vehicle type in the target prediction area within the target prediction time period to obtain n vehicle counts; Based on the temperature correction coefficient, the charging loss coefficient, and the number of n vehicles, the target charging amount for each vehicle type in the target prediction area during the target prediction time period is predicted.

[0007] Thirdly, embodiments of the present invention provide an electronic device, the electronic device including a processor and a memory, the processor being connected to the memory, the memory being used to store a computer program, and the processor being used to execute the computer program stored in the memory, so that the electronic device performs the method as described in the first aspect.

[0008] Fourthly, embodiments of the present invention provide a computer-readable storage medium storing a computer program that is executed by a processor to implement the method described in the first aspect.

[0009] Fifthly, embodiments of this application provide a computer program product, the computer program product including a non-transitory computer-readable storage medium storing a computer program, the computer being operable to perform the method as described in the first aspect.

[0010] Implementing the embodiments of this application has the following beneficial effects: In this embodiment, the target prediction area, target prediction time period, and n vehicle types input by the user are first obtained. Historical vehicle energy consumption data, historical ambient temperature data, and charging pile data for the target prediction area within the historical time period are also obtained. Then, based on the historical vehicle energy consumption data and historical ambient temperature data, a temperature correction coefficient for the target prediction area within the target prediction time period is calculated. Next, the number of vehicles of each vehicle type in the target prediction area within the target prediction time period is predicted, resulting in n vehicle numbers. Finally, based on the temperature correction coefficient, charging loss coefficient, and the n vehicle numbers, the target charging amount for each vehicle type in the target prediction area within the target prediction time period is predicted. Therefore, by calculating the corresponding temperature correction coefficient and charging loss coefficient, and classifying and predicting vehicles within the target prediction area based on these coefficients, unlike traditional single-parameter predictions, the accuracy of charging amount prediction can be improved when predicting vehicle charging demand. Attached Figure Description

[0011] To more clearly illustrate the technical solutions in the embodiments of the present invention or the background art, the drawings used in the embodiments of the present invention or the background art will be described below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0012] Figure 1 This is a schematic diagram of the architecture of a charging capacity prediction system provided in an embodiment of this application; Figure 2 This is a schematic diagram of a charging quantity prediction module provided in an embodiment of this application; Figure 3 This is a flowchart of a charging quantity prediction method provided in an embodiment of this application; Figure 4 This is a flowchart of a method for determining the charging loss coefficient provided in an embodiment of this application; Figure 5 This is a schematic diagram of multidimensional data provided in an embodiment of this application; Figure 6 This is a schematic diagram illustrating a result visualization provided in an embodiment of this application; Figure 7 This is a schematic diagram of the structure of a charging quantity prediction device provided in an embodiment of this application; Figure 8 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0013] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0014] The terms "first," "second," "third," and "fourth," etc., used in the specification, claims, and accompanying drawings of this application are used to distinguish different objects, not to describe a specific order. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or device that includes a series of steps or modules is not limited to the listed steps or modules, but may optionally include steps or modules not listed, or may optionally include other steps or modules inherent to these processes, methods, products, or devices.

[0015] In this document, the term "embodiment" means that a particular feature, result, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0016] The following describes the relevant content, concepts, technical issues, technical solutions, and beneficial effects involved in the embodiments of this application.

[0017] See Figure 1 , Figure 1 This is a schematic diagram of the architecture of a charging capacity prediction system provided in an embodiment of this application. The system specifically includes a server and an interactive interface. The interactive interface is used to acquire user input data, and the server is used to interact with the interactive interface, process the user input data, and return the processing result to the interactive interface.

[0018] See Figure 2 , Figure 2This is a schematic diagram of a charging demand prediction module provided in an embodiment of this application. Specifically, it includes a data acquisition module, a data processing module, a model building module, a prediction calculation module, and a result output module. The data acquisition module is used to acquire multi-source basic data such as vehicle classification, operation, ownership, environment, and charging facilities; the data processing module is used to process the acquired data; the model building module, based on the processed data, constructs a dedicated charging demand prediction model for different vehicle types, adapting to the operating characteristics of various vehicles; the prediction calculation module can call the corresponding model and, combined with user-set parameters such as time and region, complete the calculation and summarization of charging demand; the result output module presents the prediction results to the user in an intuitive format.

[0019] See Figure 3 , Figure 3 This is a flowchart illustrating a charging capacity prediction method provided in an embodiment of this application. The charging capacity prediction method provided in this application includes, but is not limited to, the following steps: Step S101: Obtain the target prediction area, target prediction time period, and n vehicle types input by the user; Step S102: Obtain historical vehicle energy consumption data, historical ambient temperature data, and charging pile data for the target prediction area within a historical time period; Step S103: Calculate the temperature correction coefficient for the target prediction area during the target prediction period based on historical vehicle energy consumption data and historical ambient temperature data; Step S104: Calculate the charging loss coefficient of the target prediction area during the target prediction time period based on historical charging pile data; Step S105: Predict the number of vehicles of each vehicle type in the target prediction area within the target prediction time period, and obtain n vehicle counts; Step S106: Based on the temperature correction coefficient, charging loss coefficient, and the number of n vehicles, predict the target charging amount for each vehicle type in the target prediction area within the target prediction time period.

[0020] In one possible embodiment, the system acquires the target prediction area, target prediction time period, and n vehicle types input by the user. The target prediction area refers to the geographical range specified by the user for which charging volume prediction is required; the target prediction time period is the prediction time interval set by the user, which can be multiple time dimensions such as year, quarter, month, week, day, or even a specific hour within a day; the n vehicle types are the categories of electric vehicles selected by the user for prediction, which can include buses, taxis, ride-hailing vehicles, logistics vehicles, private cars, official vehicles, etc., where n is a positive integer greater than or equal to 1. Specifically, the system receives user-inputted regional boundary parameters, such as latitude and longitude ranges and location names, as well as time start and end information and vehicle type identifiers, through an interactive interface.

[0021] In one possible embodiment, historical vehicle energy consumption data, historical ambient temperature data, and charging pile data for the target prediction area within a historical time period are obtained. The historical time period is a past time interval that is concurrent with or related to the target prediction time period. For example, if the target prediction time period is the summer of 2025, the historical time period can be selected from the concurrent data of the summers of 2023 and 2024, as well as the data of the spring of 2025. Among them, historical vehicle energy consumption data are the actual energy consumption records of n types of vehicles within the target prediction area during historical time periods, including data such as energy consumption per 100 kilometers, daily average energy consumption, and energy consumption fluctuations under different operating conditions. This data is obtained through channels such as connecting to vehicle management department databases, vehicle-mounted terminal data collection systems, and charging facility operation platforms. Historical ambient temperature data are meteorological temperature records of the target prediction area during historical time periods, including daily average temperature, monthly average temperature, and extreme temperature values. This data is collected through public interfaces of meteorological departments and regional environmental monitoring terminals. Charging pile data includes basic information and operational data of charging piles within the target prediction area during historical time periods, including charging pile types such as fast charging piles and slow charging piles, quantity, distribution location, charging efficiency, and usage frequency. This data is obtained synchronously through channels such as public charging facility management platforms and private charging facility registration systems.

