Electricity price model establishment method and system based on electric vehicle data analysis

Abstract

The invention discloses an electricity price model establishment method and system based on electric vehicle data analysis, and mainly relates to the technical field of application of an electricity price model in electric vehicle data analysis. Comprising the following steps: collecting charging time, frequency and charging amount information of different vehicle owners on an electric vehicle, and combining the collected information with a traffic jam condition to construct an electric vehicle charging demand model; establishing a traffic congestion degree electricity price mechanism in a transformer area and determining a traffic congestion degree coefficient through the charging load capacity obtained by the electric vehicle charging demand model; sS-AE is used to carry out threshold adaptive calculation of the traffic congestion degree coefficient; and setting a vehicle owner individual charging willingness feedback mechanism, determining a charging / discharging willingness index, and establishing a dynamic electricity price model. The method has the advantages that traffic congestion can be relieved, power grid load distribution is optimized, and the overall system efficiency is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of application of electricity price models in electric vehicle data analysis, and in particular to a method and system for establishing an electricity price model based on electric vehicle data analysis. Background Art

[0002] With the continuous advancement of urbanization and the rapid increase in the number of vehicles, traffic congestion is becoming increasingly serious. At the same time, the penetration rate of electric vehicles is also increasing year by year. Although electric vehicles have significant advantages in reducing carbon emissions and reducing energy consumption, their charging needs also bring new challenges to the power grid. Therefore, how to effectively manage electric vehicle charging behavior to alleviate traffic congestion and power grid pressure has become one of the current urgent issues that need to be solved.

[0003] At present, there are many technologies and methods trying to solve the problem of electric vehicle charging and traffic congestion, such as time-of-use electricity prices, smart grid charging / discharging load management and other technical methods. Although the above technologies have alleviated the problem of electric vehicle charging and traffic congestion to a certain extent, there are still the following shortcomings:

[0004] Most existing electricity price models lack a dynamic adjustment mechanism and are unable to respond in a timely manner to changes in car owners' charging needs and traffic conditions. In addition, existing solutions usually do not fully consider the personal charging willingness of electric car owners, resulting in poor user experience and difficulty in achieving optimal scheduling effects. There is a lack of effective integration between traffic management and power management systems, which makes it difficult for the two to work together and fail to fully realize the overall benefits.

[0005] Therefore, a dynamic charging / discharging electricity price model that comprehensively considers traffic congestion and the personal charging willingness of electric vehicle owners is urgently needed to solve the above problems. Summary of the invention

[0006] The purpose of the present invention is to provide a method and system for establishing an electricity price model based on electric vehicle data analysis, which can optimize power grid load distribution and improve overall system efficiency while alleviating traffic congestion.

[0007] In order to achieve the above object, the present invention is implemented through the following technical solutions:

[0008] On the one hand, a method for establishing an electricity price model based on electric vehicle data analysis is provided, comprising the following steps:

[0009] S1: Collect the charging time, frequency and charging amount information of electric vehicles from different car owners, and combine the collected information with traffic congestion to build an electric vehicle charging demand model;

[0010] S2: Using the charging load obtained by the electric vehicle charging demand model in step S1, establish a traffic congestion electricity price mechanism within the substation and determine the traffic congestion coefficient;

[0011] S3: Use SS-AE to perform adaptive threshold calculation of traffic congestion coefficient;

[0012] S4: Set up a feedback mechanism for the owner’s personal charging willingness, determine the charging / discharging willingness index, and establish a dynamic electricity price model.

[0013] Preferably, in step S1, constructing an electric vehicle charging demand model includes:

[0014] The power consumption behavior is described by the probability distribution model of the initial charging period and the probability model of the daily mileage:

[0015]

[0016] Among them, t s Represents the end of charging time, μ s represents the average end-of-charge time, σ s Represents the standard deviation of the end charging time;

[0017] Mileage:

[0018]

[0019] Among them, l represents the mileage, σ l Represents the standard deviation of mileage, μ l Represents the average mileage;

[0020] The charging time calculation formula is:

[0021]

[0022] Among them, T c represents the charging time, e represents the electric energy consumed by the electric vehicle for every 100 kilometers traveled, and P c represents the charging power of electric vehicles, and ηc represents the charging efficiency of electric vehicles;

[0023] Sampling electric vehicles and calculating the daily basic charging load of electric vehicles:

[0024] W c =P c T c

[0025] Among them, W c is the charging load.

