Electric vehicle charging service fee optimization method fusing user psychological effect

Abstract

The invention relates to the technical field of electric vehicle charging service and energy optimization, in particular to an electric vehicle charging service fee optimization method fusing user psychological effects. According to the technical scheme, the method comprises the following steps: establishing a historical data model of electric vehicle charging characteristics, and obtaining regional resident load and a light-storage model; and generating an initial electric vehicle user behavior matrix based on the data, and constructing an electric vehicle cluster load curve. According to the method, a charging behavior response mechanism is constructed by quantifying the psychological effect of the user, so that more accurate guidance for selection of the charging time period of the user is realized; a multi-target collaborative optimization model considering the power grid load, the user cost, the photovoltaic consumption and the energy storage income is established; the adaptability and robustness of the strategy are enhanced by adopting a closed-loop iteration mechanism; and a decision basis for quantitative evaluation is provided for a new energy aggregator, and the problems of inaccurate behavior description, single target and insufficient practicability in a traditional optimization method are systematically solved.

Description

Technical Field

[0001] This invention relates to the field of electric vehicle charging services and energy optimization technology, and in particular to a method for optimizing electric vehicle charging service fees by incorporating user psychological effects. Background Technology

[0002] With the rapid growth of electric vehicle (EV) ownership, large-scale EV grid connection for charging has become a key factor affecting the safety, stability, and economic operation of the power distribution network. Against this backdrop, renewable energy aggregators, acting as intermediaries connecting EV users, grid operators, and distributed photovoltaic and energy storage systems, directly influence users' charging service pricing strategies. These strategies also affect the load distribution of the power distribution network, the efficiency of renewable energy absorption, and the arbitrage potential of energy storage systems.

[0003] Currently, optimization methods for new energy aggregator service fees mainly focus on mathematical modeling and solving for single objectives, such as smoothing distribution network load fluctuations and reducing peak-valley differences, or simply optimizing pricing to increase aggregator revenue. While these methods can theoretically achieve certain optimization results, they often face problems such as low user responsiveness and less-than-expected strategy implementation in practical applications. The reason for this is that existing models generally treat users as completely rational decision-makers, responding solely based on electricity price signals, while ignoring users' psychological perceptions and behavioral characteristics in real-world scenarios.

[0004] In reality, electric vehicle users' charging decisions are a psychological and behavioral process influenced by multiple factors. Factors such as the urgency of their charging needs (e.g., low remaining battery capacity, subsequent travel plans), their psychological sensitivity to price changes, and their subjective perception of the convenience of charging times all significantly affect their final choice of charging time and location. Traditional optimization models, failing to incorporate these psychological and behavioral characteristics, result in optimized service fee strategies that are ineffective in guiding users to shift load, leading to secondary problems such as peak-hour congestion, solar curtailment, and insufficient utilization of energy storage.

[0005] The Weber-Fechner law, a classic law in psychophysics, reveals a logarithmic relationship between the intensity of subjective perception of physical stimuli (such as light, sound, and price) and the intensity of the objective stimulus. This law provides a theoretical framework for quantifying users' psychological perception of external stimuli such as price differences and the urgency of demand. Introducing it into the field of electric vehicle charging service fee optimization is expected to more accurately characterize and predict user behavioral responses, thereby designing service fee strategies that both meet grid optimization goals and are widely acceptable to users. Therefore, this application proposes a method for optimizing electric vehicle charging service fees that incorporates user psychological effects. Summary of the Invention

[0006] The purpose of this invention is to address the problem that existing methods for optimizing service fees of new energy aggregators ignore the psychological perception and behavioral characteristics of users in real-world scenarios, and to propose a method for optimizing electric vehicle charging service fees that incorporates user psychological effects.

[0007] In a first aspect, the present invention provides a method for optimizing electric vehicle charging service fees by incorporating user psychological effects, comprising the following steps:

[0008] S1. Establish a historical data model of electric vehicle charging characteristics and obtain regional residential load and photovoltaic-storage model;

[0009] S2. Based on the data from step S1, generate an initial electric vehicle user behavior matrix and construct an electric vehicle cluster load curve;

[0010] S3. Define the multi-objective model and constraint model for regional service fee optimization, and use the NSGA-II algorithm to solve for the current optimal aggregator service fee sequence.

