Charging scheduling method, device and equipment

Abstract

The invention provides a charging scheduling method, device and equipment. The method provided by the invention comprises the following steps: when a preset monitoring period arrives, if the electricity price is a non-valley price, predicting a weighted electricity conversion demand quantity of a next time period; if the total number of batteries held by the to-be-dispatched battery swap station at the current moment exceeds the weighted battery swap demand quantity and the first number of ready batteries of the to-be-dispatched battery swap station at the current moment does not exceed the weighted battery swap demand quantity, the charging time consumption of the batteries which are not ready at present is predicted; predicting whether the battery is ready in the next time period according to the charging time consumption; and when the first battery number and the number of the ready batteries in the next time period are greater than or equal to the weighted battery replacement demand, controlling the battery replacement station to be dispatched to charge the ready batteries in the next time period. According to the charging scheduling method, device and equipment provided by the invention, the purposes of peak load shifting and power utilization cost reduction can be realized on the premise of meeting the power conversion requirement.

Description

Technical Field

[0001] The present application relates to the technical field of charging management of battery swap stations, and in particular to a charging scheduling method, device and equipment. Background Art

[0002] In the battery swap business scenario, the charging strategy of the battery swap station is generally plug-and-charge, that is, once a battery is replaced and enters the charging compartment, the rack load unit will immediately start charging the battery until the battery is fully charged or the power reaches a specified threshold. In actual use, it is often encountered that the battery is placed in the charging compartment and fully charged during the peak electricity price period, and then is not taken out for a long time. When the electricity price is low, all the batteries in the battery swap station are fully charged, resulting in the current charging strategy not only increasing the fluctuation of the grid load, but also causing economic losses. Summary of the invention

[0003] In view of this, the present application provides a charging scheduling method, device and equipment to achieve the purpose of peak shaving and valley filling and reduce electricity costs while meeting the demand for battery replacement.

[0004] Specifically, the present application is implemented through the following technical solutions:

[0005] A first aspect of the present application provides a charging scheduling method, the method comprising:

[0006] For the battery swap station to be scheduled, when the preset monitoring period arrives, if the current electricity price is not a valley price, multiple battery swap demand prediction models are used to predict the battery swap demand in the next period; wherein each of the multiple battery swap demand prediction models is different, and the multiple battery swap demand prediction models include at least one model for predicting the battery swap demand based on the real-time power consumption of the vehicle, and a model for predicting the battery swap demand based on the historical battery swap data of the battery swap station;

[0007] The battery swap demand predicted by each battery swap demand prediction model is weighted according to the weight of each battery swap demand prediction model at the current moment to obtain the weighted battery swap demand for the next period; wherein the weight of each battery swap demand prediction model is dynamically updated based on the prediction error of the battery swap demand prediction model;

[0008] If the total number of batteries held by the battery swap station to be scheduled at the current moment exceeds the weighted battery swap demand, and the number of first batteries that are ready at the battery swap station to be scheduled at the current moment does not exceed the weighted battery swap demand, then for each battery that is currently not ready in the battery swap station to be scheduled, predict the charging time required for the battery to change from not ready to ready, and predict whether the battery will be ready in the next time period based on the charging time;

[0009] When the sum of the first number of batteries and the number of batteries that are ready for the next time period is greater than or equal to the weighted battery swapping demand, the battery swapping station to be scheduled is controlled to charge the batteries that are ready for the next time period.

[0010] A second aspect of the present application provides a charging scheduling device, the device comprising a prediction module, an update module and a control module; wherein:

[0011] The prediction module is used for predicting the battery swap demand in the next period of time using multiple battery swap demand prediction models for the battery swap station to be scheduled when the preset monitoring period arrives, if the electricity price at the current moment is not a valley price; wherein each of the multiple battery swap demand prediction models is different, and the multiple battery swap demand prediction models include at least one model for predicting the battery swap demand based on the real-time power consumption of the vehicle, and a model for predicting the battery swap demand based on the historical battery swap data of the battery swap station;

[0012] The updating module is used to perform weighted processing on the battery swap demand predicted by each battery swap demand prediction model according to the weight of each battery swap demand prediction model at the current moment, so as to obtain the weighted battery swap demand for the next period; wherein the weight of each battery swap demand prediction model is dynamically updated based on the prediction error of the battery swap demand prediction model;

[0013] The prediction module is further used to, when it is judged that the total number of batteries held by the battery swap station to be scheduled at the current moment exceeds the weighted battery swap demand, and the number of first batteries that are ready at the battery swap station to be scheduled at the current moment does not exceed the weighted battery swap demand, predict the charging time required for each battery that is currently not ready in the battery swap station to be scheduled to change from not ready to ready, and predict whether the battery will be ready in the next time period based on the charging time;

[0014] The control module is used to control the battery swap station to be scheduled to charge the batteries that are ready for the next time period when the sum of the first number of batteries and the number of batteries that are ready for the next time period is greater than or equal to the weighted battery swap demand.

[0015] The third aspect of the present application provides a charging scheduling device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the steps of any one of the methods provided in the first aspect of the present application are implemented.

[0016] The charging scheduling method, device and equipment provided in the present application, firstly, during the peak electricity price period, while ensuring that the battery swap service is not affected, priority is given to reducing the pressure on the power grid, and the charging rhythm is controlled by accurately predicting the weighted battery swap demand in the next period. When the number of batteries currently held exceeds the predicted weighted battery swap demand but the ready batteries are insufficient, by predicting the readiness of the batteries in the next period, only those batteries that can be charged in the next period are charged, which can not only avoid excessive charging operations during peak hours and save electricity resources, but also ensure that there are enough batteries available during peak demand, which can improve the operating efficiency of the battery swap station; secondly, during non-peak hours, the low valley electricity price can be fully utilized to charge the battery to ensure that there is enough battery supply during future peak hours. In this way, the fluctuation of electricity prices can be used to achieve more efficient use of energy and reduce the overall electricity cost; finally, when predicting the demand for battery swapping, the demand for battery swapping is predicted by multiple battery swapping demand prediction models, and the weight of each model is adjusted by dynamically updating the prediction error, which helps to more comprehensively understand the driving factors of the battery swapping demand and accurately predict the weighted battery swapping demand.

[0017] Furthermore, the charging scheduling method, apparatus and equipment provided in the present application selects a specific number of target batteries from the batteries that are ready in the next time period in order of remaining power from low to high when the sum of the first number of batteries and the number of batteries that can be ready in the next time period is greater than the weighted battery replacement demand, and then controls the battery replacement station to be scheduled to charge the target batteries. In this way, when the actual battery replacement demand in the next time period is greater than the weighted battery replacement demand, the remaining batteries can be charged immediately to reduce waiting time.

[0018] Furthermore, the charging scheduling method, apparatus and equipment provided in the present application, by fitting the first function with the battery's own historical charging record, can accurately reflect the individual charging behavior of the battery through the first function. Furthermore, by fitting the second function with the overall charging record of the battery swap station to be scheduled before the current moment, the second function can capture factors such as the usage of charging resources in the battery swap station and the overall performance of the charging equipment, and reflect the charging performance of the battery swap station. In this way, when the first function and the second function are combined to estimate the charging time, the individual charging performance of the battery and the overall charging performance of the battery swap station to be scheduled can be combined, and the differences between individual batteries and the overall charging environment can be accurately analyzed to improve the accuracy of the charging time prediction and avoid errors caused by individual differences and environmental factors. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 A flowchart of a first embodiment of a charging scheduling method provided in this application;

[0020] Figure 2A flowchart of a second embodiment of the charging scheduling method provided in this application;

[0021] Figure 3 A flowchart of the third embodiment of the charging scheduling method provided in this application;

[0022] Figure 4 A flowchart of a fourth embodiment of the charging scheduling method provided in this application;

[0023] Figure 5 A hardware structure diagram of the charging scheduling device where the charging scheduling device provided in this application is located;

[0024] Figure 6 This is a structural diagram of the first embodiment of the charging scheduling device provided in this application. DETAILED DESCRIPTION

[0025] Here, exemplary embodiments will be described in detail, and examples thereof are shown in the accompanying drawings. When the following description refers to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The implementations described in the following exemplary embodiments do not represent all implementations consistent with the present application.

