Energy storage vehicle reverse charging behavior prediction and regulation method suitable for V2G technology

Abstract

The invention provides an energy storage vehicle reverse charging behavior prediction, regulation and control method suitable for a V2G technology, and the method comprises the steps: predicting the batch reverse charging behavior of vehicles through probability statistics and deep learning based on V2G equipment, time, date and discharge SOC interval information, estimating the electric energy obtained by different V2G equipment in different time periods, and carrying out the prediction of the batch reverse charging behavior of vehicles, and power dispatching is carried out according to the actual power consumption demand and the load condition of each power grid node. According to the method, the reverse charging behavior of the vehicle can be guided and regulated in cooperation with the regulation means of time-sharing independent pricing of the V2G facility, the utilization efficiency of energy is improved, the impact and transformation cost of local electric energy excess and scattered vehicle-mounted electric energy on a power grid are avoided, design planning of V2G equipment can be assisted, and the V2G technology can better play a proper role.

Description

Technical Field

[0001] The present invention belongs to the technical fields such as V2G, and particularly relates to a method for predicting and regulating the reverse charging behavior of energy storage vehicles applicable to V2G technology. Background Art

[0002] At present, with the rapid growth of the number of new energy electric vehicles and charging demands, the impacts and fluctuations received by the power grid are more obvious. The load growth during peak electricity consumption periods is 20% - 30% higher than that during off-peak periods, which poses new challenges to the safe and stable operation of the power grid for power dispatching. For this reason, some existing technologies have proposed the concept of V2G, aiming to reverse charge the electric energy stored in the vehicle's power battery to a large-capacity energy storage device or send it to the power grid through vehicle-grid interaction, so as to relieve the power grid pressure during peak periods. However, it is difficult to accurately match the reverse charging behavior of energy storage vehicles with the actual power demand. When the electric energy from the vehicle is in oversupply, inevitably sending the excess electric energy into the power grid in a scattered manner will cause power conversion burdens and loads, and may make the final cost-effectiveness of V2G technology not significantly better than the traditional power supply-side power grid peak shaving method directly using generator sets, thus restricting the popularization and application of V2G technology. Regarding the balance between power demand and supply, the common regulation method is to set different electricity prices for peak and off-peak electricity consumption periods. However, in the context that new energy power generation is prone to overproduction and "negative electricity prices" occur in some regions at the present stage, this traditional method is no longer applicable to the regulation of power supply and demand in V2G scenarios with highly uncertain energy supply. Summary of the Invention

[0003] In view of this, aiming at the technical problems existing in this field, the present invention provides a method for predicting and regulating the reverse charging behavior of energy storage vehicles applicable to V2G technology.

[0004] The present invention specifically adopts the following technical solutions: A method for predicting and regulating the reverse charging behavior of energy storage vehicles applicable to V2G technology: Based on the information of V2G devices, time, date, and discharge SOC intervals, through probability statistics and deep learning to predict the batch reverse charging behavior of vehicles, estimate the electric energy that can be obtained by different V2G devices at different times, so as to conduct power dispatching according to the actual power demand and load conditions of each power grid node.

[0005] Further, the prediction of the batch reverse charging behavior of vehicles through probability statistics and deep learning based on the information of V2G devices, time, date, and discharge SOC interval is specifically as follows: Extract historical operation data of electric vehicles as energy storage sources. After data cleaning and division of operation segments, charging data segments and discharge data segments are obtained respectively; based on vehicle speed, battery current, SOC change under parking state, parking location, and communication tags between the vehicle and the V2G device in the discharge data segments, reverse charging data segments in which each vehicle performs reverse charging to the V2G device are screened and obtained; and reverse charging geographical coordinates, start and end times, dates, and reverse charging start and end SOC parameters are extracted from the reverse charging data segments; count each parameter and determine the geographical coordinates of the V2G device where each vehicle is accustomed to performing reverse charging, the start time and date of reverse charging start and end, and the reverse charging start and end SOC, calculate the probability of each parameter as the prior distribution; calculate the conditional probability of different reverse charging start and end times, dates, and start and end SOC respectively with the geographical coordinates where different V2G devices are located as the condition; based on the assumption that the probability of a vehicle selecting different V2G facilities under the conditions of specific reverse charging start and end times, dates, and start and end SOC is independent of other conditions, combine the prior distribution to calculate the posterior distribution of selecting each V2G device under different conditions; establish a V2G device prediction model based on deep learning, with the peak power consumption period, date, and corresponding reverse charging start SOC data of each vehicle as the input, and the calculated posterior distribution of the V2G device and the SOC change amount as the output, and use the historical operation data to train the model until it converges stably.