[0022] In one possible embodiment, a temperature correction coefficient for the target prediction area within the target prediction time period is calculated based on historical vehicle energy consumption data and historical ambient temperature data. The temperature correction coefficient is a parameter used to calibrate the impact of ambient temperature on vehicle energy consumption, correcting the deviation between actual and standard energy consumption under different temperature conditions. Since battery performance is significantly affected by temperature, low or high temperatures can lead to increased vehicle energy consumption. Specifically, the baseline energy consumption per 100 kilometers for each of the n vehicle types is extracted from historical vehicle energy consumption data, and the average temperature for the corresponding time period is extracted from historical ambient temperature data. A nonlinear relationship model between historical temperature and energy consumption deviation is established using statistical regression analysis to quantify the impact of different temperature ranges on vehicle energy consumption. For example, analysis of historical data shows that outside the normal temperature range of 10℃-28℃, for every 5℃ change in temperature, the average energy consumption per 100 kilometers increases by 2.4 kWh. The predicted temperature data for the target prediction time period is then substituted into a preset temperature correction coefficient calculation formula. Where T is the target predicted temperature. For reference, a normal temperature, for example, 19℃, is the midpoint between 10℃ and 28℃. The temperature correction coefficient corresponding to the target prediction time period is calculated based on the vehicle's energy consumption per 100 kilometers.

[0023] In one possible embodiment, the charging loss coefficient for the target prediction area during the target prediction time period is calculated based on historical charging pile data. The charging loss coefficient is a parameter characterizing the proportion of energy loss during charging. Energy loss mainly includes the charging pile's own losses and power transmission losses, and is related to the type of charging pile and the charging method. Specifically, based on historical charging pile data, the proportion of different types of charging piles and their respective basic charging efficiencies are statistically analyzed within the target prediction area. The charging efficiency of fast charging piles is typically 75%-80%, and that of slow charging piles is typically 85%-90%. A weighted average method is used to calculate the overall charging efficiency of the area, with the weights being the usage frequency or quantity proportion of each type of charging pile. The charging loss coefficient is derived from the overall charging efficiency, and the calculation formula is as follows: =1 / charging efficiency-1. For example, if the overall charging efficiency is 86.25%, then the charging loss coefficient is about 15%.

[0024] In one possible implementation, the number of vehicles of each vehicle type in the target prediction area within the target prediction time period is predicted to obtain n vehicle counts. Specifically, historical inventory data and growth trend data of n vehicle types in the target prediction area are obtained. Using prediction models such as linear regression and exponential growth, the inventory base for the target prediction time period is determined. The inventory base is then adjusted based on factors such as the infrastructure of the target prediction area, including the number of new charging piles and bus station expansion plans. For operational vehicles, such as taxis, ride-hailing vehicles, and logistics vehicles, the number of active vehicles within the target time period is predicted by combining historical operational data, order volume trends, and freight volume fluctuations. For non-operational vehicles such as private cars and official vehicles, the prediction results are optimized by combining user travel habits and seasonal travel characteristics, ultimately obtaining the number of each vehicle type within the target prediction time period.

[0025] In one possible embodiment, the target charging amount for each vehicle type in the target prediction area during the target prediction period is predicted based on a temperature correction coefficient, a charging loss coefficient, and the number of n vehicles. The target charging amount is the total charging demand for each vehicle type during the target prediction period. For each vehicle type, considering its average daily mileage and basic energy consumption per 100 kilometers, the average daily energy consumption per vehicle is first calculated: Average daily energy consumption = Average daily mileage × Energy consumption per 100 kilometers / 100. Then, the average daily energy consumption per vehicle is multiplied by the number of vehicles of that type and the number of days in the target prediction period to obtain the basic charging amount. Subsequently, the basic charging amount is multiplied by the temperature correction coefficient and (1 + charging loss coefficient) to correct for temperature effects and charging losses, resulting in the target charging amount for that vehicle type. Finally, the target charging amounts for all n vehicle types are summed to obtain the total charging demand for the target prediction area during the target prediction period.

[0026] In this embodiment, by integrating multi-dimensional data fusion and coefficient calibration, and taking into account key factors such as differences in vehicle type, influence of ambient temperature, and charging loss, the accuracy of charging quantity prediction is improved, and the prediction precision of charging quantity is increased.

[0027] Optionally, step S103, calculating the temperature correction coefficient for the target prediction area during the target prediction period based on historical vehicle energy consumption data and historical ambient temperature data, may include the following steps: Step S201: Based on historical vehicle energy consumption data and historical ambient temperature data, determine the first mapping relationship between the actual change in vehicle energy consumption and the change in ambient temperature. Step S202: Obtain the predicted temperature of the target prediction area during the target prediction time period; Step S203: Calculate the temperature difference between the predicted temperature and the preset reference temperature; Step S204: Determine the temperature correction coefficient based on the temperature difference and the first mapping relationship.

[0028] In one possible embodiment, a first mapping relationship between the actual change in vehicle energy consumption and the change in ambient temperature is determined based on historical vehicle energy consumption data and historical ambient temperature data. Here, the actual change in vehicle energy consumption is the difference between the actual energy consumption and the standard energy consumption of the vehicle under different ambient temperatures within a historical period; the standard energy consumption is the calibrated energy consumption of the vehicle under ideal ambient temperature conditions; the change in ambient temperature is the difference between the ambient temperature at different observation points within a historical period and a preset reference temperature; and the first mapping relationship is a functional relationship or corresponding rule characterizing the change in actual vehicle energy consumption with the change in ambient temperature, used to quantify the impact of temperature fluctuations on vehicle energy consumption. Specifically, the actual energy consumption data per 100 kilometers for each of the n vehicle types at different historical time points is extracted from historical vehicle energy consumption data. Similarly, ambient temperature data at corresponding time points is extracted from historical ambient temperature data, ensuring a one-to-one correspondence between the energy consumption data and temperature data in terms of time dimensions. Using a preset benchmark temperature (typically the median of a relatively stable ambient temperature range, such as 19℃), the ambient temperature change at each time point is calculated—the difference between that temperature and the benchmark temperature. The actual energy consumption change at the corresponding time point is also calculated—the difference between the actual energy consumption per 100 kilometers and the standard energy consumption per 100 kilometers. Modeling and analysis of these two sets of data are performed. After removing outlier data points, a numerical relationship model between the actual energy consumption change and the ambient temperature change is fitted. This model is the first mapping relationship. For example, by analyzing historical data, a linear mapping relationship can be obtained where, for every 1℃ change in ambient temperature, the average energy consumption per 100 kilometers changes by 0.48 kWh. Alternatively, a piecewise nonlinear mapping relationship can be established based on the energy consumption change patterns in different temperature ranges.

[0029] In one possible embodiment, the predicted temperature of the target prediction area within the target prediction time period is obtained. When the target prediction time period is monthly, the predicted temperature can be the daily average predicted temperature or the monthly average predicted temperature for that month; if it is a prediction for a specific time period within a day, the predicted temperature can be the hourly predicted temperature for that time period. Predicted temperature data at the corresponding time granularity within the target prediction time period is obtained by connecting to a professional temperature prediction interface of the meteorological department, calling a regional meteorological prediction model, or by extrapolating based on historical temperature data from the same period combined with seasonal variation trends and climate anomaly factors.