[0026] Preferably, in the step S2, establishing a traffic congestion degree electricity price mechanism in the power distribution area and determining a traffic congestion degree coefficient includes:

[0027] The traffic congestion degree in the power distribution area is defined as:

[0028]

[0029] Among them, V i，t represents the average vehicle speed of the i-th power distribution area at time t, and V i，max represents the routine vehicle speed of the i-th power distribution area, and ρ i，t represents the traffic flow density of the i-th power distribution area at time t, and ρ i，max represents the maximum traffic flow density of the i-th power distribution area;

[0030] The traffic congestion coefficient is denoted as: TPF,

[0031]

[0032] When TPF≥M, it means that the traffic in the power distribution area is not congested. When TPF<M, it means that the traffic in the power distribution area is congested, where M is the threshold of the traffic congestion coefficient.

[0033] Preferably, the SS-AE includes two AEs, each AE includes an encoder and a decoder, and both the encoder and the decoder are composed of three dense layers. The encoder part includes:

[0034] δ and δ′ respectively represent the historical traffic congestion coefficient and the real-time traffic congestion coefficient, respectively represent the reconstruction of the historical traffic congestion coefficient and the real-time traffic congestion coefficient through the SS-AE, r, are the latent representations of δ and δ′ respectively, r, is expressed as:

[0035] r = μ ε (δ)+σ ε (δ)⊙η

[0036]

[0037] The posterior probability q ε (r|δ, 1):

[0038] q ε (r|δ, 1)~N(r|μ ε (δ), σ 2 (δ))

[0039] The posterior probability of the low-confidence sample is expressed as:

[0040]

[0041] The decoder part includes:

[0042] The decoder in SS-AE is viewed as a conditional probability distribution model express Reconstruction of δ based on r:

[0043]

[0044] The loss function of conversion reconstruction is expressed as:

[0045]

[0046] By maximizing Optimize the decoder parameters, D KL (q ε (r|δ, 1)||p φ (r)) indicates q ε (r|δ, 1) and p φ (r) KL divergence between;

[0047] The reconstruction error between δ and δ′ is ||δ-δ′|| 2 , The reconstruction error between

[0048] Solve the threshold M of the traffic congestion coefficient:

[0049]

[0050] Preferably, in step S4, the charge / discharge willingness index is specifically: the vehicle owner's response charge / discharge willingness index, expressed as:

[0051] W j，t =W j，or +SoC j，t

[0052] Among them, W j，t represents the response charge / discharge willingness index of the jth electric vehicle, W j，or represents the charging / discharging willingness of the owner of the jth electric vehicle;

[0053] The establishment of a dynamic electricity price model includes: establishing a charging price model and establishing a discharging price model. The establishment of a charging price model includes:

[0054]

[0055] in, is the charging price of the jth electric vehicle at time t, C t is the basic charging electricity price in the area at time t, is the penalty price due to charging congestion in the area during period t, SoC j，t The SiC at the initial time t of charging the jth electric vehicle, The penalty price generated by whether the j-th electric vehicle responds to the dispatch;

[0056] Establish a discharge price model, including:

[0057]

[0058] in, is the discharge price of the jth electric vehicle at time t, D t is the basic discharge price in the area at time t, is the penalty price caused by charging congestion in the area during period t, The penalty price generated by whether the j-th electric vehicle responds to the dispatch.

[0059] On the other hand, a system based on the above-mentioned method for establishing an electricity price model based on electric vehicle data analysis is provided, comprising:

[0060] The data collection and electric vehicle charging demand model building module is used to: collect the charging time, frequency and charging amount information of different car owners for electric vehicles, and combine the collected information with traffic congestion conditions to build an electric vehicle charging demand model;

[0061] The threshold calculation module of the traffic congestion coefficient is used to: establish the traffic congestion electricity price mechanism in the substation area and determine the traffic congestion coefficient based on the charging load obtained by the electric vehicle charging demand model; use SS-AE to perform adaptive threshold calculation of the traffic congestion coefficient;

[0062] The dynamic electricity price model building module is used to: set up the owner's personal charging willingness feedback mechanism, determine the charging / discharging willingness index, and establish a dynamic electricity price model.