[0006] Further, based on the trained V2G device prediction model, predict the V2G device with the highest probability that may be selected currently and the available reverse charging capacity according to the time, date, and SOC parameters collected from each vehicle, and determine the available reverse charging capacity of each V2G device according to the prediction result.

[0007] Further, combined with the corresponding grid load at each V2G device during peak hours, the maximum storage capacity of the V2G device, and the available reverse charging capacity, set corresponding reverse charging electricity prices for different V2G devices respectively, and perform corresponding power peak shaving dispatching.

[0008] Further, the method for screening and obtaining the reverse charging data segments in which each vehicle performs reverse charging to the V2G device is specifically as follows: By traversing the segments that simultaneously meet the conditions of vehicle speed being 0, battery current being positive, SOC dropping by more than a certain value, and communication occurring between the vehicle and the V2G device, and then matching the geographical coordinates of these segments during parking discharge with the coordinates of each V2G device in the digital map within a certain range, so as to screen out the reverse charging data segments from the discharge segments.

[0009] Further, the specific process of calculating the prior distribution of each parameter includes: after matching the parking location geographical coordinates of the reverse charging data segments with the geographical coordinates of each V2G device in the digital map according to the distance between them, adding the corresponding V2G device label Ln to each reverse charging data segment, indicating the coordinate L of the nth V2G device; then counting and sorting the number of V2G device markings of each reverse charging data segment to obtain the V2G devices that each vehicle is accustomed to using and the corresponding probability P(Ln), where n represents the nth V2G device; extracting the start and end times of reverse charging in the reverse charging data segments, and dividing each day into several time periods at equal time intervals and adding the time period label hm, where m represents the mth time period in a day; then counting and sorting the time period labels at the start and end of reverse charging each day to obtain the start and end time periods of reverse charging that each vehicle is accustomed to and the corresponding probability P(hm); extracting or converting the date of the reverse charging data segment to obtain the corresponding weekly calendar date and adding the weekly calendar label wi, where i = 1~7; then counting and sorting the weekly calendar labels to obtain the weekly calendar dates of reverse charging habits and the corresponding probability P(wi); dividing the battery SOC from 100% to 0% into several SOC intervals at equal intervals and adding the corresponding SOC interval label SOCi for the SOC change of each reverse charging data segment, where i represents the ith SOC interval; then counting and sorting each SOC interval label to obtain the SOC intervals consumed by each vehicle's reverse charging habits and the corresponding probability P(SOCi).

[0010] Further, in the process of the posterior calculation: first, calculate the conditional probabilities of the geographical coordinates of the V2G device for different reverse charging start and end times, dates, and start and end SOCs respectively: P(hm | Ln), P(wi | Ln), P(SOCi | Ln); For a combination Ω = [hm, wi, SOCi] of specific reverse charging start and end times, dates, and start and end SOC conditions, calculate the prior probability: P(Ω) = P(hm) × P(wi) × P(SOCi) And the conditional probability: P(Ω | Ln) = P(hm | Ln) × P(wi | Ln) × P(SOCi | Ln); Use Bayes' formula to calculate the posterior distribution of selecting each V2G device under different conditions: P(Ln | Ω) = P(Ln) × P(Ω | Ln) / P(Ω).

[0011] Further, during the training process of the V2G device prediction model, vehicle VIN, time and date tags, start and end SOC tags, V2G device tags, and the calculated posterior distribution are extracted from the reverse charging data segments during peak electricity consumption periods to construct corresponding training sets and validation sets, and the model is trained and the training effect is verified; at each specified time interval, by re-collecting the historical operation data of energy storage vehicles and executing the data processing and model training processes again, the V2G device prediction model is updated.

[0012] Moreover, an electronic device includes a memory, a processor, and a computer program stored on the memory and executable on the processor, wherein the processor implements the steps of the method for predicting and regulating the reverse charging behavior of energy storage vehicles applicable to V2G technology as described above when executing the program.

[0013] A non-transitory computer-readable storage medium stores a computer program, wherein the computer program implements the steps of the method for predicting and regulating the reverse charging behavior of energy storage vehicles applicable to V2G technology as described above when executed by a processor.