[0026] The terms used in this application are only for the purpose of describing specific embodiments and are not intended to limit this application. The singular forms of "a", "said" and "the" used in this application are also intended to include plural forms, unless the context clearly indicates other meanings. It should also be understood that the term "and / or" used in this article refers to and includes any or all possible combinations of one or more associated listed items.

[0027] It should be understood that although the terms first, second, third, etc. may be used in the present application to describe various information, these information should not be limited to these terms. These terms are only used to distinguish the same type of information from each other. For example, without departing from the scope of the present application, the first information may also be referred to as the second information, and similarly, the second information may also be referred to as the first information. Depending on the context, the word "if" as used herein may be interpreted as "at the time of" or "when" or "in response to determining".

[0028] Specific embodiments are given below to introduce the technical solution of the present application in detail.

[0029] Figure 1 This is a flow chart of the first embodiment of the charging scheduling method provided by this application. Please refer to Figure 1 , the method provided in this embodiment may include:

[0030] S101. For the battery swap station to be scheduled, when the preset monitoring period arrives, if the current electricity price is not a valley price, multiple battery swap demand prediction models are used to predict the battery swap demand in the next time period; wherein, each of the multiple battery swap demand prediction models is different, and the multiple battery swap demand prediction models include at least one model for predicting the battery swap demand based on the real-time power consumption of the vehicle, and a model for predicting the battery swap demand based on the historical battery swap data of the battery swap station.

[0031] It should be noted that the charging scheduling method and device provided in this application are applied to the charging scheduling device. The charging scheduling device can be a server. Furthermore, the charging scheduling method, device and equipment provided in this application are used to control the battery swap station under the control of the charging scheduling device, specifically, to control the charging behavior of the battery at the battery swap station, in other words, to control which batteries are charged at which time at the battery swap station.

[0032] In a specific implementation, the charging scheduling device will preset a monitoring period to monitor whether the battery needs to be charged and which batteries need to be charged when the preset monitoring period arrives.

[0033] Referring to the previous description, it can be understood that the battery swap station to be scheduled is any battery swap station under the control of the charging scheduling device. Furthermore, the specific value of the preset monitoring period is set according to actual needs, and is not limited in this embodiment. For example, in one possible implementation, the preset monitoring period is 10 minutes, that is, monitoring is performed every 10 minutes.

[0034] It should be noted that in order to adapt to the power demand and supply in different time periods, power companies will set different electricity prices in different time periods according to the load of the power grid, that is, the electricity prices are divided into multiple categories (different time periods correspond to different categories). Among them, valley price refers to a category in electricity price, which is usually used to describe the lowest point of electricity price in a day. Non-valley price refers to other categories in the electricity price category except valley price.

[0035] Specifically, when classifying electricity prices, different power companies and regions may adopt different segmentation strategies. For example, in one possible implementation, electricity prices are divided into three categories, including peak prices, valley prices, and par prices. At this time, non-valley prices refer to peak prices and par prices. For another example, in another possible implementation, the electricity usage time of a day is divided into four different time periods, and each time period corresponds to a different electricity price, that is, the electricity price is divided into four categories, including peak time periods (time periods with the highest and most intense electricity demand, and electricity prices are usually the most expensive in a day, for example, 1 pm to 3 pm or 6 pm to 8 pm), peak time periods (regular peak time periods, electricity prices are higher than normal time periods, for example, 7 am to 9 am, 5 pm to 7 pm), normal time periods (time periods between peak time periods and valley time periods, electricity prices are relatively stable, for example, 9 am to 12 pm, 3 pm to 5 pm), valley time periods (time periods with lower electricity demand, and lower electricity prices, for example, 8 pm to 7 am), at this time, non-valley prices include peak time periods, peak time periods, and parity time periods.

[0036] In combination with the above introduction, for example, in one embodiment, the power company of city A stipulates that the time period from 8:00 to 22:00 is the peak electricity consumption period, and the electricity price during this period is the peak price; the time period from 22:00 to 8:00 is the low electricity consumption period, and the electricity price during this period is the valley price; for another example, in another embodiment, the power company of city B stipulates that the time periods from 10:00 to 15:00 and 18:00 to 21:00 are the peak electricity consumption periods, and the electricity price during this period is the peak price; the time periods from 7:00 to 10:00, 15:00 to 18:00, and 21:00 to 23:00 are the flat electricity price time periods, and the electricity price during this time period is the flat price; the time period from 23:00 to 7:00 is the low electricity consumption period, and the electricity price during this time period is the valley price.

[0037] Specifically, when the preset monitoring period arrives, if the current electricity price is not a valley price, charging scheduling is performed according to the method provided in this embodiment. Furthermore, if the current electricity price is a valley price, at this time, there is no need to consider the cost issue, and the battery swap station to be scheduled is directly controlled to charge all batteries that are not ready in the battery swap station.

[0038] Specifically, in this embodiment, when the forecast monitoring period arrives, if the current electricity price is not a valley price, multiple battery swap demand forecasting models are used to predict the battery swap demand in the next period. It should be noted that the battery swap demand in the next period is used to characterize how many vehicles will need to replace batteries in the next period, that is, how many ready batteries will need to be replaced in the next period.

[0039] It should be noted that the next time period refers to the preset time period after the current moment. The specific duration of the preset time period is set according to actual needs and is not limited in this embodiment. For example, in one possible implementation, the next time period is 30 minutes after the current moment. Combined with the previous example, when the preset monitoring period is 10 minutes, at this time, the charging scheduling equipment monitors every 10 minutes. Furthermore, during monitoring, if the current electricity price is not a valley price, the battery swap demand prediction model is used to predict the battery swap demand within 30 minutes after the current moment.

[0040] Furthermore, the battery swap demand prediction model can predict the battery swap demand in the next period based on the input. For example, in one possible implementation, the battery swap demand prediction model includes two models, one of which predicts that the battery swap demand in the next period is 5, and the other predicts that the battery swap demand in the next period is 7.

[0041] It should be noted that each of the multiple battery swap demand prediction models is different. Further, the multiple battery swap demand prediction models include at least one model for predicting the battery swap demand based on the real-time power of the vehicle, and a model for predicting the battery swap demand based on the historical battery swap data of the battery swap station.

[0042] Specifically, the model for predicting the demand for battery replacement based on the real-time power of the vehicle can predict the probability of the vehicle replacing the battery in the next period based on the real-time power of the vehicle, and then determine the demand for battery replacement in the next period based on the probability of the vehicle replacing the battery in the next period. The input of the model is the power of a vehicle at the current moment, and the output is the probability of the vehicle replacing the battery in the next period.

[0043] It should be noted that the battery swap demand prediction model based on the real-time power of the vehicle predicts the demand for battery swapping is obtained by fitting based on the vehicle's status data, wherein the vehicle's status data includes the vehicle's real-time power and the vehicle's battery swapping behavior data; further, the vehicle's battery swapping behavior data may include the vehicle's identification information, battery power information, the start time and end time of the battery swapping operation, the specific date of the battery swapping operation, etc. The vehicle's real-time power and battery swapping behavior data can be used to fit the linear relationship between the vehicle's current power and the probability of battery swapping in the future, thereby realizing the ability to predict future battery swapping behavior based on real-time power. Furthermore, please refer to the description of the relevant prior art for the specific training method of the model, which will not be repeated here.