[0030] In one possible embodiment, the temperature difference between the predicted temperature and the preset reference temperature is calculated. The temperature difference refers to the difference between the predicted temperature and the preset reference temperature during the target prediction period, and is used to quantify the degree of deviation of the temperature during the target prediction period from the ideal ambient temperature.

[0031] In one possible embodiment, a temperature correction coefficient is determined based on the temperature difference and a first mapping relationship. The calculated temperature difference is substituted into the established first mapping relationship, and the corresponding predicted value of the actual energy consumption change of the vehicle is calculated through the mapping relationship model. Subsequently, combined with the vehicle's standard energy consumption per 100 kilometers, the temperature correction coefficient is derived according to a preset temperature correction coefficient calculation formula. For example, based on the rule of temperature correction coefficient = 1 + (temperature difference × unit temperature energy consumption change) / standard energy consumption per 100 kilometers, if the unit temperature energy consumption change is 0.48 kWh / ℃, the standard energy consumption per 100 kilometers is 105 kWh, and the temperature difference is 11℃, then the temperature correction coefficient = 1 + (11 × 0.48) / 105 ≈ 1.050. This coefficient represents the correction ratio of the vehicle's energy consumption relative to the standard energy consumption at the target predicted temperature, and is used for temperature-based calibration of energy consumption data in subsequent charging calculations.

[0032] In this embodiment, by constructing a quantitative mapping relationship between temperature and energy consumption, and combining the predicted temperature for the target time period to calculate the temperature correction coefficient, the accuracy of temperature-induced calibration is improved, and the accuracy of the charging quantity prediction results is increased.

[0033] Optional, see below Figure 4 , Figure 4 This is a flowchart of a method for determining the charging loss coefficient provided in an embodiment of this application. Step S104, calculating the charging loss coefficient of the target prediction area within the target prediction time period based on historical charging pile data, may include the following steps: Step S301: Based on historical charging pile data, determine the number of charging piles corresponding to each type of charging pile within the target prediction area; Step S302: Obtain the charging efficiency of each charging station; Step S303: Determine the charging efficiency of each type of charging pile based on the charging efficiency of each charging pile; Step S304: Determine the usage ratio of each charging pile type based on the number of charging piles corresponding to each charging pile type; Step S305: Determine the charging loss coefficient based on the usage ratio and charging efficiency of each type of charging pile.

[0034] In one possible embodiment, the number of charging piles corresponding to each type within the target prediction area is determined based on historical charging pile data. Here, charging pile type refers to the category of charging piles classified according to charging power and charging method, which may include fast charging piles (high-power fast charging) and slow charging piles (low-power conventional charging); the number of charging piles refers to the actual number of charging piles of each type within the target prediction area. Specifically, basic information of all charging piles within the target prediction area is extracted from historical charging pile data, including charging pile type identification, installation location, registration and filing information, etc.; charging pile data outside the area is filtered by geographical scope; and the number of fast charging piles and slow charging piles is accumulated separately according to charging pile type to form the statistical results of the number of charging piles for each category, so that the quantity data corresponds to the target prediction area and charging pile type.

[0035] In one possible embodiment, the charging efficiency of each charging pile is obtained. Charging efficiency is the proportion of grid energy converted by the charging pile into energy that can be stored in the electric vehicle's battery; that is, the ratio of output energy to input energy. Its value is affected by factors such as the charging pile's hardware performance, service life, and charging conditions, and ranges from 0 to 1. Specifically, operational record data for each charging pile is extracted from historical charging pile data, including the input energy (energy obtained from the grid), output energy (energy charged into the vehicle's battery), and charging time for each charge. For each charging pile, the single-charge efficiency is calculated according to the rule: charging efficiency = single-charge output energy / single-charge input energy. Then, an average is obtained by statistical averaging, for example, taking the arithmetic mean of valid charging records over the past 3-6 months. For charging piles without direct operational records, the average charging efficiency of charging piles of the same type, brand, and service life is used as a reference value to ensure that each charging pile has corresponding charging efficiency data.

[0036] In one possible embodiment, the charging efficiency of each charging pile type is determined based on the charging efficiency of each charging pile. Here, the charging efficiency of each charging pile type refers to the average charging efficiency of a certain category of charging piles. Specifically, the charging efficiency of individual charging piles is grouped according to their charging pile type, that is, the charging efficiency of all fast charging piles is grouped into one group, and the charging efficiency of all slow charging piles is grouped into another group. A weighted average method is used to calculate the average charging efficiency of each group, and the weight can be set as the proportion of the historical usage frequency of an individual charging pile. The higher the usage frequency, the greater the weight. The average charging efficiency of the fast charging pile type and the average charging efficiency of the slow charging pile type are obtained, which are the charging efficiency of the corresponding charging pile type.

[0037] In one possible implementation, the usage ratio of each charging pile type is determined based on the number of charging piles corresponding to each type. The usage ratio is the proportion of the usage frequency of a certain type of charging pile to the total usage frequency of all charging piles in the target prediction area, or an equivalent usage ratio derived from the quantity ratio, used to reflect the actual usage intensity of different types of charging piles. Specifically, the total number of all charging piles in the target prediction area is counted, i.e., the sum of the number of fast charging piles and the number of slow charging piles; for each type of charging pile, its quantity ratio is calculated: quantity ratio = number of charging piles of that type / total number of charging piles; if historical charging pile data includes usage frequency data for each type of charging pile, such as average daily charging times and cumulative charging time, the actual usage ratio is calculated based on the usage frequency: usage ratio = total usage frequency of charging piles of that type / total usage frequency of all charging piles; if usage frequency data is lacking, the quantity ratio can be directly used as the equivalent usage ratio, or the quantity ratio can be corrected by combining regional charging demand characteristics, such as high slow charging frequency in residential areas and high fast charging frequency in commercial areas, so that the usage ratio can truly reflect the actual usage of various types of charging piles.

[0038] In one possible embodiment, a charging loss coefficient is determined based on the usage ratio and charging efficiency of each charging pile type. The charging loss coefficient is the proportion of energy loss to output energy during charging, used to quantify energy loss in the charging process. It is negatively correlated with charging efficiency; the higher the charging efficiency, the smaller the loss coefficient. Specifically, the charging efficiency of each type of charging pile is weighted and calculated to obtain the comprehensive charging efficiency of the target prediction area. The calculation formula is: Comprehensive charging efficiency = Σ(Usage ratio of a certain type of charging pile × Charging efficiency of that type of charging pile). The charging loss coefficient is derived from the comprehensive charging efficiency using the formula: Charging loss coefficient = (1 / Comprehensive charging efficiency) - 1. For example, if the comprehensive charging efficiency is 86.25%, then the charging loss coefficient = (1 / 0.8625) - 1 ≈ 0.1596, or 15.96%. This coefficient is used in subsequent charging quantity calculations to compensate for energy loss during the charging process, ensuring that the predicted charging demand covers the actual loss.

[0039] In this embodiment of the application, the number of charging piles is counted by type, the average charging efficiency of a single pile and type is calculated, and the comprehensive charging efficiency and loss coefficient are calculated by using a weighted average, thereby quantifying the charging loss and improving the accuracy of the loss coefficient calculation.