[0014] Compared with the prior art, the present invention and its preferred solutions, for information such as V2G devices, time, date, and SOC intervals selected by user habits in vehicle-grid interaction, realize the prediction of batch reverse charging behaviors of vehicles through probability statistics and deep learning methods, so as to estimate the available capacity that can be obtained by different V2G devices at different times, which helps to perform more accurate power dispatching according to the actual electricity consumption demands and load conditions of each power grid node. With the regulation means of separately pricing V2G facilities by time, it is possible to guide and regulate the reverse charging behaviors of vehicles, not only improving the energy utilization efficiency, avoiding local electricity overproduction and the impact and conversion costs of scattered in-vehicle electricity on the power grid, but also assisting in the design and planning of V2G devices, enabling V2G to better play its due role. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] The following further details the present invention in conjunction with the drawings and specific embodiments: Figure 1 It is a schematic flowchart of the method provided by the embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0016] Hereinafter, specific embodiments of the present application will be described in detail with reference to the drawings. According to these detailed descriptions, those skilled in the art can clearly understand the present application and can implement the present application. Without departing from the principles of the present application, the features in different embodiments can be combined to obtain new implementation manners, or some features in certain embodiments can be replaced to obtain other preferred implementation manners.

[0017] To make the features and advantages of the present invention more obvious and understandable, specific embodiments are given below and described in detail in conjunction with the accompanying drawings as follows: An embodiment of the present invention provides a method for predicting and regulating the reverse charging behavior of an energy storage vehicle applicable to V2G technology, as Figure 1 shown, which specifically includes the following steps: Step 1: Extract the historical operation data of the electric vehicle serving as an energy storage source; perform data cleaning and operation segment division on the historical operation data to obtain charging data segments and discharging data segments respectively; Step 2: Screen out the reverse charging data segments in which each vehicle performs reverse charging to V2G devices such as bidirectional charging piles and large-capacity energy storage cabinets from the discharging data segments based on vehicle speed, battery current, SOC change under parking state, parking location, communication tags between the vehicle and V2G devices, etc.; extract the reverse charging geographical coordinates, start and end times, dates, and reverse charging start and end SOC parameters from the reverse charging data segments; count each parameter and determine the geographical coordinates of the V2G devices where each vehicle is accustomed to performing reverse charging, the start time and date of reverse charging start and end, and the reverse charging start and end SOC, and calculate the probability of each parameter as the prior distribution; Step 3: Calculate the conditional probabilities of different reverse charging start and end times, dates, and start and end SOC respectively with the geographical coordinates where different V2G devices are located as conditions; based on the assumption that the probability of a vehicle selecting different V2G facilities under specific reverse charging start and end times, dates, and start and end SOC conditions is independent of other conditions, calculate the posterior distribution of each V2G device selected under different conditions in combination with the prior distribution obtained in Step 2; Step 4: Establish a V2G device prediction model based on deep learning, with the data of each vehicle at peak power consumption times, dates, and corresponding reverse charging start SOC as inputs, and the calculated posterior distribution of the V2G device and the SOC change amount as outputs, and use the historical operation data to train the model until it converges stably; Step 5: Apply the trained V2G device prediction model online, predict the V2G device with the highest probability of being selected and the available reverse charging capacity according to the time, date, and SOC parameters collected from each vehicle, and determine the available reverse charging capacity of each V2G device according to the prediction results; of course, the available reverse charging capacity should also consider the part of the electric energy stored by the V2G device; Step 6: Combine the corresponding grid loads at each V2G device during peak hours, the maximum storage capacity of the V2G device, and the available reverse charging capacity, and set corresponding reverse charging electricity prices for different V2G devices respectively, and perform corresponding power peak regulation and dispatching.

[0018] By performing the above five steps, the historical reverse charging behavior habits and patterns of a large number of energy storage vehicles are statistically analyzed and deeply learned, enabling the prediction of the time, space, and available power supply options when individual vehicles perform reverse charging. As a result, targeted peak shaving and valley filling can be achieved for each power grid node. By using price regulation measures, it is also possible to guide and adjust the reverse charging habits of each vehicle user based on the actual electricity demand within a certain period, such as monthly or seasonal, to improve the energy utilization rate.