[0044] Furthermore, the model for predicting the demand for battery swapping based on the historical battery swapping data of the battery swapping station can predict the demand for battery swapping in the next period based on the historical battery swapping data of the battery swapping station to be scheduled. The input of the model is the historical battery swapping data of the battery swapping station to be scheduled, and the output is the demand for battery swapping in the next period.

[0045] Furthermore, the battery swap demand prediction model for predicting the battery swap demand based on the historical battery swap data of the battery swap station is pre-trained based on the historical battery swap data of the battery swap station, wherein the historical battery swap data of the battery swap station may include the number of batteries swapped out of the battery swap station at different times. Further, for the specific training method of the model, please refer to the description of the relevant prior art, which will not be repeated here.

[0046] Optionally, in a possible implementation, the multiple battery swap demand prediction models include a first battery swap demand prediction model that predicts the battery swap demand based on the real-time power of the vehicle; the first battery swap demand prediction model is used to predict the battery swap probability of the vehicle in the next time period based on the real-time power of the vehicle; the first battery swap demand prediction model is obtained by fitting based on the status data of the vehicle in a historical time period; the status data includes real-time power and battery swap behavior data; wherein, the process of using the first battery swap demand prediction model to predict the battery swap demand in the next time period includes:

[0047] (1) Obtaining the current power of the target vehicle at the current moment; wherein the target vehicle is a vehicle within a preset distance of the battery swap station to be scheduled.

[0048] Specifically, the preset distance is set according to actual needs, and is not limited in this embodiment. For example, in one embodiment, the preset distance is 5 km. Further, the current power of the vehicle can be obtained through a system or sensor in the vehicle.

[0049] (2) Predicting the probability of the vehicle replacing its battery in the next time period based on the current power level and the first battery replacement demand prediction model.

[0050] Specifically, referring to the previous description, it can be seen that the first battery swap demand prediction model is a model obtained by fitting the real-time power of the vehicle and the battery swap behavior data in the historical time period. After the model is fitted, the current power of a vehicle can be input into the model, and the model outputs the probability of the vehicle swapping batteries in the next time period.

[0051] For example, in one embodiment, the current power level of the target vehicle 1 is 30%. Furthermore, in this step, the first battery replacement demand prediction model predicts that the probability of the target vehicle 1 replacing the battery in the next time period is 90%.

[0052] (3) Determine the battery replacement demand for the next time period based on the probability of each target vehicle replacing the battery in the next time period, and determine the battery replacement demand as the battery replacement demand for the next time period predicted by the first battery replacement demand prediction model.

[0053] Optionally, in a possible implementation, for example, when the probability of a vehicle replacing a battery in the next time period is greater than a preset value, it is considered that the vehicle needs to replace a battery in the next time period; further, the total number of vehicles that need to replace a battery in the next time period is counted, and this total number is the battery replacement demand in the next time period.

[0054] Specifically, the preset value is set according to actual needs and is not limited in this embodiment. For example, in one possible implementation, when the probability of the target vehicle replacing the battery in the next period is greater than 70%, it can be considered that the target vehicle needs to perform a battery replacement operation in the next period.

[0055] Furthermore, in a possible implementation, the multiple battery swap demand prediction models further include a second battery swap demand prediction model for predicting battery swap demand based on historical battery swap data of the battery swap station and a third battery swap demand prediction model for predicting battery swap demand based on historical battery swap data of the battery swap station; the second battery swap demand prediction model is different from the third battery swap demand prediction model; wherein the process of using the second battery swap demand prediction model to predict the battery swap demand in the next period includes:

[0056] Inputting the historical battery swap volume of the battery swap station to be scheduled within the fifth specified time period before the current moment into the second battery swap demand prediction model, so that the second battery swap demand prediction model predicts the battery swap demand volume in the next time period;

[0057] The process of using the third battery swap demand prediction model to predict the battery swap demand in the next period includes:

[0058] The historical battery swap volume of the battery swap station to be scheduled within the sixth specified time period before the current moment is input into the third battery swap demand prediction model, so that the third battery swap demand prediction model can predict the battery swap demand in the next time period.

[0059] Specifically, the historical battery swap volume of the battery swap station to be scheduled within the fifth specified time period before the current moment is input into the second battery swap demand prediction model, and the battery swap demand volume for the next time period is output by the second battery swap demand prediction model. Among them, the fifth specified time period is set according to actual needs, and in this embodiment, it is not limited. For example, in one possible implementation, the fifth specified time period is the past hour; for another example, in another possible implementation, the fifth specified time period is the past three hours, etc.

[0060] Furthermore, the third battery swap demand prediction model inputs the historical battery swap amount of the battery swap station to be scheduled within the sixth specified time period before the current moment, and the third battery swap demand prediction model outputs the battery swap demand for the next time period. Among them, the sixth specified time period is set according to actual needs, and in this embodiment, it is not limited. In addition, the sixth specified time period may be the same as the fifth specified time period, or it may be different, and in this embodiment, it is not limited. For example, in one possible implementation, the sixth specified time period may be the past hour.

[0061] It should be noted that the historical battery swap volume refers to the number of batteries actually replaced at the battery swap station to be scheduled.

[0062] Furthermore, the second battery swap demand prediction model is different from the third battery swap demand prediction model. For example, in one possible implementation, the second battery swap demand prediction model may be an LSTM model (Long Short-Term Memory, long short-term memory neural network), and the third battery swap demand prediction model may be an ARIAM model (Auto Regressive Integrated Moving Average, time series analysis model).

[0063] It is understandable that the LSTM model is a special type of recurrent neural network. By introducing "memory units" and "gating mechanisms", it can handle long-term dependencies well, maintain and utilize long-term information. The LSTM model can learn long-term dependencies and nonlinear relationships from complex patterns. Furthermore, the ARIAM model is a model for time series prediction that can effectively handle non-stationary time series data and convert it into a stationary series through differentiation. At the same time, the model parameters have clear interpretability. By combining the LSTM model with the ARIAM model, the prediction performance can be improved and adapted to more complex time series characteristics.

[0064] S102. Perform weighted processing on the battery swap demand predicted by each battery swap demand prediction model according to the weight of each battery swap demand prediction model at the current moment to obtain the weighted battery swap demand for the next time period; wherein the weight of each battery swap demand prediction model is dynamically updated based on the prediction error of the battery swap demand prediction model.

[0065] Specifically, the battery swap demand predicted by each battery swap demand prediction model can be weighted by the following formula:

[0066] ;

[0067] in, is the weighted battery replacement demand in the next period;

[0068] m is the total number of multiple battery swap demand prediction models;

[0069] is the weight of the j-th battery swap demand prediction model at the current moment;

[0070] The battery swap demand predicted by the j-th battery swap demand prediction model.

[0071] It should be noted that the weight of each battery swap demand prediction model is dynamically updated based on the prediction error of the battery swap demand prediction model. Through the prediction error, the weight corresponding to each battery swap demand prediction model can be dynamically updated, that is, the corresponding weight of each model can be adjusted according to the prediction error, so that the model with low prediction error is rewarded, and the model with high prediction error is punished, that is, the weight of the model with low prediction error is increased, and the weight of the model with high prediction error is reduced. Among them, the prediction error can be the mean square error or absolute error between the battery swap demand predicted by each battery swap demand prediction model and the actual battery swap demand. The specific update process of the weights will be described in detail in the following embodiments and will not be repeated here.