[0063] Compared with the prior art, the beneficial effects of the present invention are:

[0064] 1. Realize peak load shaving and valley filling of power grid load to relieve power grid pressure: The dynamic electricity price model adjusts electricity prices to guide car owners to charge when the power grid load is low, thereby balancing the load of the power grid and avoiding the collapse of the power system;

[0065] 2. Optimize traffic congestion: This model takes traffic congestion into consideration and encourages car owners to charge during off-peak hours through dynamic electricity prices, thereby reducing traffic pressure during peak hours. For example, during peak traffic hours, the charging fee is higher, so car owners are more inclined to charge during off-peak hours, thereby reducing travel during peak hours.

[0066] 3. Reduce the charging cost of car owners: Through dynamic electricity prices, car owners can choose to charge during periods with lower electricity prices, thereby reducing charging costs. Especially for private car owners, the reduction in charging costs can significantly increase their enthusiasm and economic benefits for using electric vehicles, and even increase the market acceptance of electric vehicles, which is of great significance for promoting the popularization and development of electric vehicles. BRIEF DESCRIPTION OF THE DRAWINGS

[0067] Figure 1 It is a front view of the method flow of the present invention;

[0068] Figure 2 It is a schematic diagram of the system structure of the present invention. DETAILED DESCRIPTION

[0069] The present invention will be further described below in conjunction with specific embodiments. It should be understood that these embodiments are only used to illustrate the present invention and are not intended to limit the scope of the present invention. In addition, it should be understood that after reading the content taught by the present invention, those skilled in the art can make various changes or modifications to the present invention, and these equivalent forms fall within the scope limited by the application equally.

[0070] In the present invention, terms such as "upper", "lower", "left", "right", "front", "back", "vertical", "horizontal", "side", "bottom", etc. indicate directions or positional relationships based on the directions or positional relationships shown in the accompanying drawings. They are relational words determined only for the convenience of describing the structural relationships of the various parts or elements of the present invention, and do not specifically refer to any part or element in the present invention and should not be understood as limitations on the present invention.

[0071] In the present invention, terms such as "fixed connection", "connected", "connection", etc. should be understood in a broad sense, indicating that it can be fixedly connected, integrally connected or detachably connected; it can be directly connected or indirectly connected through an intermediate medium. Relevant scientific research or technical personnel in this field can determine the specific meanings of the above terms in the present invention according to specific circumstances, and they should not be understood as limiting the present invention.

[0072] Example:

[0073] like Figure 1 As shown, this embodiment provides a method for establishing an electricity price model based on electric vehicle data analysis, comprising the following steps:

[0074] S1: Collect the charging time, frequency and charging amount information of electric vehicles from different car owners, and combine the collected information with traffic congestion to build an electric vehicle charging demand model;

[0075] S2: Using the charging load obtained by the electric vehicle charging demand model in step S1, establish a traffic congestion electricity price mechanism within the substation and determine the traffic congestion coefficient;

[0076] S3: Use SS-AE to perform adaptive threshold calculation of traffic congestion coefficient;

[0077] S4: Set up a feedback mechanism for the owner’s personal charging willingness, determine the charging / discharging willingness index, and establish a dynamic electricity price model.