[0040] Optionally, step S105, predicting the number of vehicles of each vehicle type in the target prediction area within the target prediction time period to obtain n vehicle counts, may include the following steps: Step S401: Obtain historical vehicle driving data and current vehicle driving data for the target prediction area; Step S402: Based on historical vehicle driving data, determine the number of first reference vehicles for each vehicle type within the target prediction time period; Step S403: Based on the current vehicle driving data, determine the number of second reference vehicles for each vehicle type within the target prediction time period; Step S404: Determine the number of n vehicles based on the number of the first reference vehicles and the number of the second reference vehicles corresponding to each vehicle type.

[0041] In one possible embodiment, historical vehicle driving data and current vehicle driving data for the target prediction area are acquired. Historical vehicle driving data refers to vehicle operation records within the target prediction area over a certain period, including data such as the average daily number of trips, peak-hour distribution density, regional dwell time, mileage distribution, and vehicle ownership growth trend for each of the n vehicle types. Sources include vehicle management department registration databases, traffic monitoring system data, data uploaded by vehicle-mounted terminals, and vehicle identification data associated with charging facilities. Current vehicle driving data is the short-term vehicle operation record most recent to the target prediction time period, including the current real-time ownership of each vehicle type, recent average daily driving frequency, regional activity, number of newly registered vehicles, and number of scrapped / transferred vehicles. This data is obtained through channels such as traffic management real-time monitoring platforms, vehicle operation management systems, and the latest registration and filing data.

[0042] In one possible embodiment, the number of first reference vehicles for each vehicle type within the target prediction period is determined based on historical vehicle driving data. The first reference vehicle number is the theoretical number of vehicles in operation or the number of active vehicles for each vehicle type within the target prediction period, derived from historical data trends, reflecting long-term operational patterns. For each vehicle type, historical monthly and yearly vehicle ownership data is extracted from historical vehicle driving data. A time series analysis method is used to fit a model of ownership growth trends, combined with historical vehicle ownership fluctuation characteristics, such as changes in the number of operational vehicles due to holidays and seasonal factors, to predict the base number of vehicles in operation within the target prediction period. For operational vehicles such as taxis and ride-hailing vehicles, the number of active vehicles in the target period is estimated by combining the historical average daily active vehicle ratio (which can be the number of active vehicles / total ownership). Finally, the first reference vehicle number for each vehicle type is obtained.

[0043] In one possible embodiment, a second reference vehicle quantity for each vehicle type is determined based on current vehicle driving data within the target prediction period. This second reference vehicle quantity is the projected quantity of each vehicle type within the target prediction period, dynamically adjusted based on recent short-term data, reflecting the immediate impact of recent market changes, policy adjustments, and environmental factors on vehicle quantities. For each vehicle type, using the real-time vehicle inventory in the current driving data as a benchmark, the recent growth rate of new vehicle additions and the attenuation rate of vehicles transferred out / scrapped are statistically analyzed to calculate the short-term net growth trend. The net growth trend is then corrected based on immediate influencing factors such as changes in infrastructure within the current region (e.g., adjustments to the layout of new bus stations and logistics parks) and fluctuations in travel demand (e.g., recent changes in business district activity and commuting demand). Finally, the vehicle quantity within the target prediction period is extrapolated to obtain the second reference vehicle quantity for each vehicle type.

[0044] In one possible implementation, the number of vehicles (n) is determined based on the number of first and second reference vehicles for each vehicle type. Specifically, for each vehicle type, the number of first and second reference vehicles is weighted and fused by combining the reliability weight of historical data and the timeliness weight of current data. The weight allocation can be dynamically adjusted according to data quality. For example, when historical data samples are sufficient and the trend is stable, the weight of the first reference vehicle number can be set to 0.6-0.7, and the weight of the second reference vehicle number to 0.3-0.4. When recent market changes are drastic and adjustments have a significant impact, the weight of the second reference vehicle number can be increased to 0.5-0.6 to highlight the impact of short-term dynamic factors. The final number of vehicles (n) is calculated using the weighted formula: Final number of vehicles = First reference vehicle number × Historical weight + Second reference vehicle number × Current weight, thus obtaining the accurate number of each vehicle type within the target prediction period.

[0045] In this embodiment, by fusing historical long-term trend data with current short-term dynamic data, a weighted fusion method is used to predict the number of each vehicle type, avoiding the limitations of prediction based on a single data dimension and improving the accuracy of vehicle number prediction.

[0046] Optionally, step S106, predicting the target charging amount for each vehicle type in the target prediction area within the target prediction time period based on the temperature correction coefficient, charging loss coefficient, and the number of n vehicles, may include the following steps: Step S501: Determine the first sub-time period and the second sub-time period based on the target prediction time period; Step S502: Based on the charging quantity prediction model, temperature correction coefficient, charging loss coefficient, and number of vehicles of each vehicle type, determine the first charging quantity of each vehicle type in the first sub-time period of the target prediction area. Step S503: Determine the charging time distribution coefficient of the target prediction area; the charging time distribution coefficient is used to reflect the charging distribution in each time period; Step S504: Based on the charging quantity prediction model, charging time distribution coefficient, temperature correction coefficient, charging loss coefficient, and number of vehicles of each vehicle type, determine the second charging quantity of each vehicle type in the target prediction area during the second sub-time period. Step S505: Determine the target charging amount for each vehicle type in the target prediction area within the target prediction time period based on the first charging amount and the second charging amount for each vehicle type.

[0047] In one possible embodiment, a first sub-time period and a second sub-time period are determined based on the target prediction time period. The first sub-time period is a coarser-grained interval within the target prediction time period that does not require further subdivision of charging distribution across time periods; it is typically a day or longer. For example, if the target prediction time period is monthly, the first sub-time period could be a complete natural day within that month. The second sub-time period is a finer-grained interval within the target prediction time period that requires precise consideration of differences in charging distribution across time periods; it is typically a specific hourly segment within a day. Specifically, based on the total duration and time accuracy requirements of the target prediction period, it is divided into segments. Long-cycle periods that do not require time-segment differentiation analysis are assigned to the first sub-period, while short-cycle periods with significant time-segment differences in charging behavior are assigned to the second sub-period. For example, if the target prediction period is from 0:00 on June 1, 2025 to 12:00 on June 3, 2025, since the charging behavior on adjacent days is relatively uniform, the period from 0:00 on June 1, 2025 to 0:00 on June 3, 2025 can be assigned to the first sub-period. Since the time-segment distribution within the day is significantly different, the period from 0:00 on June 3, 2025 to 12:00 on June 3, 2025 can be assigned to the second sub-period. After the division, the first and second sub-periods must not overlap and must completely cover the target prediction period.

[0048] In one possible embodiment, the first charging volume for each vehicle type within the first sub-time period is determined based on the charging volume prediction model for each vehicle type, the temperature correction coefficient, the charging loss coefficient, and the number of vehicles of each vehicle type. The charging volume prediction model is a customized charging volume calculation model for different vehicle types, whose parameters include the vehicle's average daily mileage, energy consumption per 100 kilometers, and charging frequency. It is a dedicated calculation model built based on historical charging data and vehicle operating characteristics; for example, the bus model focuses on fixed operating mileage and concentrated charging characteristics, while the private car model focuses on daily commuting mileage and distributed charging characteristics. The first charging volume refers to the total charging demand for a certain vehicle type within the first sub-time period, without needing to distinguish between time periods. Specifically, for each vehicle type, its corresponding dedicated charging volume prediction model is invoked. The number of vehicles of this type, the temperature correction coefficient for the target prediction time period, and the charging loss coefficient are input, along with the basic parameters of this vehicle type, such as average daily mileage and energy consumption per 100 kilometers. The model first calculates the basic charging volume of a single vehicle in the first sub-time period: Basic charging volume = average daily mileage × energy consumption per 100 kilometers / 100 × number of days in the first sub-time period. Then, the basic charging volume is multiplied by the temperature correction coefficient to calibrate the impact of temperature on energy consumption, and multiplied by (1 + charging loss coefficient) to compensate for energy loss during the charging process. Finally, it is multiplied by the number of vehicles of this type to obtain the total charging volume of this vehicle type in the first sub-time period, which is the first charging volume.