[0019] In a preferred embodiment of the present invention, in step two, the specific process of traversing the segments that meet the conditions such as a vehicle speed of 0, a positive battery current, an SOC drop of more than 10%, and the communication tag between the vehicle and the V2G device is carried out. And the geographical coordinates of the parked vehicle discharging in these segments are matched with the coordinates of each V2G device in the digital map within a certain range. For example, when the distance from a certain V2G device is within 30m, it can be considered the same geographical coordinate. Thus, the short-term parking power consumption data segments unrelated to reverse charging are removed from the discharging segments, and the reverse charging data segments are screened out.

[0020] In a preferred embodiment of the present invention, the specific process of calculating the prior distribution of each parameter in step two includes: ① After matching the geographical coordinates of the parking positions of the reverse charging data segments with the geographical coordinates of each V2G device in the digital map according to the distance between them, add the corresponding V2G device label Ln to each reverse charging data segment, indicating the nth V2G device coordinate L; then count and sort the number of V2G device markings of each reverse charging data segment to obtain the V2G devices that each vehicle is accustomed to using and the corresponding probability P(Ln), where n represents the nth V2G device; ② Extract the start and end times of reverse charging in the reverse charging data segments, and divide each day into several time periods at equal time intervals, such as half an hour, and add the time period label hm, where m represents the mth daily time period; then count and sort the time period labels at the start and end of reverse charging each day to obtain the start and end time periods of reverse charging that each vehicle is accustomed to and the corresponding probability P(hm); ③ Extract or convert the date of the reverse charging data segment to obtain the corresponding weekly calendar date and add the weekly calendar label wi, where i = 1~7; then count and sort the weekly calendar labels to obtain the weekly calendar dates of the accustomed reverse charging and the corresponding probability P(wi); ④ Divide the battery SOC from 100% to 0% into several SOC intervals at equal intervals, such as 10% SOC, and add the corresponding SOC interval label SOCi to the SOC change of each reverse charging data segment, where i represents the ith SOC interval; then count and sort each SOC interval label to obtain the SOC intervals consumed by each vehicle's accustomed reverse charging and the corresponding probability P(SOCi).

[0021] In a preferred embodiment of the present invention, in step three, the conditional probabilities of the geographical coordinates of the V2G device for different reverse charging start and end times, dates, and start and end SOCs are first calculated respectively: P(hm | Ln), P(wi | Ln), P(SOCi | Ln); For a combination Ω = [hm, wi, SOCi] of specific reverse charging start and end times, dates, and start and end SOC conditions, such as the corresponding parameter conditions for the peak electricity consumption period from 15:00 to 17:00 during the night commute on Monday, the prior probability is calculated: P(Ω) = P(hm) × P(wi) × P(SOCi) And the conditional probability: P(Ω | Ln) = P(hm | Ln) × P(wi | Ln) × P(SOCi | Ln); Finally, the posterior distribution of selecting each V2G device under different conditions is calculated using Bayes' formula: P(Ln | Ω) = P(Ln) × P(Ω | Ln) / P(Ω).

[0022] In a preferred embodiment of the present invention, in step four, the vehicle VIN, time and date tags, start and end SOC tags, V2G device tags, and the calculated posterior distribution in the reverse charging data segment during the peak electricity consumption period are specifically extracted to construct the corresponding training set and validation set, and the neural network model based on deep learning is trained and the training effect is verified; at each specified time interval, by re-collecting the historical operation data of the energy storage vehicles and executing the data processing and model training processes of steps one to four again, the V2G device prediction model is updated according to the change of the reverse charging habit rules of the vehicle group within a certain period.

[0023] In a preferred embodiment of the present invention, in step six, the power grid also sends electricity consumption requests during the peak electricity consumption period to each energy storage vehicle in advance based on the model prediction results, and determines the available capacity that each V2G device can actually obtain based on the response of the vehicle to the electricity consumption request. On this basis, power dispatching is coordinated with the generator set. And corresponding electricity prices are set for the vehicles going to different V2G devices, so as to guide some price-sensitive users to adjust their reverse charging habits and better play the role of V2G devices in the overall power grid.

[0024] Based on the same inventive concept, the present invention further provides a computer device, which includes: one or more processors, and a memory for storing one or more computer programs; the program includes program instructions, and the processor is configured to execute the program instructions stored in the memory. The processor may be a Central Processing Unit (CPU), or may also be other general-purpose processors, Digital Signal Processors (DSPs), Application Specific Integrated Circuits (ASICs), Field-Programmable Gate Arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. It is the computing core and control core of the terminal, and is used to implement one or more instructions. Specifically, it is used to load and execute one or more instructions in the computer storage medium to implement the above method.