[0072] In specific implementation, combined with the above description, when multiple battery swap demand prediction models include a first battery swap demand prediction model M1, a second battery swap demand prediction model M2 and a third battery swap demand prediction model M3, further, for example, the weight of M1 at the current moment is 0.2, and its predicted battery swap demand is 10; the weight of M2 at the current moment is 0.35, and its predicted battery swap demand is 11; the weight of M3 at the current moment is 0.45, and its predicted battery swap demand is 14; at this time, after weighted processing of the battery swap demand predicted by each battery swap demand prediction model according to the weights of the three battery swap demand prediction models at the current moment, the battery swap demand for the next time period is 13.

[0073] It should be noted that the model for predicting the demand for battery swapping based on the real-time power of the vehicle predicts the demand for battery swapping based on the actual usage of the vehicle, while the model for predicting the demand for battery swapping based on the historical battery swapping data of the battery swapping station starts from the overall operating trend of the battery swapping station and uses long-term trends and rules to predict the demand for battery swapping. By combining multiple battery swapping demand prediction models to predict the demand for battery swapping, it is helpful to have a more comprehensive understanding of the driving factors of the demand for battery swapping and accurately predict the demand for battery swapping.

[0074] S103. If the total number of batteries held by the battery swap station to be scheduled at the current moment exceeds the weighted battery swap demand, and the number of the first batteries that are ready at the battery swap station to be scheduled at the current moment does not exceed the weighted battery swap demand, then for each battery that is currently not ready in the battery swap station to be scheduled, predict the charging time required for the battery to change from not ready to ready, and predict whether the battery will be ready in the next time period based on the charging time.

[0075] Specifically, the total number of batteries held by the battery swap station to be scheduled at the current moment refers to the total number of all batteries (including ready batteries and not ready batteries) contained in the battery swap station at the current moment; further, the first number of batteries ready at the battery swap station to be scheduled at the current moment refers to the total number of all ready batteries contained in the battery swap station at the current moment.

[0076] It should be noted that a ready battery refers to a battery whose battery power is greater than a preset ready power. The preset ready power is set according to actual needs and is not limited in this embodiment. For example, in a possible implementation, the preset ready power is 85%, that is, when the real-time power of a battery is greater than 85%, the battery is considered to be a ready battery.

[0077] Optionally, the total number of batteries held by the battery swap station to be scheduled at the current moment exceeds the weighted battery swap demand, which means that the total number of batteries held by the battery swap station to be scheduled at the current moment is greater than or equal to the weighted battery swap demand; similarly, the number of first batteries ready at the battery swap station to be scheduled at the current moment does not exceed the weighted battery swap demand, which means that the number of first batteries ready at the battery swap station to be scheduled at the current moment is less than the weighted battery swap demand.

[0078] Optionally, when the total number of batteries held by the battery swap station to be scheduled at the current moment is less than the weighted battery swap demand, at this time, the total number of batteries held by the battery swap station to be scheduled cannot meet the battery swap demand of the next time period; directly control the battery swap station to be scheduled to charge all unready batteries in the battery swap station. Furthermore, when the total number of batteries held by the battery swap station to be scheduled at the current moment is greater than or equal to the weighted battery swap demand, and the number of first batteries that are ready at the battery swap station to be scheduled at the current moment is greater than or equal to the weighted battery swap demand, at this time, the number of ready batteries can meet the battery swap demand of the next time period, and charging is stopped at this time.

[0079] In this step, if the total number of batteries held by the battery swap station to be scheduled at the current moment exceeds the weighted battery swap demand, and the number of the first batteries that are ready at the battery swap station to be scheduled at the current moment does not exceed the weighted battery swap demand, then for each battery that is currently not ready in the battery swap station to be scheduled, the charging time required for the battery to change from not ready to ready is predicted, and based on the charging time, it is predicted whether the battery will be ready in the next time period.

[0080] In a specific implementation, optionally, in a possible implementation, for each battery that is not currently ready in the battery swap station to be scheduled, the amount to be charged can be determined according to the current power of the battery and the preset ready power, and further, the charging time can be determined according to the charging power and the amount to be charged, as shown below:

[0081] ;

[0082] Among them, t is the charging time;

[0083] is the preset ready power;

[0084] is the current charge of the battery;

[0085] W is the charging power.

[0086] It should be noted that the charging power can be a preset empirical value, or it can be a value determined based on the historical charging data of the battery swap station or the historical charging data of the battery. In this embodiment, it is not limited to this.

[0087] For another example, in another possible implementation, the charging time can be predicted by fitting a first function of how the battery charge changes over time during the charging process based on the charging record of the battery within a first specified time period before the current moment. Furthermore, based on the current charge of the battery and a preset ready charge, the first function is used to calculate the charging time required for the battery to charge from the current charge to the ready charge.

[0088] It should be noted that the battery charging record may include information such as the start time of charging, the initial power during charging, the power at various moments during the charging process, and the end power at the end of charging. Furthermore, through the charging record of the battery in a first specified time period before the current moment, a first function of the battery power changing with time during the charging process is fitted, and the first function can reflect the charging performance of the battery; further, the current power of the battery and the preset ready power can be used as dependent variables, respectively, and these two dependent variables are substituted into the first function for calculation, and the charging start time corresponding to the current power of the battery and the charging end time corresponding to the preset ready power are obtained respectively, and then the functional relationship between the charging start time and the charging end time is used to determine the charging amount per unit time, and then the charging time is calculated based on the charging amount and the amount to be charged per unit time.

[0089] For another example, in another possible implementation, a second function of the battery charge changing over time during the charging process can be fitted based on the charging records of the battery swap station to be scheduled within a second specified time period before the current moment. Furthermore, based on the current charge of the battery and the preset ready charge, the second function is used to calculate the charging time required for the battery to charge from the current charge to the ready charge.

[0090] It should be noted that the charging record of the battery swap station to be scheduled may include the charging activity information of all batteries in the battery swap station, and the charging activity information may include the charging start time, charging end time, battery charging mode, and the power at different times during the charging process. Further, through the charging record of the battery swap station to be scheduled in the second specified time period before the current time, a second function of the battery power changing with time during the charging process can be fitted, and the second function can characterize the charging performance of the battery swap station; further, the current power of the battery and the preset ready power can be used as dependent variables, respectively, and these two dependent variables are substituted into the second function for calculation, respectively, to obtain the charging start time corresponding to the current power of the battery, and the charging end time corresponding to the preset ready power, and then use the functional relationship between the charging start time and the charging end time to determine the charging amount per unit time, and then calculate the charging time based on the charging amount and the amount to be charged per unit time.

[0091] Furthermore, for a battery that is not ready, after calculating the charging time of the battery, it is possible to determine whether the battery is ready in the next time period based on the charging time. Specifically, the time interval between the current time and the end time of the next time period can be calculated, and further, it is determined whether the charging time of a battery that is not ready is less than or equal to the above time interval. If so, it is determined that the battery that is not ready is ready in the next time period, otherwise, it is determined that the battery that is not ready is not ready in the next time period.

[0092] For example, in one possible implementation, the current time is 10:00, and the next time period is from 10:00 to 10:30. At this time, for a battery that is not ready, the predicted charging time of the battery is 20 minutes. At this time, it is determined that the battery will be ready in the next time period. For another example, for another battery that is not ready, the predicted charging time of the battery is 35 minutes. At this time, it is determined that the battery will not be ready in the next time period.

[0093] S104. When the sum of the first number of batteries and the number of batteries that are ready for use in the next time period is greater than or equal to the weighted battery swapping demand, control the battery swapping station to be scheduled to charge the batteries that are ready for use in the next time period.