[0049] In one possible embodiment, a charging time distribution coefficient for the target prediction area is determined. This coefficient reflects the charging distribution across different time periods. The charging time distribution coefficient is the proportion of charging volume in a specific time period to the total daily charging volume; its value is a decimal between 0 and 1, and the sum of the coefficients for all time periods is 1. This coefficient quantifies the concentration of charging demand at different times of the day. Specifically, historical charging data for a certain period is acquired within the target prediction area, including charging volume records for each vehicle type at different times. The historical charging data is grouped into time periods according to a second sub-time period division standard, such as hourly time periods. The cumulative charging volume for each sub-time period is calculated against the total daily charging volume. The charging percentage for each sub-time period is calculated as: Period Percentage = Cumulative Charging Volume for that Period / Total Daily Charging Volume. The percentage is then calibrated based on seasonal characteristics and date attributes. Finally, a charging time distribution coefficient matrix adapted to the target prediction area is obtained. Each element in the matrix corresponds to the distribution coefficient for a specific time period. For example, the distribution coefficient for the morning peak (07:00-09:00) is 0.25, representing that the charging volume during this period accounts for 25% of the total daily charging volume.

[0050] In one possible embodiment, based on the charging volume prediction model for each vehicle type, the charging time distribution coefficient, the temperature correction coefficient, the charging loss coefficient, and the number of vehicles of each vehicle type, the second charging volume for each vehicle type within the second sub-time period in the target prediction area is determined. The second charging volume is the total charging demand for a certain vehicle type within the second sub-time period, which is a refined charging volume calculated based on the time period distribution coefficient. For each vehicle type, its dedicated charging volume prediction model is also invoked. First, the total daily charging volume for that type of vehicle in the second sub-time period is calculated. Then, based on the charging time distribution coefficient matrix, the distribution coefficients corresponding to each sub-time period of the second sub-time period are extracted, and the charging volume for each sub-time period within the second sub-time period is calculated: Time period charging volume = total daily charging volume × corresponding time period distribution coefficient. The charging volumes of all sub-time periods within the second sub-time period are accumulated to obtain the total charging volume for a single vehicle in the second sub-time period. Finally, multiplied by the number of vehicles of that type, the total charging demand for that vehicle type in the second sub-time period is obtained, i.e., the second charging volume.

[0051] In one possible embodiment, the target charging amount for each vehicle type within the target prediction time period is determined based on the first charging amount and the second charging amount for each vehicle type. Specifically, for each vehicle type, the first charging amount in the first sub-time period and the second charging amount in the second sub-time period are summed to calculate the target charging amount = first charging amount + second charging amount.

[0052] In this embodiment of the application, by dividing the time period and adapting it to a dedicated model, and by combining the charging time distribution coefficient to quantify the differences in time periods, the accuracy of charging quantity prediction is improved.

[0053] Optionally, step S503, determining the charging time distribution coefficient of the target prediction region, may include the following steps: Step S601: Obtain historical charging data for the target prediction area; historical charging data includes the charging amount for each time period; Step S602: Determine the charging amount ratio for each time period based on historical charging data; Step S603: Determine the charging time distribution coefficient of the target prediction area based on the charging amount ratio of each time period and the time period type of each time period.

[0054] In one possible embodiment, historical charging data for the target prediction area is acquired; the historical charging data includes the charging amount for each time period. This historical charging data includes key information such as charging start and end times, charging amount, vehicle type, and charging pile number, and its sources include public charging facility operation platform databases, private charging facility registration and usage records, and regional charging service management systems. Each time period refers to a fixed time interval divided according to time granularity. Based on the time distribution characteristics of charging behavior, it is typically divided into 24 time periods, usually in 1-hour units, such as 00:00-01:00, 01:00-02:00…23:00-24:00. It can also be adjusted to 2-hour or 30-minute units according to actual needs to distinguish the differences in charging amount across different time periods.

[0055] In one possible embodiment, the charging volume ratio for each time period is determined based on historical charging data. The charging volume ratio is the proportion of the cumulative charging volume in a given time period to the total charging volume within the historical charging data statistical period. It is used to quantify the proportion of charging demand in each time period, and its value is a decimal between 0 and 1. The sum of the charging volume ratios for all time periods is 1. Specifically, the total charging volume within the historical charging data statistical period is calculated, i.e., the sum of the cumulative charging volume in all time periods. For each time period, the cumulative charging volume for that time period is calculated. The charging volume ratio is calculated according to the formula: Charging Volume Ratio = Cumulative Charging Volume in a Time Period / Total Charging Volume in the Statistical Period, for each of the 24 time periods, or all time periods under other classification criteria.

[0056] In one possible embodiment, a charging time distribution coefficient for the target prediction area is determined based on the charging volume ratio of each time period and the time period type. Here, the time period type refers to the category of time periods classified according to charging scenario characteristics and user travel patterns, including weekday commuting hours, weekday non-commuting hours, and holiday periods. The charging demand distribution characteristics of different time period types differ significantly. The charging time distribution coefficient is used to accurately match the time period characteristics of the target prediction time period, improving the accuracy of time-segmented charging volume prediction. Specifically, each time period is labeled with a time period type. For example, 07:00-09:00 and 18:00-20:00 are labeled as weekday commuting periods, 09:00-18:00 are labeled as weekday non-commuting periods, and the entire day of holidays is labeled as holiday periods. Then, the date attributes of the target prediction time period are obtained, such as whether it is a weekday or a holiday, to determine the time period type corresponding to the target prediction time period. Based on the target time period type, the original charging volume ratio is calibrated. For example, if the target prediction time period includes weekday commuting periods, and the historical data shows that the charging volume ratio for weekday commuting periods is 0.25, but the recent addition of large enterprises in the region has led to an increase in commuting travel demand, the ratio for this time period can be calibrated to 0.28 based on the new travel data. If the target prediction time period is a holiday, the original ratio is adjusted by referring to the charging volume ratio characteristics of historical holiday periods to remove the influence of the ratio deviation for weekday commuting periods. After calibration, the charging time distribution coefficient corresponding to each time period is obtained.

[0057] In this embodiment of the application, by combining historical charging volume statistics with precise calibration of time period type, the charging time distribution coefficient is determined, thereby improving the matching degree between the coefficient and the target prediction scenario, making the time-based charging volume prediction more accurate.