[0011] S4. Based on the optimal aggregator service fee sequence, calculate the transition probability of each user in each time slot according to the Weber-Fechner effect, and generate a transition probability matrix;

[0012] S5. Update the user behavior matrix based on the transition probability matrix and compare it with the behavior matrix of the previous iteration. If the two are consistent, the model is determined to have converged and proceed to the next step; otherwise, return to step S2 to continue iterating.

[0013] S6. Based on the converged user behavior matrix, generate the optimized electric vehicle cluster load curve.

[0014] Optionally, step S1 specifically includes:

[0015] S1.1 Obtaining data under disordered charging conditions Charging characteristic sequence of electric vehicles ,in, , Indicates the first Charging characteristic sequence of electric vehicles Indicates the first The time when an electric vehicle begins charging. Indicates the first The battery level of an electric vehicle when it begins charging. Indicates the first The time it takes for an electric vehicle to finish charging;

[0016] S1.2 Obtaining the regional feature sequence ,in, This indicates the residential electricity load sequence for the area. Represents the residential load sequence. This represents the photovoltaic power output sequence.

[0017] Optionally, step S2 specifically includes:

[0018] S2.1 Initialize the user behavior matrix Matrix elements Indicates the first electric vehicles in the first Charging status during a given period;

[0019] S2.2, Based on behavior matrix and charging power of a single electric vehicle Calculate the initial cluster load ;

[0020] S2.3 Calculate the initial total charging cost for users. .

[0021] Optionally, step S3 specifically includes:

[0022] S3.1 Define the optimization variable vector This includes service fees for each time period and energy storage charging and discharging power;

[0023] S3.2 Constructing a multi-objective optimization function ,in, Due to poor load on the distribution network, Total cost for electric vehicle users This refers to the amount of solar power curtailment. For energy storage arbitrage profits, This represents the transpose of a matrix.

[0024] S3.3, Set constraints on energy storage charging and discharging power and SOC evolution;

[0025] S3.4 Solve using the NSGA-II algorithm and output the optimal service fee sequence. .

[0026] Optionally, step S4 includes:

[0027] S4.1, Calculate the first... Actual price paid by users during the time period , This indicates that the power grid is in The price of electricity at that time;

[0028] S4.2 Calculate the probability of necessity for user switching based on the Weber-Fechner effect. ;

[0029] S4.3 Calculating suitability probability based on price difference ;

[0030] S4.4, Weighted fusion yields the overall transition probability. Construct the transition probability matrix.

[0031] Optionally, the probability of necessity The calculation formula is:

[0032]

[0033] in, , This is the difference between the current time period and the original charging time period. , For necessity probability parameters; For the first Initial SOC of an electric vehicle; The time difference between the initial start time of the connection, in hours; , This represents the normalized range of the necessity index.

[0034] Optionally, the calculation of the suitability probability includes three user psychological states: normal state, positive state, and anxious state, respectively corresponding to... , , The final suitability probability is its mean.

[0035] Optionally, step S5 specifically includes:

[0036] S5.1, Based on the transition probability threshold Update user behavior matrix ;

[0037] S5.2 Calculate the updated cluster load ;

[0038] S5.3 Determine whether the behavior matrix has converged. If it has not converged and the maximum number of iterations has not been reached, return to step two to continue iterating.

[0039] Optionally, the final electric vehicle cluster load calculation formula in step six is:

[0040]

[0041] in, For the final number Total load of electric vehicle cluster during the time period This is the optimal behavior matrix after convergence. Charging power for a single electric vehicle.

[0042] In a second aspect, the present invention provides an electronic device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the method described in the first aspect.

[0043] Compared with the prior art, this application includes at least one of the following beneficial technical effects:

[0044] By introducing the Weber-Fechner law, we can quantitatively model users' perception of the urgency of charging needs (perceived necessity) and their sensitivity to price changes (perceived suitability), making service fee optimization strategies closer to users' real psychological decision-making mechanisms. This will more effectively guide users to adjust their charging times and achieve load shifting.