[0094] Referring to the previous description, for a battery that is not currently ready, by predicting its charging time, it is possible to determine whether the battery will be ready in the next time period. Furthermore, based on the determination result, each unready battery can be labeled to identify by the label whether it will be ready in the next time period. For example, the label can be ready or not ready (specifically, when a battery is labeled as ready, it indicates that the battery is not ready at the current moment but will be ready in the next time period; similarly, when a battery is labeled as not ready, it indicates that the battery is not ready at the current moment and will not be ready in the next time period).

[0095] In specific implementation, the number of batteries labeled as ready can be counted, which is the number of batteries that can be ready in the next time period. Furthermore, the sum of the first number of batteries and the number of batteries that can be ready in the next time period can be calculated, and the sum can be compared with the weighted battery replacement demand.

[0096] It should be noted that when the sum is less than the weighted battery swapping demand, it means that even if the batteries that are ready for the next time period are charged immediately, the number of ready batteries in the next time period cannot meet the battery swapping demand. At this time, the battery swap stations to be scheduled are controlled to charge all batteries that are not ready.

[0097] Furthermore, when the sum is greater than or equal to the weighted battery swap demand, the batteries that are ready for the next period are immediately charged. In the next period, these batteries that are ready can be converted to ready after charging, so that the number of ready batteries can meet the battery swap demand in the next period. Therefore, at this time, the battery swap station to be scheduled is controlled to charge the batteries that are ready for the next period.

[0098] In one possible implementation, the battery swap station to be scheduled can be controlled to directly charge all batteries that are ready for the next time period. Furthermore, in another possible implementation, when the sum is equal to the weighted battery swap demand, all batteries that are ready for the next time period are directly charged, and when the sum is greater than the weighted battery swap demand, the difference between the weighted battery swap demand and the first number of batteries can be calculated first, and the difference is used as the actual number of batteries to be charged at the current moment. Furthermore, the actual number of batteries is randomly selected from all batteries that are ready for the next time period, and the battery swap station to be scheduled is controlled to charge the selected batteries.

[0099] It should be noted that in this embodiment, through accurate demand forecasting and battery scheduling, it is possible to ensure that there are sufficient batteries available for swapping during peak electricity prices, thereby ensuring that the use of power resources is optimized to the maximum extent possible without affecting the battery swap service, reducing operating costs, and ensuring the stability of the battery swap service.

[0100] The charging scheduling method provided in this embodiment, first, during the peak electricity price period, while ensuring that the battery swap service is not affected, priority is given to reducing the pressure on the power grid, and the charging rhythm is controlled by accurately predicting the weighted battery swap demand in the next period. When the number of batteries currently held exceeds the predicted weighted battery swap demand but the ready batteries are insufficient, by predicting the readiness of the batteries in the next period, only those batteries that can be charged in the next period are charged, which can not only avoid excessive charging operations during peak hours and save power resources, but also ensure that there are enough batteries available during peak demand, which can improve the operating efficiency of the battery swap station; secondly, during non-peak hours, the low valley electricity price can be fully utilized to charge the battery to ensure that there is enough battery supply during future peak hours. In this way, the fluctuation of electricity prices can be used to achieve more efficient use of energy and reduce the overall electricity cost; finally, when predicting the battery swap demand, multiple battery swap demand prediction models are used to predict the battery swap demand, and the weight of each model is adjusted by dynamically updating the prediction error, which helps to more comprehensively understand the driving factors of the battery swap demand and accurately predict the weighted battery swap demand.

[0101] Optionally, in a possible implementation, the method further includes:

[0102] Regularly update the parameters of the multiple battery swap demand prediction models.

[0103] Specifically, the specific update time of the regular update is set according to actual needs, and is not limited in this embodiment. For example, in one possible implementation, the parameters of the multiple battery swap demand prediction models can be regularly updated every week or every month.

[0104] The method provided in this embodiment can improve the prediction accuracy by regularly updating the parameters of the multiple battery swap demand prediction models.

[0105] Optional, Figure 2 This is a flow chart of the second embodiment of the charging scheduling method provided in this application. Please refer to Figure 2 In a possible implementation, when the sum of the first number of batteries and the number of batteries that are ready for use in the next time period is greater than the weighted battery swapping demand, controlling the battery swap station to be scheduled to charge the batteries that are ready for use in the next time period includes:

[0106] S201. Select a specific number of target batteries from the batteries that are ready for the next time period in order of remaining power from low to high; wherein the specific number is equal to the difference between the weighted battery replacement demand and the first number of batteries.

[0107] Specifically, when the first number of batteries and the number of batteries ready for the next time period are greater than the weighted battery replacement demand, the difference between the weighted battery replacement demand and the first number of batteries can be calculated, and the difference can be used as the actual number of batteries (i.e., the specific number) that actually need to be charged at the current moment.

[0108] Furthermore, a specific number of target batteries are selected from the batteries that are ready for use in the next period in order of remaining power from low to high.

[0109] S202: Control the battery swap station to be scheduled to charge the target battery.

[0110] For example, in one possible implementation, the number of batteries that are ready for the next time period is 10, the weighted battery replacement demand is 20, and the number of first batteries is 13. At this time, the sum of the number of first batteries and the number of batteries that are ready for the next time period is 23, which is greater than the battery replacement demand. In this step, the difference between the weighted battery replacement demand and the first number of batteries can be calculated, and the difference is 7. Therefore, in the actual charging process, only 7 ready batteries need to be charged to meet the battery replacement demand for the next time period. Furthermore, 7 target batteries are selected from the batteries that are ready for the next time period in order of remaining power from low to high, and these 7 target batteries are charged.

[0111] It should be noted that the target batteries are selected in order of remaining power from low to high, and the target batteries are charged. In this way, the batteries with higher power levels that are ready for the next time period can be retained and not charged at the current moment. In this way, once the actual battery replacement demand in the next time period is greater than the weighted battery replacement demand, the remaining batteries can be charged immediately. Since the remaining batteries have higher power levels, their charging time is relatively low, which can reduce waiting time.

[0112] The method provided in this embodiment selects a specific number of target batteries from the batteries that are ready in the next time period in order of remaining power from low to high when the sum of the first number of batteries and the number of batteries that can be ready in the next time period is greater than the weighted battery replacement demand, and then controls the battery replacement station to be scheduled to charge the target batteries. In this way, when the actual battery replacement demand in the next time period is greater than the weighted battery replacement demand, the remaining batteries can be charged immediately to reduce waiting time.

[0113] Figure 3 This is a flow chart of the third embodiment of the charging scheduling method provided in this application. Please refer to Figure 3 Based on the above embodiment, the step of predicting the charging time required for each battery that is not ready at present in the battery swap station to be scheduled to change from being not ready to being ready includes:

[0114] S301. Fit a first function of how the battery charge changes over time during the charging process, and a second function of how the battery charge changes over time during the charging process, based on the charging record of the battery within a first specified time period before the current moment, and the charging record of the battery swap station to be scheduled within a second specified time period before the current moment.

[0115] Specifically, a first function of how the battery charge changes over time during the charging process can be fitted based on the charging record of the battery within a first specified time period before the current moment. Similarly, a second function of how the battery charge changes over time during the charging process can be fitted based on the charging record of the battery swap station to be scheduled within a second specified time period before the current moment. Among them, the first function reflects the relationship between the battery charge and time during the charging process from the perspective of the battery itself, which can reflect the charging performance of the battery. Furthermore, the second function reflects the relationship between the battery charge and time during the charging process from the perspective of the battery swap station to be scheduled, which can reflect the charging performance of the battery swap station to be scheduled.