[0058] Optionally, step S504, determining the second charging amount for each vehicle type in the target prediction area within the second sub-time period based on the charging amount prediction model for each vehicle type, the charging time distribution coefficient, the temperature correction coefficient, the charging loss coefficient, and the number of vehicles of each vehicle type, may include the following steps: Step S701: Determine the charging amount ratio of the second sub-time period based on the charging time distribution coefficient and the time period type of the second sub-time period; Step S702: Determine the candidate time period corresponding to the second sub-time period; Step S703: Based on the charging quantity prediction model, temperature correction coefficient, charging loss coefficient, and number of vehicles of each vehicle type, determine the candidate charging quantity of each vehicle type in the target prediction area within the candidate time period. Step S704: Based on the charging amount ratio and candidate charging amount, determine the second charging amount for each vehicle type in the target prediction area during the second sub-time period.

[0059] In one possible embodiment, the charging volume ratio of the second sub-time period is determined based on the charging time distribution coefficient and the time period type of the second sub-time period. The charging volume ratio of the second sub-time period is the proportion of the charging volume of the second sub-time period within its corresponding complete cycle. The specific time range and corresponding time period type of the second sub-time period are determined, such as 07:00-09:00, corresponding to weekday commuting hours. From the determined charging time distribution coefficient matrix, the distribution coefficients of each sub-time period under this time period type are extracted. These sub-time period distribution coefficients are summed to obtain the total proportion of the second sub-time period within the complete cycle, i.e., the charging volume ratio of the second sub-time period. For example, if the distribution coefficient for 07:00-08:00 is 0.13 and the distribution coefficient for 08:00-09:00 is 0.15, then the charging volume ratio of the second sub-time period is 0.28.

[0060] In one possible embodiment, a candidate time period corresponding to the second sub-time period is determined. The candidate time period refers to a complete time interval used for calculating the reference charging capacity that has the same time cycle characteristics as the second sub-time period. It is typically the calendar day to which the second sub-time period belongs. For example, if the second sub-time period is from 07:00 to 09:00 on a certain workday, then the candidate time period is that complete workday. Specifically, based on the start and end times of the second sub-time period, its complete time cycle is determined. If the second sub-time period is a daytime segment, then the candidate time period is the calendar day to which that segment belongs; if the second sub-time period is a specific segment within a week, then the candidate time period is the complete calendar week to which that segment belongs. This ensures that the candidate time periods can completely cover the second sub-time period and possess the same date attributes, seasonal characteristics, etc., as the second sub-time period.

[0061] In one possible embodiment, based on the charging quantity prediction model for each vehicle type, the temperature correction coefficient, the charging loss coefficient, and the number of vehicles of each vehicle type, the candidate charging quantity for each vehicle type within the candidate time period in the target prediction area is determined. The candidate charging quantity is the total charging demand for a certain vehicle type within the candidate time period, and its calculation logic is consistent with that of the first charging quantity. Specifically, for each vehicle type, its dedicated charging quantity prediction model is invoked, and the number of vehicles of that type, the temperature correction coefficient corresponding to the target prediction time period, and the charging loss coefficient are input, along with the basic parameters of that vehicle type, such as average daily mileage, energy consumption per 100 kilometers, and charging frequency. The model first calculates the basic charging quantity for a single vehicle within the candidate time period; then, the basic charging quantity is multiplied by the temperature correction coefficient to calibrate the impact of ambient temperature on vehicle energy consumption, and multiplied by (1 + charging loss coefficient) to compensate for energy loss during the charging process; finally, the corrected charging quantity for a single vehicle is multiplied by the number of vehicles of that type to obtain the total charging demand for that vehicle type within the candidate time period, i.e., the candidate charging quantity.

[0062] In one possible embodiment, the second charging amount for each vehicle type in the target prediction area within the second sub-time period is determined based on the charging amount ratio and candidate charging amounts. For each vehicle type, the charging amount ratio of the second sub-time period is multiplied by the candidate charging amount to calculate the second charging amount: Second charging amount = Candidate charging amount × Charging amount ratio of the second sub-time period. After calculation, the rationality can be verified by combining the charging behavior characteristics of the vehicle type. For example, the second charging amount of commercial vehicles during peak commuting hours should be consistent with the historical peak charging trend, and the second charging amount of private cars should match commuting travel demand. If there is a deviation in the verification, the charging amount ratio can be fine-tuned by combining the charging preferences of the vehicle type during different time periods, such as buses charging more during off-peak hours in the daytime, so that the second charging amount can accurately reflect the actual charging demand of the vehicle type within the second sub-time period. Finally, the refined charging demand of each vehicle type within the second sub-time period, i.e., the second charging amount, is obtained.

[0063] In this embodiment of the application, the charging amount of the second sub-time period is determined by accurately combining the calculation of the charging amount of the baseline cycle with the time period ratio, thereby improving the accuracy of the target charging amount prediction.

[0064] In a specific embodiment, see Figure 5 , Figure 5 This is a schematic diagram of multidimensional data provided in an embodiment of this application. The data acquisition module obtains vehicle classification data, operation data, energy consumption data, environmental data, and charging facility data. Among them, vehicle classification data includes types such as buses, taxis, ride-hailing vehicles, logistics vehicles, private cars, and official vehicles, clearly defining the vehicle coverage of the data collection; operation data refers to vehicle dynamic information, specifically including average daily mileage data; energy consumption data can be obtained for different vehicle brands, such as brand one, brand two, brand three, etc.; environmental data refers to external influencing factors, including average temperature and precipitation; charging facility data refers to charging resources, including information such as charging pile type, charging pile distribution, and charging efficiency.

[0065] Furthermore, customized charging demand prediction models will be built to address the differentiated operational characteristics of buses, taxis, ride-hailing vehicles, logistics vehicles, private cars, and official vehicles.

[0066] The bus model is built upon the characteristics of fixed routes, stable daily mileage, and well-defined vehicle parameters, employing an energy consumption statistical model based on operational characteristics. Using mileage data from fixed operating routes, daily operating frequency, and vehicle rated energy consumption parameters as core inputs, the model quantifies energy consumption patterns under fixed routes by statistically analyzing the correlation between energy consumption and charging demand over historical operating cycles, thus constructing a charging demand prediction model that accurately matches the bus operation mode.

[0067] The taxi model is built based on vehicle operation data and location trajectories, employing a statistical model based on the number of active vehicles and average daily energy consumption. Dynamic information such as actual mileage and operating time of taxis is extracted from location trajectory data. Combined with statistics on the number of active vehicles and average daily energy consumption data, a quantitative correlation between operating status and charging demand is established, enabling dynamic prediction of taxi charging needs.

[0068] The ride-hailing model focuses on distinguishing between high-frequency and low-frequency operating states, employing a classification and statistical model based on order data and mileage. By breaking down order data into mileage, order frequency, and charging interval characteristics under different operating states, a mapping relationship between energy consumption and charging demand is established for each of the two states. This mapping is then combined with classification weight coefficients to form a complete predictive model.

[0069] The logistics vehicle model is constructed to fully consider the impact of monthly freight volume fluctuations, and adopts an energy consumption model based on average daily mileage and monthly correction coefficients. Based on historical average daily mileage and vehicle load energy consumption characteristics, monthly correction coefficients are generated by analyzing the monthly freight volume variation patterns to calibrate energy consumption deviations under different freight loads and improve the model's adaptability to freight fluctuations.

[0070] The private car model integrates data on car ownership, vehicle type composition, and user surveys, employing an energy consumption model based on sampling statistics and annual mileage estimation. By sampling and statistically analyzing the annual mileage and charging habits of different car types, combined with regional private car ownership and vehicle type proportions, the overall energy consumption level and charging demand distribution are estimated, enabling accurate prediction of decentralized charging demand.