[0078] In step S1, an electric vehicle charging demand model is constructed, including:

[0079] The charging load of electric vehicles usually depends on the owner's car usage and charging habits, and lacks obvious rules. Therefore, this embodiment describes its power consumption behavior through the probability distribution model of the initial charging time period and the probability model of the daily mileage:

[0080]

[0081] Among them, t s Represents the end of charging time, μ s represents the average end-of-charge time, σ s Represents the standard deviation of the end charging time,

[0082] The mileage is as follows:

[0083]

[0084] Among them, l represents the mileage, σ l Represents the standard deviation of mileage, μ l Represents the average mileage, and the charging time calculation formula for private cars is as follows:

[0085]

[0086] Among them, T c represents the charging time, e represents the electric energy consumed by the electric vehicle for every 100 kilometers traveled, and P c represents the charging power of electric vehicles, η c Represents the charging efficiency of electric vehicles,

[0087] The Monte Carlo method is used to sample private cars to determine the starting charging time and charging duration, thereby calculating the daily basic charging load of private cars:

[0088] W c =P c T c

[0089] Among them, W c is the charging load;

[0090] Since taxis and buses have long operating hours and large mileage, the electricity consumption patterns of these two types of electric vehicles are relatively regular. Assuming that they need to be charged twice a day, the SoC and time of charging start are normally distributed, as shown in Table 1:

[0091] Table 1. Charging start time and SoC distribution of taxis and buses

[0092]

[0093]

[0094] t1, t2, t3, and t4 represent the expected charging start and end times of taxis and buses, respectively; σ t1 , σ t2 , σ t3 , σ t4 represents the variance of the charging start and end time of taxis and buses; SoC1 and SoC2 represent the expected SoC at the charging start and end time of taxis and buses, respectively; denotes the variance of SoC at the start and end of charging for taxis and buses, respectively.

[0095] Charging time for taxis and buses:

[0096]

[0097] SoC e is the remaining power percentage of the two types of electric vehicles, C e is the battery capacity, η e is the charging efficiency P of the two types of electric vehicles e is the charging power of two types of electric vehicles;

[0098] A charging load model for three types of electric vehicles, namely private cars, taxis and buses, was established for the subsequent dynamic electricity price analysis.

[0099] The charging demand of electric vehicles has a significant impact on the grid load, especially during peak hours. By formulating a dynamic electricity price strategy, car owners can be guided to charge during periods of low grid load, thereby balancing the grid load and reducing grid pressure. At the same time, a reasonable electricity price strategy can reduce the charging cost of car owners, and flexibly choose when and where to charge based on charging demand and car owners' willingness to charge.

[0100] Due to the strong mobility of various electric vehicles and the lack of fixed charging / discharging areas, it is necessary to set congestion-based electricity prices for each charging / discharging area. By real-time monitoring the congestion level of charging piles in each area, the charging / discharging electricity prices are dynamically adjusted to adapt to the free and dispersed charging behaviors of users in different areas. When the traffic condition in a certain area becomes congested, the charging electricity price will increase while the discharging compensation electricity price will decrease to encourage users to choose other areas for charging or discharging, thus alleviating the traffic congestion problem in that area. Through this congestion-based electricity price mechanism, the charging and discharging resources can be optimized and the charging and discharging efficiency of the distribution network can be improved. Therefore, steps S2 - S3 include:

[0101] 1. Define the traffic congestion level in the area:

[0102]

[0103] Among them, V i，t represents the average vehicle speed of the i-th area at time t, V i，max represents the routine vehicle speed of the i-th area, ρ i，t represents the traffic flow density of the i-th area at time t, ρ i，max represents the maximum traffic flow density of the i-th area. To quantitatively analyze the traffic congestion conditions in different areas, a Traffic Performance Factor (TPF) is introduced for digital analysis of traffic congestion conditions:

[0104]

[0105] When TPF ≥ M, it indicates that the traffic in the area is not congested; when TPF < M, it indicates that the traffic in the area is congested, where M is the threshold of the traffic congestion coefficient.

[0106] 2. Adaptively solve the threshold M of the traffic congestion coefficient:

[0107] Between the charging demand of electric vehicles and the power grid load management, optimizing the charging resources and improving the efficiency of the distribution network are the keys. To achieve this goal, this embodiment proposes a method of using SS - AE for adaptive threshold calculation. SS - AE adaptively determines the traffic congestion threshold and dynamically adjusts the electricity price strategy, thereby optimizing the charging / discharging behavior.