[0116] It should be noted that the battery charging record may include information such as the start time of charging, the initial power during charging, the power at each moment during the charging process, and the end power at the end of charging; further, the charging record of the battery swap station to be scheduled may include the charging activity information of all batteries in the battery swap station, and the charging activity information may include information such as the start time of charging, the end time of charging, the battery charging mode, the power at different moments during the charging process, and so on.

[0117] Furthermore, the first designated time period and the second designated time period are set according to actual needs and are not limited in this embodiment.

[0118] It should be noted that the first designated time period and the second designated time period may be the same or different, and this embodiment does not limit them. The following description is made by taking the first designated time period as 30 days and the second designated time period as 7 days as an example.

[0119] In specific implementation, the time and power of the charging process can be used as a set of data pairs, and then multiple sets of data pairs can be used to fit the functional relationship of the battery power changing with time during the charging process.

[0120] For example, when fitting the first functional relationship, multiple data pairs can be obtained through the charging records of the battery in the past 30 days. Further, the first functional relationship of the battery power changing with time during the charging process is fitted through these multiple data pairs.

[0121] For another example, when fitting the second functional relationship, multiple sets of data pairs can be obtained through the charging records of the battery swap station to be scheduled in the past 7 days. Furthermore, the second functional relationship of the battery charge changing over time during the charging process can be fitted through these multiple sets of data pairs.

[0122] S302: According to the current power of the battery and the preset ready power, the first function and the second function are used to calculate a first time required for the battery to be charged from the current power to the ready power and a second time required for the battery to be charged from the current power to the ready power.

[0123] In a specific implementation, a first function can be used to calculate a first duration required for charging the battery from the current power level to the ready power level based on the current power level of the battery and the preset ready power level; and a second function can be used to calculate a second duration required for charging the battery from the current power level to the ready power level based on the current power level of the battery and the preset ready power level.

[0124] It is understandable that in the first function, the power is the dependent variable and the time is the independent variable. In specific implementation, the current power of the battery and the preset ready power can be used as dependent variables, and the two dependent variables are substituted into the first function to calculate the charging start time corresponding to the current power and the charging end time corresponding to the preset ready power, and then the time difference between the two times is calculated to obtain the first duration. It is understandable that the first duration, from the perspective of the battery itself, describes the estimated charging time when charging the battery.

[0125] Similarly, when calculating the second duration, the current power of the battery and the preset ready power are used as dependent variables, and the two dependent variables are substituted into the second function to calculate the charging start time corresponding to the current power and the charging end time corresponding to the preset ready power, and then the time difference between the two times is calculated to obtain the second duration. It can be understood that the second duration, from the perspective of the battery swap station to be scheduled, indicates the estimated charging time when the battery is charged using the battery swap station to be scheduled.

[0126] Optionally, in another possible implementation, the current power of the battery and the preset ready power can be used as dependent variables, respectively, and the two dependent variables are substituted into the first function or the second function, respectively, to calculate the charging start time corresponding to the current power and the charging end time corresponding to the preset ready power. Furthermore, the functional relationship between the charging start time and the charging end time is used to determine the charging amount per unit time, and then the charging time is calculated based on the charging amount per unit time and the amount to be charged.

[0127] S303, weighting the first duration and the second duration according to the first weight of the battery at the current moment and the second weight of the battery swap station to be scheduled at the current moment to obtain a weighted duration, and determining the weighted duration as the charging time required for the battery to change from not ready to ready.

[0128] Specifically, the weighted duration can be calculated by the following formula:

[0129] ;

[0130] Among them, T is the weighted duration;

[0131] is the first weight;

[0132] is the second weight;

[0133] For the first duration;

[0134] For the second duration.

[0135] It should be noted that the first weight and the second weight may be fixed, that is, the first weight and the second weight are fixed values ​​preset in advance. Specifically, when the first weight and the second weight are fixed, the specific values ​​of the first weight and the second weight are set according to actual needs, and are not limited in this embodiment. For example, in one possible implementation, the first weight and the second weight are the same, and the first weight and the second weight are both 0.5. For another example, in another possible implementation, the first weight is 0.4 and the second weight is 0.6.

[0136] Furthermore, the first weight and the second weight may also be dynamically updated. Furthermore, when the first weight and the second weight are dynamically updated, in a possible implementation, a method for determining the first weight at a current moment and the second weight at a current moment may include:

[0137] (1) Calculate the sum of the specified value and the number of batteries that have been charged at the battery swap station to be scheduled within a third specified time period before the current moment; wherein the specified value is 1.

[0138] Specifically, the designated value and the third designated time period are set according to actual needs, and are not limited in this embodiment. For example, in one embodiment, the designated value is 1, and the third designated time period is 7 days before the current time.

[0139] Furthermore, the specified value and the sum of the number of batteries that have been charged by the battery swap station to be scheduled in the third specified time period before the current moment can be calculated by the following formula:

[0140] ;

[0141] Among them, n is the number of batteries that have been charged by the battery swap station to be scheduled within the third specified time period before the current moment, that is, the number of batteries that have been charged by the battery swap station to be scheduled within the third specified time period before the current moment.

[0142] (2) Determine the ratio of the number of batteries that have been charged at the battery swap station to be scheduled within the third specified time period before the current moment to the sum value as the second weight at the current moment.

[0143] Specifically, in combination with the above description, the second weight can be calculated by the following formula:

[0144] ;

[0145] (3) The ratio of the specified value to the sum value is determined as the first weight at the current moment.

[0146] Specifically, in combination with the above description, the first weight can be calculated by the following formula:

[0147] ;

[0148] The method provided in this embodiment fits the first function by utilizing the battery's own historical charging record to fit the first function, and the individual charging behavior of the battery can be accurately reflected through the first function. Furthermore, the second function is fit by the overall charging record of the battery swap station to be scheduled before the current moment, and the second function can be used to capture factors such as the usage of charging resources in the battery swap station and the overall performance of the charging equipment to reflect the charging performance of the battery swap station. In this way, when the first function and the second function are combined to estimate the charging time, the individual charging performance of the battery and the overall charging performance of the battery swap station to be scheduled can be combined, and the differences between individual batteries and the overall charging environment can be accurately analyzed to improve the accuracy of the charging time prediction and avoid errors caused by individual differences and environmental factors.

[0149] In addition, by assigning different weights to batteries and battery swap stations to be dispatched for weighted processing, we can flexibly respond to needs in different situations and improve the accuracy of prediction.

[0150] Furthermore, in the process of dynamically updating the weight, by calculating the sum of the designated value and the number of batteries that have been charged by the battery swap station to be scheduled in the third designated period before the current moment, the ratio of the number of batteries that have been charged by the battery swap station to be scheduled in the third designated period before the current moment to the sum is determined as the second weight at the current moment, and the ratio of the designated value to the sum is determined as the first weight at the current moment. In this way, the relative influence of the first weight (individual battery) and the second weight (overall charging trend) can be dynamically adjusted to ensure that the influence of the overall trend increases when the charging records are abundant, and when the charging records are less, it is more dependent on the historical data of individual batteries to improve the accuracy of the prediction of charging time.

[0151] Figure 4 This is a flowchart of the fourth embodiment of the charging scheduling method provided by this application. Please refer to Figure 4 Based on the above embodiment, the weight updating process of each battery swap demand prediction model includes:

[0152] S401. When the preset update cycle arrives, for each battery swap demand prediction model, calculate the prediction error of the battery swap demand prediction model based on the predicted battery swap demand predicted by the battery swap demand prediction model in the fourth specified time period before the current moment, and the actual battery swap demand in the fourth specified time period.