[0071] The model for official vehicles is constructed based on their relatively fixed daily mileage usage characteristics, employing a statistical model based on vehicle ownership and daily energy consumption. Using data on the registered ownership of official vehicles as a foundation, combined with historical usage records of daily mileage and rated energy consumption parameters, the correlation between energy consumption and charging demand under fixed usage patterns is quantified, leading to the construction of a concise and efficient predictive model.

[0072] Furthermore, the calculation of bus charging volume is as follows: Total power consumption = Average daily power consumption × Number of predicted days × Number of buses × Temperature correction factor × (1 + Charging loss factor). Where, Average daily power consumption = Average daily operating mileage × Energy consumption per 100 kilometers / 100; Number of buses refers to the number of buses in the designated area; Temperature correction factor is 1 at normal temperature, and calculated based on temperature difference at extreme temperatures; Charging loss factor takes into account the losses of charging equipment and power transmission, and is taken as 15%.

[0073] Taxi charging calculation: Total power consumption = Average daily power consumption × Number of taxis × Number of forecast days × Temperature correction factor × (1 + Charging loss factor). Average daily power consumption = Average daily operating mileage × Energy consumption per 100 kilometers / 100; The number of taxis is the average number of active taxis in the forecast area, and the other parameters are consistent with the bus model.

[0074] Ride-hailing vehicle charging calculation: Total power consumption = (Daily average power consumption of high-frequency ride-hailing vehicles × Number of high-frequency ride-hailing vehicles + Daily average power consumption of low-frequency ride-hailing vehicles × Number of low-frequency ride-hailing vehicles) × Number of forecast days × Temperature correction coefficient × (1 + Charging loss coefficient). The number of high-frequency and low-frequency ride-hailing vehicles in the forecast area needs to be counted separately, and the daily average power consumption calculation logic is the same as before.

[0075] Calculation of charging capacity for logistics vehicles: Total power consumption = Average daily power consumption × × Number of logistics vehicles × Forecast number of days × Temperature correction factor × (1 + Charging loss factor). This is a monthly correction factor used to adjust for the impact of freight volume fluctuations on energy consumption.

[0076] Private vehicle charging consumption calculation: Includes sedans and sport utility vehicles (SUVs). Total power consumption = (average daily power consumption of sedans × number of sedans + average daily power consumption of SUVs × number of SUVs) × number of forecast days × temperature correction factor × (1 + charging loss factor). Calculated separately based on vehicle type, sedans consume approximately 13.5 kWh per 100 km, and SUVs approximately 16.5 kWh. Parameters such as average annual mileage can be adjusted based on regional survey data.

[0077] Calculation of charging capacity for official vehicles: Total power consumption = Average daily power consumption × Number of official vehicles × Number of forecast days × Temperature correction factor × (1 + Charging loss factor). Average daily power consumption is calculated based on the average daily mileage and energy consumption per 100 kilometers of official vehicles. The number of vehicles is the number of official vehicles in the forecast area.

[0078] In a specific embodiment, see Figure 6 , Figure 6This is a schematic diagram illustrating the visualization of results provided in an embodiment of this application. After predicting the target charging volume, the prediction results are presented to the user in an intuitive and easy-to-understand form. The main outputs include: total charging demand, displaying the total charging demand within the predicted time range in numerical and graphical form and comparing it with historical data from the same period; charging demand by vehicle type, showing the proportion and absolute amount of charging demand for different types of vehicles, presented in pie charts and bar charts; time distribution map, showing the daily, weekly, and monthly distribution characteristics of charging demand to help identify peak charging periods; geographical distribution heat map, visually displaying the charging density in different areas in heat map form, providing a reference for charging facility layout; peak-valley analysis map, analyzing the peak-valley characteristics of charging load, calculating the peak-valley ratio, assessing grid pressure, and providing a basis for off-peak charging strategies. Furthermore, it supports exporting data in multiple formats such as Excel spreadsheets, PDF reports, and images, and also provides an Application Programming Interface (API) interface, allowing other systems to call the prediction results, achieving seamless integration with charging facility management systems, grid dispatching systems, etc.

[0079] In this embodiment, the target prediction area, target prediction time period, and n vehicle types input by the user are first obtained. Historical vehicle energy consumption data, historical ambient temperature data, and charging pile data for the target prediction area within the historical time period are also obtained. Then, based on the historical vehicle energy consumption data and historical ambient temperature data, a temperature correction coefficient for the target prediction area within the target prediction time period is calculated. Next, the number of vehicles of each vehicle type in the target prediction area within the target prediction time period is predicted, resulting in n vehicle numbers. Finally, based on the temperature correction coefficient, charging loss coefficient, and the n vehicle numbers, the target charging amount for each vehicle type in the target prediction area within the target prediction time period is predicted. Therefore, by calculating the corresponding temperature correction coefficient and charging loss coefficient, and classifying and predicting vehicles within the target prediction area based on these coefficients, unlike traditional single-parameter predictions, the accuracy of charging amount prediction can be improved when predicting vehicle charging demand.

[0080] The methods of the embodiments of the present invention have been described in detail above, and the apparatus of the embodiments of the present invention is provided below.

[0081] See Figure 7 , Figure 7 This is a schematic diagram of the structure of a charging quantity prediction device provided in an embodiment of this application. Figure 7As shown, the charging quantity prediction device 800 includes an acquisition unit 801 and a processing unit 802. The acquisition unit 801 is used to acquire the target prediction area, target prediction time period, and n vehicle types input by the user; acquire historical vehicle energy consumption data, historical ambient temperature data, and charging pile data of the target prediction area within the historical time period. The processing unit 802 is used to calculate the temperature correction coefficient of the target prediction area within the target prediction time period based on the historical vehicle energy consumption data and historical ambient temperature data; calculate the charging loss coefficient of the target prediction area within the target prediction time period based on the historical charging pile data; predict the number of vehicles of each vehicle type in the target prediction area within the target prediction time period, obtaining n vehicle numbers; and predict the target charging quantity of each vehicle type in the target prediction area within the target prediction time period based on the temperature correction coefficient, the charging loss coefficient, and the n vehicle numbers.

[0082] In specific implementations, the acquisition unit 801 and processing unit 802 in this application embodiment can also execute other implementation methods described in the charging quantity prediction method of this application embodiment, which will not be repeated here.

[0083] See Figure 8 , Figure 8 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Figure 8 As shown, the electronic device 900 includes a transceiver 901, a processor 902, and a memory 903, which are connected via a bus 904. The memory 903 stores computer programs and data, and can transmit the data stored in the memory 903 to the processor 902. The electronic device 900 can be the aforementioned charge level prediction device 800, and the processor 902 can be the aforementioned acquisition unit 801 or processing unit 802. In this embodiment, the processor 902 is used to read the computer program in the memory 903 and execute some or all of the steps of the aforementioned charge level prediction method.

[0084] This application also provides a computer-readable storage medium storing a computer program that is executed by a processor to implement some or all of the steps of any of the charge prediction methods described in the above method embodiments.

[0085] This application also provides a computer program product, which includes a non-transitory computer-readable storage medium storing a computer program operable to cause a computer to perform some or all of the steps of any of the charge prediction methods described in the above method embodiments.