[0108] Specifically, SS - AE consists of two Autoencoders (AE). Each AE contains an encoder and a decoder. Both the encoder and the decoder are composed of three dense layers. The two encoders share weights and have the same structure, and the two decoders also have the same structure. The core of SS - AE lies in its ability to adaptively determine the threshold.

[0109] When applied to the management of electric vehicle charging stations, SS-AE uses real-time monitoring data to learn and identify charging behavior patterns within the station. SS-AE will dynamically adjust the charging price and discharge compensation price of the station according to its adaptively calculated threshold. In this way, when a station becomes crowded, the charging price will increase, while the discharge compensation price will decrease, which will encourage users to choose other stations for charging and discharging, thereby effectively alleviating traffic congestion problems.

[0110] Among them, the encoder part:

[0111] δ and δ′ represent the historical traffic congestion coefficient and the real-time traffic congestion coefficient respectively. They represent the historical traffic congestion coefficient and the real-time traffic congestion coefficient reconstructed by SS-AE, r, are the potential representations of δ and δ′, r, It is expressed as:

[0112] r=μ ε (δ)+σ ε (δ)⊙η

[0113]

[0114] The posterior probability q ε (r|δ, 1):

[0115] q ε (r|δ,1)~N(r|μ ε (δ), σ 2 (δ)

[0116] The posterior probability of a low confidence sample is expressed as:

[0117]

[0118] Decoder part:

[0119] The decoder in SS-AE is viewed as a conditional probability distribution model express Reconstruction of δ based on r:

[0120]

[0121] The loss function of conversion reconstruction is expressed as:

[0122]

[0123] By maximizing Optimize the decoder parameters, D KL (q ε (r|δ, 1)||p φ (r)) indicates qε (r|δ, 1) and p φ (r) KL divergence between;

[0124] The reconstruction error between δ and δ′ is ||δ-δ′|| 2 , The reconstruction error between

[0125] Solve the threshold M of the traffic congestion coefficient:

[0126]

[0127] This embodiment uses the above formula to balance the historical traffic data and the real-time traffic data, and solves the threshold M of the traffic congestion coefficient based on the historical data and the real-time data.

[0128] Define the car owner's willingness index for charging / discharging:

[0129] W j，t =W j，or +SoC j，t

[0130] Among them, W j，t represents the response charge / discharge willingness index of the jth electric vehicle, W j，or represents the charging / discharging willingness of the owner of the jth electric vehicle;

[0131] The owner's willingness to charge / discharge is affected by multiple factors, including battery power, remaining driving distance of the vehicle, and road traffic conditions. Specifically, when road traffic is smooth, the owner will feel more flexible and the vehicle's movement within and between regions is unrestricted. Therefore, the owner's willingness to charge is relatively high, and the willingness to discharge may also increase, especially when the system needs electricity. At this time, electric vehicles, as a mobilizable resource for regulating energy demand, can respond quickly and participate in charging / discharging scheduling;

[0132] In the case of road congestion, due to the reduction in vehicle speed and increased traffic density, the flexibility of the vehicle is reduced, and the owner will tend to reserve power to ensure that the vehicle can successfully complete the travel task, and the charging / discharging willingness index will decrease accordingly. This is because the owner is worried that he will not be able to reach the charging station in time or affect his travel due to insufficient power or road congestion during driving. Therefore, the charging demand and discharging willingness may be significantly weakened in traffic congestion, and even completely unwilling to participate in charging / discharging scheduling;

[0133] From the above situation, it can be seen that the owner's charging and discharging willingness index is not only affected by the vehicle's power level, but also by traffic conditions and travel needs, which in turn affects the responsiveness of electric vehicles as energy regulation resources;

[0134] When the road is less congested, the traffic density in each area of ​​the city is low, the speed is fast, and the movement of vehicles within and between areas is very smooth, which makes electric vehicles more flexible as a movable resource, and car owners are willing to charge / discharge in such areas; when the road is seriously congested, the traffic density in each area increases, the speed decreases, and the movement and transfer of vehicles are restricted, which reduces the flexibility of electric vehicles as a movable resource and makes scheduling difficult;

[0135] Therefore, the dynamic charging and discharging price model formulated in this embodiment is as described in step S4, including:

[0136] 1. Establish a charging price model:

[0137]

[0138] in, is the charging price of the jth electric vehicle at time t, C t is the basic charging electricity price in the area at time t, is the penalty price due to charging congestion in the area during period t, SoC j，t The SoC at the initial time t of charging the j-th electric vehicle, The penalty price generated by whether the j-th electric vehicle responds to the dispatch;

[0139] 2. Establish a discharge price model:

[0140]

[0141] in, is the discharge price of the jth electric vehicle at time t, D t is the basic discharge electricity price in the area at time t, is the penalty price caused by charging congestion in the area during period t, The penalty price generated by whether the j-th electric vehicle responds to the dispatch;

[0142] From the above model, it can be seen that when electric vehicle owners have a higher willingness to respond to scheduling, the charging price is lower; when the congestion in the electric vehicle area is lower, the charging price is lower. Similarly, when electric vehicle owners have a higher willingness to respond to scheduling, the owners will gain greater benefits from responding to the scheduling policy. When the charging congestion in the electric vehicle area is higher, the discharge price is higher, and the corresponding distributed power supply is more needed in the area.

[0143] like Figure 2 As shown, this embodiment also provides a system based on the above-mentioned method for establishing an electricity price model based on electric vehicle data analysis, including:

[0144] The data collection and electric vehicle charging demand model building module is used to: collect the charging time, frequency and charging amount information of different car owners for electric vehicles, and combine the collected information with traffic congestion conditions to build an electric vehicle charging demand model;

[0145] The threshold calculation module of the traffic congestion coefficient is used to: establish the traffic congestion electricity price mechanism in the substation area and determine the traffic congestion coefficient based on the charging load obtained by the electric vehicle charging demand model; use SS-AE to perform adaptive threshold calculation of the traffic congestion coefficient;

[0146] The dynamic electricity price model building module is used to: set up the owner's personal charging willingness feedback mechanism, determine the charging / discharging willingness index, and establish a dynamic electricity price model.

[0147] The above is a specific description of the preferred implementation of the present invention, but the present invention is not limited to the described embodiments. Those skilled in the art may make various equivalent modifications or substitutions without violating the spirit of the present invention. These equivalent modifications or substitutions are all included in the scope defined by the claims of this application.

Claims

1. A method for establishing an electricity price model based on electric vehicle data analysis, characterized in that: Including the following steps: S1: Collect the charging time, frequency, and charging amount information of electric vehicles for different vehicle owners, and combine the collected information with the traffic congestion situation to construct an electric vehicle charging demand model; S2: Based on the charging load obtained from the electric vehicle charging demand model in step S1, establish a traffic congestion degree electricity price mechanism within the substation area and determine the traffic congestion degree coefficient; S3: Use SS-AE to perform adaptive calculation of the threshold of the traffic congestion degree coefficient; S4: Set up a feedback mechanism for the personal charging willingness of vehicle owners, determine the charge / discharge willingness index, and establish a dynamic electricity price model.

2. The method for establishing an electricity price model based on electric vehicle data analysis according to claim 1, characterized in that: In step S1, constructing the electric vehicle charging demand model includes: Describing its electricity consumption behavior through the probability distribution model of the initial charging time period and the probability model of the daily driving mileage: Among them, t s Represents the end of charging time, μ s represents the average end-of-charge time, σ S Represents the standard deviation of the end-of-charge time.

3. The method for establishing an electricity price model based on electric vehicle data analysis according to claim 2 is characterized in that: The driving mileage is: Among them, l represents the mileage, σ l Represents the standard deviation of mileage, μ l Represents the average mileage.

4. The method for establishing an electricity price model based on electric vehicle data analysis according to claim 3 is characterized in that: The charging duration calculation formula is: Among them, T c represents the charging time, e represents the electric energy consumed by the electric vehicle for every 100 kilometers traveled, and P c represents the charging power of electric vehicles, η c Represents the charging efficiency of electric vehicles; Sample electric vehicles and calculate the daily basic charging load of electric vehicles: W c =P c T c Among them, W c is the charging load.