[0045] A multi-objective optimization model was constructed that simultaneously covers distribution network load balancing, user charging cost savings, maximization of photovoltaic absorption, and maximization of energy storage arbitrage profits. This model overcomes the limitations of traditional single-objective optimization strategies and takes into account the interests of the power grid, users, aggregators, and energy storage systems.

[0046] By dynamically updating the user behavior matrix and re-optimizing service fees, the final strategy can adapt to load changes caused by user psychological responses, thereby improving the stability and reliability of the strategy in practical applications.

[0047] The final output is the optimal service fee sequence and the corresponding electric vehicle cluster load curve, providing new energy aggregators with a clear and quantitative basis for pricing decisions, and allowing for an intuitive evaluation of the specific effects of the strategy on load smoothing, cost savings, and photovoltaic consumption.

[0048] In summary, this invention constructs a charging behavior response mechanism by quantifying user psychological effects, thereby enabling more precise guidance for users' charging time selection; it establishes a multi-objective collaborative optimization model that takes into account grid load, user costs, photovoltaic consumption, and energy storage revenue; it enhances the adaptability and robustness of the strategy by adopting a closed-loop iterative mechanism; and it provides new energy aggregators with quantifiable evaluation basis for decision-making, systematically solving the problems of inaccurate behavior characterization, single objective, and insufficient practicality in traditional optimization methods. Attached Figure Description

[0049] Figure 1 A flowchart illustrating a method for optimizing electric vehicle charging service fees that incorporates user psychological effects. Detailed Implementation

[0050] The following specific examples illustrate the implementation of the present invention. Those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and various details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of the present invention. It should be noted that, unless otherwise specified, the following embodiments and features described therein can be combined with each other.

[0051] Example: Figure 1 As shown in the figure, the present invention proposes an electric vehicle charging service fee optimization method that integrates user psychological effects. The steps of the method are described in detail below.

[0052] Step 1: Establish a historical data model of electric vehicle charging characteristics and obtain a regional residential load and photovoltaic-storage model; specifically including the following steps:

[0053] Step 1.1: Obtain the charging characteristic sequence of m electric vehicles under disordered charging conditions in a certain area. ,in, Indicates the first The charging characteristic sequence of a vehicle, and , Indicates the first The time when an electric vehicle begins charging. Indicates the first The battery level of an electric vehicle when it begins charging. Indicates the first The time it takes for an electric vehicle to finish charging;

[0054] Step 1.2: Obtain the region feature sequence ,in This represents the electricity sales price sequence on the power grid, and = , Indicates the first Hourly electricity sales price As the benchmark service fee sequence, and Indicates the first The hourly load of residents in this area, , Indicates the first The photovoltaic output of the charging station is measured in hours.

[0055] Step 2: Generate an initial electric vehicle user behavior matrix based on the data from Step 1, and use this matrix to construct the electric vehicle cluster load curve; including the following steps:

[0056] Step 2.1: Initialize the electric vehicle user behavior matrix. Assume the research time scale is T=24 hours, discretized into t=1,2,...24 time periods, and construct an m×T dimensional initial behavior matrix. The matrix elements are defined as follows:

[0057]

[0058]

[0059] In the formula: For electric vehicles, For discrete time periods; , The first The arrival and departure times of the electric vehicles; Before characterization optimization electric vehicles in the first Charging status during a given time period. Step 2.2: Construct the initial electric vehicle cluster load curve, assuming the charging power of a single electric vehicle is... (Constant value), calculate cluster load using the initial behavior matrix:

[0060]

[0061] In the formula: To optimize the first Total load of electric vehicle cluster during the time period (unit: kW); Constant charging power for a single electric vehicle (unit: kW). Let i be the charging state of the i-th EV during time period t. Step 2.3: Calculate the initial user fee using formula (1.3):

[0062]

[0063] In the formula: Initial total charging cost (unit: yuan); For the first Time-based service fee benchmark price (unit: yuan / kWh);

[0064] Set the maximum number of iterations Initial iteration counter Convergence indicators .