[0025] It should be further noted that, based on the same inventive concept, the present invention further provides a computer storage medium, on which a computer program is stored, and the computer program, when run by a processor, executes the above method. The storage medium may be any combination of one or more computer-readable media. The computer-readable medium may be a computer-readable signal medium or a computer-readable storage medium. The computer-readable storage medium may, for example, but is not limited to, an electrical, magnetic, optical, electrical, magnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the above. More specific examples (non-exhaustive list) of the computer-readable storage medium include: an electrical connection having one or more wires, a portable computer disk, a hard disk, a Random Access Memory (RAM), a Read Only Memory (ROM), an Erasable Programmable Read Only Memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present invention, the computer-readable storage medium may be any tangible medium that contains or stores a program, and the program may be used by or in conjunction with an instruction execution system, apparatus, or device.

[0026] In the description of this specification, the description with reference to terms such as "one embodiment", "example", "specific example", etc. means that the specific features, structures, materials, or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the present disclosure. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described may be combined in any one or more embodiments or examples in a suitable manner.

[0027] The above has shown and described the basic principles, main features and advantages of the present disclosure. Those skilled in the art should understand that the present disclosure is not limited by the above embodiments. What is described in the above embodiments and the specification only illustrates the principles of the present disclosure. Without departing from the spirit and scope of the present disclosure, the present disclosure will have various changes and improvements, and these changes and improvements all fall within the scope of the present disclosure claimed.

[0028] The present invention is not limited to the above best implementation manners. Anyone can derive various other forms of methods for predicting and regulating the reverse charging behavior of energy storage vehicles applicable to V2G technology under the inspiration of the present invention. All equal changes and modifications made according to the scope of the patent application of the present invention shall fall within the scope covered by the present invention.

Claims

1. A method for predicting and regulating the reverse charging behavior of energy storage vehicles applicable to V2G technology, characterized in that: Based on the information of V2G equipment, time, date, and discharge SOC range, predict the batch reverse charging behavior of vehicles through probability statistics and deep learning, estimate the electric energy that can be obtained by different V2G equipment at different times, and conduct power dispatching according to the actual electricity demand and load conditions of each power grid node.

2. The method for predicting and regulating the reverse charging behavior of an energy storage vehicle applicable to V2G technology according to claim 1, characterized in that: The prediction of the batch reverse charging behavior of vehicles through probability statistics and deep learning based on the information of V2G equipment, time, date, and discharge SOC range is specifically as follows: Extract historical operation data from electric vehicles used as energy storage sources. After data cleaning and division of operation segments, obtain charging data segments and discharge data segments respectively; Based on vehicle speed, battery current, SOC change under parking state, parking location, and communication tags between the vehicle and V2G equipment from the discharge data segments, screen and obtain the reverse charging data segments of each vehicle for reverse charging to the V2G equipment. And extract the reverse charging geographical coordinates, start and end times, dates, and reverse charging start and end SOC parameters from the reverse charging data segments. Statistical parameters and determine the geographical coordinates of V2G equipment where each vehicle is accustomed to reverse charging, the start time and date of reverse charging start and end, the start and end SOC of reverse charging, calculate the probability of each parameter as the prior distribution; Taking the geographical coordinates where different V2G equipment is located as the condition, calculate the conditional probabilities of different reverse charging start and end times, dates, and start and end SOC respectively; Based on the assumption that the probability of a vehicle selecting different V2G facilities under specific reverse charging start and end times, dates, and start and end SOC conditions is independent of other conditions, combine the prior distribution to calculate the posterior distribution of selecting each V2G equipment under different conditions; Establish a prediction model of V2G equipment based on deep learning, using the data of each vehicle during peak electricity consumption hours, dates, and corresponding reverse charging start SOC as input, and the calculated posterior distribution of V2G equipment and SOC change amount as output, and use historical operation data to train the model until it converges stably.

3. The method for predicting and regulating the reverse charging behavior of an energy storage vehicle applicable to V2G technology according to claim 2, wherein: Based on the trained prediction model of V2G equipment, predict the V2G equipment and the available reverse charging capacity that are most likely to be selected currently according to the time, date, and SOC parameters collected from each vehicle, and determine the available reverse charging capacity of each V2G equipment according to the prediction results.