[0086] It should be noted that, for the sake of simplicity, the foregoing method embodiments are all described as a series of actions. However, those skilled in the art should understand that this application is not limited to the described order of actions, as some steps may be performed in other orders or simultaneously according to this application. Furthermore, those skilled in the art should also understand that the embodiments described in the specification are all optional embodiments, and the actions and modules involved are not necessarily essential to this application.

[0087] In the above embodiments, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions in other embodiments.

[0088] In the several embodiments provided in this application, it should be understood that the disclosed apparatus can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of modules is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple modules or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between devices or modules may be electrical or other forms.

[0089] The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical modules; that is, they may be located in one place or distributed across multiple network modules. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0090] Furthermore, the functional modules in the various embodiments of this application can be integrated into one processing module, or each module can exist physically separately, or two or more modules can be integrated into one module. The integrated modules described above can be implemented in hardware or as software program modules.

[0091] If the integrated module is implemented as a software program module and sold or used as an independent product, it can be stored in a computer-readable storage device (CMD). Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a memory and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned memory includes various media capable of storing program code, such as USB flash drives, read-only memory (ROM), random access memory (RAM), portable hard drives, magnetic disks, or optical disks.

[0092] The embodiments of this application have been described in detail above. Specific examples have been used to illustrate the principles and implementation methods of this application. The description of the above embodiments is only for the purpose of helping to understand the method and core ideas of this application. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of this application. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. A method for predicting charging capacity, characterized in that, include: Obtain the target prediction area, target prediction time period, and n vehicle types input by the user; Acquire historical vehicle energy consumption data, historical ambient temperature data, and charging pile data for the target prediction area within a historical time period; Based on the historical vehicle energy consumption data and the historical ambient temperature data, calculate the temperature correction coefficient for the target prediction area during the target prediction time period; Based on the historical charging pile data, calculate the charging loss coefficient of the target prediction area during the target prediction time period; Predict the number of vehicles of each vehicle type in the target prediction area within the target prediction time period to obtain n vehicle counts; Based on the temperature correction coefficient, the charging loss coefficient, and the number of n vehicles, the target charging amount for each vehicle type in the target prediction area during the target prediction time period is predicted.

2. The method as described in claim 1, characterized in that, The step of calculating the temperature correction coefficient for the target prediction area during the target prediction time period based on the historical vehicle energy consumption data and the historical ambient temperature data includes: Based on the historical vehicle energy consumption data and the historical ambient temperature data, a first mapping relationship between the actual change in vehicle energy consumption and the change in ambient temperature is determined. Obtain the predicted temperature of the target prediction area during the target prediction time period; Calculate the temperature difference between the predicted temperature and the preset reference temperature; The temperature correction coefficient is determined based on the temperature difference and the first mapping relationship.

3. The method as described in claim 1, characterized in that, The step of calculating the charging loss coefficient of the target prediction area during the target prediction time period based on the historical charging pile data includes: Based on the historical charging pile data, determine the number of charging piles corresponding to each type of charging pile within the target prediction area; Obtain the charging efficiency of each charging station; Based on the charging efficiency of each charging pile, determine the charging efficiency of each type of charging pile; The usage ratio of each charging pile type is determined based on the number of charging piles corresponding to each type. The charging loss coefficient is determined based on the usage ratio of each type of charging pile and the charging efficiency of each type of charging pile.

4. The method as described in claim 1, characterized in that, The prediction of the number of vehicles of each vehicle type in the target prediction area within the target prediction time period yields n vehicle counts, including: Obtain historical vehicle driving data and current vehicle driving data for the target prediction area; Based on the historical vehicle driving data, determine the number of first reference vehicles of each vehicle type within the target prediction time period; Based on the current vehicle driving data, determine the number of second reference vehicles of each vehicle type within the target prediction time period; The number of n vehicles is determined based on the number of the first reference vehicles and the number of the second reference vehicles corresponding to each vehicle type.

5. The method as described in claim 4, characterized in that, The step of predicting the target charging amount for each vehicle type in the target prediction area within the target prediction time period based on the temperature correction coefficient, the charging loss coefficient, and the number of n vehicles includes: Based on the target predicted time period, determine the first sub-time period and the second sub-time period; Based on the charging volume prediction model for each vehicle type, the temperature correction coefficient, the charging loss coefficient, and the number of vehicles of each vehicle type, the first charging volume of each vehicle type in the target prediction area within the first sub-time period is determined. Determine the charging time distribution coefficient of the target prediction area; the charging time distribution coefficient is used to reflect the charging distribution in each time period; Based on the charging volume prediction model for each vehicle type, the charging time distribution coefficient, the temperature correction coefficient, the charging loss coefficient, and the number of vehicles of each vehicle type, the second charging volume of each vehicle type in the target prediction area during the second sub-time period is determined. Based on the first charging amount and the second charging amount for each vehicle type, the target charging amount for each vehicle type in the target prediction area is determined within the target prediction time period.

6. The method as described in claim 5, characterized in that, Determining the charging time distribution coefficient of the target prediction region includes: Obtain historical charging data for the target prediction area; the historical charging data includes the charging amount for each time period; Based on the historical charging data, determine the charging volume ratio for each time period; The charging time distribution coefficient of the target prediction area is determined based on the charging volume ratio and the time period type of each time period.

7. The method as described in claim 6, characterized in that, The step of determining the second charging amount for each vehicle type in the target prediction area within the second sub-time period based on the charging amount prediction model for each vehicle type, the charging time distribution coefficient, the temperature correction coefficient, the charging loss coefficient, and the number of vehicles of each vehicle type includes: Based on the charging time distribution coefficient and the time period type of the second sub-time period, the charging amount ratio of the second sub-time period is determined; Determine the candidate time period corresponding to the second sub-time period; Based on the charging volume prediction model for each vehicle type, the temperature correction coefficient, the charging loss coefficient, and the number of vehicles of each vehicle type, the candidate charging volume for each vehicle type in the target prediction area within the candidate time period is determined. Based on the charging amount ratio and the candidate charging amount, the second charging amount for each vehicle type in the target prediction area during the second sub-time period is determined.

8. A charging quantity prediction device, characterized in that, The device includes an acquisition unit and a processing unit; The acquisition unit is used to acquire the target prediction area, target prediction time period, and n types of vehicles input by the user. Acquire historical vehicle energy consumption data, historical ambient temperature data, and charging pile data for the target prediction area within a historical time period; The processing unit is used to calculate the temperature correction coefficient of the target prediction area during the target prediction time period based on the historical vehicle energy consumption data and the historical ambient temperature data. Based on the historical charging pile data, calculate the charging loss coefficient of the target prediction area during the target prediction time period; Predict the number of vehicles of each vehicle type in the target prediction area within the target prediction time period to obtain n vehicle counts; Based on the temperature correction coefficient, the charging loss coefficient, and the number of n vehicles, the target charging amount for each vehicle type in the target prediction area during the target prediction time period is predicted.

9. An electronic device, characterized in that, include: A processor and a memory, the processor being connected to the memory, the memory being used to store a computer program, the processor being used to execute the computer program stored in the memory to cause the electronic device to perform the method as described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, the computer program including program instructions that, when executed by a processor, cause the processor to perform the method as described in any one of claims 1-7.

Images