5. The method for establishing an electricity price model based on electric vehicle data analysis according to claim 4 is characterized in that: In step S2, establishing the traffic congestion degree electricity price mechanism within the substation area and determining the traffic congestion degree coefficient includes: The traffic congestion degree within the substation area is defined as: Among them, V i , t represents the average vehicle speed of the ith station at time t, V i , max represents the routine vehicle speed of the ith station, ρ i , t represents the traffic density of the ith station at time t, ρ i , max Represents the maximum traffic density of the ith area.

6. A method for establishing an electricity price model based on electric vehicle data analysis according to claim 5, characterized in that: The traffic congestion coefficient is denoted as: TPF, When TPF≥M, it indicates that the traffic within the substation area is not congested; when TPF<M, it indicates that the traffic within the substation area is congested, where M is the threshold of the traffic congestion coefficient.

7. A method for establishing an electricity price model based on electric vehicle data analysis according to claim 6, characterized in that: The SS-AE includes two AEs, each AE includes an encoder and a decoder, and both the encoder and the decoder are composed of three dense layers. The encoder part includes: δ and δ′ represent the historical traffic congestion coefficient and the real-time traffic congestion coefficient respectively. They represent the historical traffic congestion coefficient and the real-time traffic congestion coefficient reconstructed by SS-AE, r, are the potential representations of δ and δ′, r, It is expressed as: r=μ ε (d)+s ε (d)⊙h The posterior probability q ε (r|δ, 1): q ε (r|δ,1)~N(r|μ ε (d), s 2 (d)) The posterior probability of low-confidence samples is expressed as:

8. A method for establishing an electricity price model based on electric vehicle data analysis according to claim 7, characterized in that: The decoder part includes: The decoder in SS-AE is viewed as a conditional probability distribution model express Reconstruction of δ based on r: The loss function of the conversion reconstruction is expressed as: By maximizing Optimize the decoder parameters, D KL (q ε (r|δ, 1)||p φ (r)) indicates q ε (r|δ, 1) and p φ (r) KL divergence between; The reconstruction error between δ and δ′ is ‖δ-δ′‖ 2 , The reconstruction error between Solve for the threshold M of the traffic congestion coefficient:

9. A method for establishing an electricity price model based on electric vehicle data analysis according to claim 8, characterized in that: In step S4, the charge / discharge willingness index is specifically: the vehicle owner's response to the charge / discharge willingness index, expressed as: W j，t =W j，or +SoC j，t Among them, W j，t represents the response charge / discharge willingness index of the jth electric vehicle, W j，or represents the charging / discharging willingness of the owner of the jth electric vehicle; The establishment of the dynamic electricity price model includes: establishing a charging price model and establishing a discharging price model. Establishing the charging price model includes: in, is the charging price of the jth electric vehicle at time t, C t is the basic charging electricity price in the area at time t, is the penalty price due to charging congestion in the area during period t, SoC j，t The SoC at the initial time t of charging the j-th electric vehicle, The penalty price generated by whether the j-th electric vehicle responds to the dispatch; Establishing the discharging price model includes: in, is the discharge price of the jth electric vehicle at time t, D t is the basic discharge price in the area at time t, is the penalty price caused by charging congestion in the area during period t, The penalty price generated by whether the j-th electric vehicle responds to the dispatch.

10. A system based on the method for establishing an electricity price model based on electric vehicle data analysis as claimed in claim 1, characterized in that: Including: A data collection and electric vehicle charging demand model construction module, which is used to: collect the charging time, frequency, and charging amount information of electric vehicles for different vehicle owners, and combine the collected information with the traffic congestion situation to construct an electric vehicle charging demand model; A traffic congestion degree coefficient threshold calculation module, which is used to: based on the charging load obtained from the electric vehicle charging demand model, establish a traffic congestion degree electricity price mechanism within the substation area and determine the traffic congestion degree coefficient; use SS-AE to perform adaptive calculation of the threshold of the traffic congestion degree coefficient; A dynamic electricity price model construction module, which is used to: set up a feedback mechanism for the personal charging willingness of vehicle owners, determine the charge / discharge willingness index, and establish a dynamic electricity price model.

Images