4. The method for predicting and regulating the reverse charging behavior of an energy storage vehicle applicable to V2G technology according to claim 3, characterized in that: Combined with the corresponding grid load at each V2G equipment during peak hours, the maximum storage capacity of the V2G equipment, and the available reverse charging capacity, set corresponding reverse charging electricity prices for different V2G equipment respectively, and perform corresponding power peak shaving dispatching.

5. The method for predicting and regulating the reverse charging behavior of an energy storage vehicle applicable to V2G technology according to claim 2, characterized in that: The method for screening and obtaining the reverse charging data segments of each vehicle for reverse charging to the V2G equipment is specifically as follows: By traversing the segments that simultaneously meet the conditions of vehicle speed being 0, battery current being positive, SOC decreasing by more than a certain value, and communication occurring between the vehicle and V2G equipment, and then matching the geographical coordinates of these segments during parking discharge with the coordinates of each V2G equipment in the digital map within a certain range, so as to screen out the reverse charging data segments from the discharge segments.

6. The method for predicting and regulating the reverse charging behavior of an energy storage vehicle applicable to V2G technology according to claim 2, characterized in that: The specific process of calculating the prior distribution of each parameter includes: After matching the parking location geographical coordinates of the reverse charging data segment with the geographical coordinates of each V2G device in the digital map according to the distance between them, add the corresponding V2G device label Ln to each reverse charging data segment, indicating the coordinates L of the nth V2G device; then count and sort the number of V2G device markings of each reverse charging data segment to obtain the V2G devices that each vehicle is accustomed to using and the corresponding probability P(Ln), where n represents the nth V2G device; extract the start and end times of the reverse charging in the reverse charging data segment, and divide each day into several time periods at equal time intervals and add the time period label hm, where m represents the mth time period in a day; then count and sort the time period labels of the start and end of the reverse charging each day to obtain the start and end time periods of the reverse charging that each vehicle is accustomed to and the corresponding probability P(hm); extract or convert the date of the reverse charging data segment to obtain the corresponding weekly calendar date and add the weekly calendar label wi, where i = 1 to 7; then count and sort the weekly calendar labels to obtain the weekly calendar dates of the reverse charging that are accustomed to and the corresponding probability P(wi); divide the battery SOC from 100% to 0% into several SOC intervals at equal intervals and add the corresponding SOC interval label SOCi for the SOC change of each reverse charging data segment, where i represents the ith SOC interval; then count and sort each SOC interval label to obtain the SOC intervals consumed by each vehicle's accustomed reverse charging and the corresponding probability P(SOCi).

7. The method for predicting and regulating the reverse charging behavior of an energy storage vehicle applicable to V2G technology according to claim 2, characterized in that: In the process of the posterior calculation: First, calculate the conditional probabilities of the geographical coordinates of the V2G device for different reverse charging start and end times, dates, and start and end SOCs respectively: P(hm | Ln), P(wi | Ln), P(SOCi | Ln); For a combination Ω = [hm, wi, SOCi] of specific reverse charging start and end times, dates, and start and end SOC conditions, calculate the prior probability: P(Ω) = P(hm) × P(wi) × P(SOCi) And the conditional probability: P(Ω | Ln) = P(hm | Ln) × P(wi | Ln) × P(SOCi | Ln); Use Bayes' formula to calculate the posterior distribution of selecting each V2G device under different conditions: P(Ln | Ω) = P(Ln) × P(Ω | Ln) / P(Ω).

8. The method for predicting and regulating the reverse charging behavior of an energy storage vehicle applicable to V2G technology according to claim 2, characterized in that: In the training process of the V2G device prediction model, extract the vehicle VIN, time and date labels, start and end SOC labels, V2G device labels, and the calculated posterior distribution in the reverse charging data segment during peak electricity consumption periods to construct the corresponding training set and validation set, and train and verify the training effect of the model; at each specified time interval, update the V2G device prediction model by re-collecting the historical operation data of the energy storage vehicle and re-executing the data processing and model training process.

9. An electronic device, comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the steps of the method for predicting and regulating the reverse charging behavior of an energy storage vehicle applicable to V2G technology according to any one of claims 1-8.

10. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the method for predicting and regulating the reverse charging behavior of an energy storage vehicle applicable to V2G technology according to any one of claims 1-8.

Images