[0153] Specifically, the specific value of the preset update cycle is set according to actual needs, and is not limited in this embodiment. For example, in one embodiment, the preset update cycle is 1 hour, that is, the weights of each battery swap demand prediction model are updated every hour.

[0154] Furthermore, when the preset update cycle arrives, for each battery swap demand prediction model, the prediction error of the battery swap demand prediction model is calculated. The prediction error of each battery swap demand prediction model can be the mean square error or absolute error between the predicted battery swap demand and the actual battery swap quantity. Among them, the actual battery swap quantity refers to the number of batteries actually replaced by the battery swap station to be scheduled within the fourth specified time period before the current moment. For the calculation method of the prediction error, please refer to the introduction in the relevant technology, which will not be repeated here.

[0155] It should be noted that the fourth specified time period is set according to actual needs and is not limited in this embodiment. For example, in one embodiment, the fourth specified time period may be 1 hour before the current time.

[0156] S402. For each battery swap demand prediction model, calculate the weight update coefficient of the battery swap demand prediction model based on the prediction error of the battery swap demand prediction model, the maximum value of the prediction errors of all battery swap demand prediction models, and the minimum value of the prediction errors of all battery swap demand prediction models.

[0157] Specifically, the weight update coefficient can be calculated by the following formula:

[0158] ;

[0159] in, is the weight update coefficient of the i-th battery swap demand prediction model;

[0160] is the preset adjustment coefficient;

[0161] is the prediction error of the i-th battery swap demand prediction model;

[0162] It is the maximum value of the prediction error of all battery swap demand prediction models;

[0163] It is the minimum value of the prediction error of all battery swap demand prediction models.

[0164] S403. Calculate the updated weight of the battery swap demand prediction model according to the current weight of the battery swap demand prediction model and the weight update coefficient of the battery swap demand prediction model.

[0165] Specifically, the update weight can be calculated by the following formula:

[0166] ;

[0167] in, is the update weight of the i-th battery swap demand prediction model;

[0168] is the weight of the i-th battery swap demand prediction model at the current moment;

[0169] is the weight update coefficient of the i-th battery swap demand prediction model.

[0170] S404. Normalize the updated weights of all battery swap demand prediction models to obtain the updated weights of each battery swap demand prediction model.

[0171] Specifically, the sum of the updated weights of all battery swap demand prediction models can be calculated, and then the updated weights of each battery swap demand prediction model can be normalized using the sum to ensure that the sum of the updated weights is 1.

[0172] In specific implementation, the sum of updated weights can be calculated by the following formula:

[0173] ;

[0174] Among them, SUM is the sum of the update weights of all battery swap demand prediction models;

[0175] is the update weight of the i-th battery swap demand prediction model;

[0176] m is the total number of multiple battery swap demand prediction models.

[0177] Furthermore, the update weight of each battery swap demand prediction model can be normalized by the following formula:

[0178] ;

[0179] Among them, w i is the normalized weight of the i-th battery swap demand forecasting model, that is, the updated weight of the i-th battery swap demand forecasting model;

[0180] is the update weight of the i-th battery swap demand prediction model;

[0181] SUM is the sum of the updated weights of all battery swap demand prediction models.

[0182] Referring to the previous description, it can be seen that in this embodiment, the weight of each model is adjusted according to the actual prediction performance of the model. When the prediction error of a model is small, it means that the model performs better in the current situation, and its weight will increase accordingly, otherwise it will decrease. It can automatically adapt to different prediction environments and demand changes, making the prediction more accurate. In addition, because the prediction model of each model is comprehensively considered when updating the weight, the weight adjustment will tend to reduce the influence of the model with a larger error. In this way, the impact of the extreme error caused by the model with a larger error on the final weighted battery replacement demand can be reduced, and the accuracy of the prediction can be improved. Furthermore, in different time periods and situations, some models perform better. Through weight adjustment, better performing models can be flexibly used without completely abandoning the poorly performing models, and computing resources can be used efficiently.

[0183] The method provided in this embodiment can ensure a more accurate prediction by evaluating the prediction error of each battery swap demand prediction model and dynamically updating the weight of each battery swap demand prediction model based on the prediction error.

[0184] Corresponding to the aforementioned embodiment of a charging scheduling method, the present application also provides an embodiment of a charging scheduling device.

[0185] The present application provides an embodiment of a charging scheduling device that can be applied to charging scheduling equipment. The device embodiment can be implemented by software, or by hardware or a combination of software and hardware. Taking software implementation as an example, as a device in a logical sense, it is formed by the processor of the charging scheduling device in which it is located reading the corresponding computer program instructions in the non-volatile memory into the memory and running them. From the hardware level, if Figure 5 As shown, it is a hardware structure diagram of the charging scheduling device where the charging scheduling device provided by this application is located, except Figure 5 In addition to the processor, memory, network interface, and non-volatile memory shown, the charging scheduling device in which the device is located in the embodiment may also include other hardware according to the actual function of the charging scheduling device, which will not be described in detail.

[0186] Figure 6 This is a structural diagram of the first embodiment of the charging scheduling device provided by this application. Figure 6 The device provided in this embodiment includes a prediction module 610, an update module 620 and a control module 630; wherein,

[0187] The prediction module 610 is used for predicting the battery swap demand in the next period of time by using multiple battery swap demand prediction models for the battery swap station to be scheduled when the preset monitoring period arrives, if the electricity price at the current moment is not a valley price; wherein each of the multiple battery swap demand prediction models is different, and the multiple battery swap demand prediction models include at least one model for predicting the battery swap demand based on the real-time power consumption of the vehicle, and a model for predicting the battery swap demand based on the historical battery swap data of the battery swap station;

[0188] The updating module 620 is used to perform weighted processing on the battery swap demand predicted by each battery swap demand prediction model according to the weight of each battery swap demand prediction model at the current moment, so as to obtain the weighted battery swap demand for the next period; wherein the weight of each battery swap demand prediction model is dynamically updated based on the prediction error of the battery swap demand prediction model;

[0189] The prediction module 610 is further configured to, when it is determined that the total number of batteries held by the battery swap station to be scheduled at the current moment exceeds the weighted battery swap demand, and the number of first batteries that are ready at the battery swap station to be scheduled at the current moment does not exceed the weighted battery swap demand, predict the charging time required for each battery that is currently not ready in the battery swap station to be scheduled to change from being not ready to being ready, and predict whether the battery will be ready in the next time period based on the charging time;

[0190] The control module 630 is used to control the battery swap station to be scheduled to charge the batteries that are ready for the next time period when the sum of the first number of batteries and the number of batteries that are ready for the next time period is greater than or equal to the weighted battery swap demand.

[0191] The device of this embodiment can be used to perform Figure 1 The steps, specific implementation principles and implementation processes of the method embodiment shown are similar and will not be repeated here.

[0192] Please continue to refer to Figure 5 The present application also provides a charging scheduling device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the steps of any one of the methods provided in the first aspect of the present application are implemented.

[0193] The implementation process of the functions and effects of each unit in the above-mentioned device is specifically described in the implementation process of the corresponding steps in the above-mentioned method, and will not be repeated here.

[0194] For the device embodiment, since it basically corresponds to the method embodiment, the relevant parts can refer to the partial description of the method embodiment. The device embodiment described above is only schematic, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the present application scheme. A person of ordinary skill in the art can understand and implement it without paying any creative work.

[0195] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application shall be included in the scope of protection of the present application.

Claims

1. A charging scheduling method, characterized in that: The method comprises: For the battery swap station to be scheduled, when the preset monitoring period arrives, if the current electricity price is not a valley price, multiple battery swap demand prediction models are used to predict the battery swap demand in the next period; wherein each of the multiple battery swap demand prediction models is different, and the multiple battery swap demand prediction models include at least one model for predicting the battery swap demand based on the real-time power consumption of the vehicle, and a model for predicting the battery swap demand based on the historical battery swap data of the battery swap station; The battery swap demand predicted by each battery swap demand prediction model is weighted according to the weight of each battery swap demand prediction model at the current moment to obtain the weighted battery swap demand for the next period; wherein the weight of each battery swap demand prediction model is dynamically updated based on the prediction error of the battery swap demand prediction model; If the total number of batteries held by the battery swap station to be scheduled at the current moment exceeds the weighted battery swap demand, and the number of first batteries that are ready at the battery swap station to be scheduled at the current moment does not exceed the weighted battery swap demand, then for each battery that is currently not ready in the battery swap station to be scheduled, predict the charging time required for the battery to change from not ready to ready, and predict whether the battery will be ready in the next time period based on the charging time; When the sum of the first number of batteries and the number of batteries that are ready for the next time period is greater than or equal to the weighted battery swapping demand, the battery swapping station to be scheduled is controlled to charge the batteries that are ready for the next time period.

2. The method according to claim 1, characterized in that When the sum of the first number of batteries and the number of batteries that are ready for use in the next time period is greater than the weighted battery swapping demand, controlling the battery swapping station to be scheduled to charge the batteries that are ready for use in the next time period includes: Selecting a specific number of target batteries from the batteries that are ready for use in the next time period in order of remaining power from low to high; wherein the specific number is equal to the difference between the weighted battery replacement demand and the first number of batteries; Control the battery swap station to be scheduled to charge the target battery.

3. The method according to claim 1, characterized in that The predicting of the charging time required for each battery that is not currently ready in the battery swap station to be scheduled to change from being not ready to being ready includes: Fitting a first function of the battery power varying with time during the charging process and a second function of the battery power varying with time during the charging process according to the charging record of the battery in a first specified time period before the current moment and the charging record of the battery swap station to be scheduled in a second specified time period before the current moment; According to the current power of the battery and the preset ready power, respectively using the first function and the second function, calculate a first time required for the battery to be charged from the current power to the ready power and a second time required for the battery to be charged from the current power to the ready power; The first duration and the second duration are weighted according to the first weight of the battery at the current moment and the second weight of the battery swap station to be scheduled at the current moment to obtain a weighted duration, and the weighted duration is determined as the charging time required for the battery to change from being not ready to being ready.

4. The method according to claim 3, characterized in that The first weight and the second weight are dynamically updated; A method for determining a first weight at a current moment and a second weight at a current moment includes: Calculate the sum of the specified value and the number of batteries that have been charged by the battery swap station to be scheduled within a third specified time period before the current moment; wherein the specified value is 1; Determine the ratio of the number of batteries that have been charged at the battery swap station to be scheduled within the third specified time period before the current moment to the sum value as the second weight at the current moment; The ratio of the designated value to the sum value is determined as the first weight at the current moment.

5. The method according to claim 1, characterized in that The weight updating process of each battery swap demand prediction model includes: When the preset update cycle arrives, for each battery swap demand prediction model, the prediction error of the battery swap demand prediction model is calculated according to the predicted battery swap demand predicted by the battery swap demand prediction model in the fourth specified time period before the current moment and the actual battery swap demand in the fourth specified time period; For each battery swap demand prediction model, calculate the weight update coefficient of the battery swap demand prediction model according to the prediction error of the battery swap demand prediction model, the maximum value of the prediction errors of all battery swap demand prediction models, and the minimum value of the prediction errors of all battery swap demand prediction models; Calculate the updated weight of the battery swap demand prediction model according to the current weight of the battery swap demand prediction model and the weight update coefficient of the battery swap demand prediction model; The updated weights of all battery swap demand prediction models are normalized to obtain the updated weights of each battery swap demand prediction model.

6. The method according to claim 1, characterized in that The multiple battery swap demand prediction models include a first battery swap demand prediction model for predicting the battery swap demand based on the real-time power of the vehicle; the first battery swap demand prediction model is used to predict the battery swap probability of the vehicle in the next time period based on the real-time power of the vehicle; the first battery swap demand prediction model is obtained by fitting based on the state data of the vehicle in a historical time period; the state data includes real-time power and battery swap behavior data; wherein, the process of using the first battery swap demand prediction model to predict the battery swap demand in the next time period includes: Obtaining the current power of the target vehicle at the current moment; wherein the target vehicle is a vehicle within a preset distance of the battery swap station to be scheduled; Predicting the probability of the vehicle replacing a battery in the next time period according to the current power and the first battery replacement demand prediction model; According to the probability of battery replacement for each target vehicle in the next time period, the battery replacement demand for the next time period is determined, and the battery replacement demand is determined as the battery replacement demand for the next time period predicted by the first battery replacement demand prediction model.

7. The method according to claim 1 or 6, characterized in that: The multiple battery swap demand prediction models include a second battery swap demand prediction model for predicting battery swap demand based on historical battery swap data of a battery swap station and a third battery swap demand prediction model for predicting battery swap demand based on historical battery swap data of a battery swap station; the second battery swap demand prediction model is different from the third battery swap demand prediction model; wherein the process of using the second battery swap demand prediction model to predict the battery swap demand for the next period includes: Inputting the historical battery swap volume of the battery swap station to be scheduled within the fifth specified time period before the current moment into the second battery swap demand prediction model, so that the second battery swap demand prediction model predicts the battery swap demand volume in the next time period; The process of using the third battery swap demand prediction model to predict the battery swap demand in the next period includes: The historical battery swap volume of the battery swap station to be scheduled within the sixth specified time period before the current moment is input into the third battery swap demand prediction model, so that the third battery swap demand prediction model can predict the battery swap demand in the next time period.

8. The method according to claim 1, characterized in that The method further comprises: Regularly update the parameters of the multiple battery swap demand prediction models.

9. A charging scheduling device, characterized in that: The device comprises a prediction module, an update module and a control module; wherein, The prediction module is used for predicting the battery swap demand in the next period of time using multiple battery swap demand prediction models for the battery swap station to be scheduled when the preset monitoring period arrives, if the electricity price at the current moment is not a valley price; wherein each of the multiple battery swap demand prediction models is different, and the multiple battery swap demand prediction models include at least one model for predicting the battery swap demand based on the real-time power consumption of the vehicle, and a model for predicting the battery swap demand based on the historical battery swap data of the battery swap station; The updating module is used to perform weighted processing on the battery swap demand predicted by each battery swap demand prediction model according to the weight of each battery swap demand prediction model at the current moment, so as to obtain the weighted battery swap demand for the next period; wherein the weight of each battery swap demand prediction model is dynamically updated based on the prediction error of the battery swap demand prediction model; The prediction module is further used to, when it is judged that the total number of batteries held by the battery swap station to be scheduled at the current moment exceeds the weighted battery swap demand, and the number of first batteries that are ready at the battery swap station to be scheduled at the current moment does not exceed the weighted battery swap demand, predict the charging time required for each battery that is currently not ready in the battery swap station to be scheduled to change from not ready to ready, and predict whether the battery will be ready in the next time period based on the charging time; The control module is used to control the battery swap station to be scheduled to charge the batteries that are ready for the next time period when the sum of the first number of batteries and the number of batteries that are ready for the next time period is greater than or equal to the weighted battery swap demand.

10. A charging scheduling device, characterized in that: The method comprises a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the method according to any one of claims 1 to 8 when executing the program.

Images