[0065] Step 3: Define the multi-objective model and constraint model for optimizing regional service fees, and solve them using the NSGA-II algorithm to obtain the optimal aggregator service fee sequence obtained by the current model. This includes the following steps:

[0066] Step 3.1: Define the optimization variables. Let the optimization variable vector be as shown in equation (1.4):

[0067] In the formula: For the first Best service fee for the time period (unit: yuan / kWh); For the first Time-of-use energy storage charging power (unit: kW, negative value indicates charging); For the first Energy storage discharge power during a given period (unit: kW, positive value indicates discharge).

[0068] Step 3.2: Construct a multi-objective optimization function with the objectives of minimizing the load difference in the distribution network, minimizing the cost to electric vehicle users, maximizing photovoltaic absorption, and maximizing the arbitrage revenue from energy storage. The objective function is shown in formula (1.5): In the formula: Distribution network load differential (unit: kW); Total cost for electric vehicle users (unit: yuan); Photovoltaic curtailment (unit: kWh); Energy storage arbitrage profit (unit: yuan). Specific expressions for each objective:

[0069]

[0070] in:

[0071]

[0072]

[0073] in: ;

[0074]

[0075] in, The distribution network load during time period t. The amount of photovoltaic power used during time period t;

[0076] Step 3.3: Set the constraints for energy storage discharge as follows:

[0077]

[0078] In the formula: Maximum charge / discharge power of energy storage (unit: kW). Set constraints for the evolution of State of Charge (SOC):

[0079]

[0080]

[0081] In the formula: For energy storage, the lowest SOC; Initial SOC for energy storage; Total energy storage capacity (unit: kWh); , These are the energy storage charging and discharging efficiencies (dimensionless). It is an accumulated variable.

[0082] Step 3.4: Solve the model using the NSGA-II algorithm. Set parameters such as population size and maximum number of iterations, call the NSGA-II algorithm to solve the problem, and output the optimal solution.

[0083] Extract the optimal service fee sequence ,in ;

[0084] Step 4: Based on the current optimal aggregator service fee sequence obtained in Step 3, and according to the Weber-Fechner effect, calculate the transition probability for each user in each time slot, and use this as the element of the transition probability matrix to generate the transition probability matrix; specifically including the following steps: Step 4.1, based on the optimal service fee sequence, calculate the actual price paid by the user in time slot t:

[0085]

[0086] In the formula: The actual price paid by the user (unit: yuan / kWh).

[0087] Step 4.2: Considering the Weber-Fechner effect, calculate the probability of necessity for a user to switch to the current time slot using equation (1.14):

[0088]

[0089]

[0090] In the formula: , For necessity probability parameters; For the first Initial SOC of an electric vehicle; The time difference (in hours) between the initial start time of the connection. , The normalized range of the necessity index; The necessity and probability of shifting the start charging time of the nth vehicle to time period t. This represents the probability of a user's necessary migration to time slot t before normalization. Step 4.3: Calculate the suitability probability based on the price difference using equation (1.16):

[0091]

[0092]

[0093]

[0094]

[0095]

[0096]

[0097] In the formula: Price difference (unit: yuan / kWh); This represents the probability of suitability for the user under normal circumstances. This represents the probability of user migration suitability under positive circumstances. The probability of user switching under anxiety conditions. The maximum value of the suitability probability. The critical price for the dead zone-linear zone. The critical price of the linear region-cutoff region. The probability of user switching when there is no price difference and the user is in an active state; Let k be the suitability probability, and k be the slope of the user switching probability curve under normal circumstances. The slope of the user switching probability curve under positive conditions. The slope of the user switching probability curve under anxiety conditions.

[0098] Step 4.4: Weighted fusion is used to obtain the comprehensive transition probability, and an m×T dimensional transition probability matrix is ​​constructed. :

[0099]

[0100] In the formula: , Let be the transition probability weight parameters, and satisfy . ; This represents the overall transition probability.

[0101] Step 5: Calculate the new user behavior matrix based on the user transition probability matrix obtained in Step 4, and compare it with the user behavior matrix generated in the previous iteration. If the two matrices are equal, the model is considered to have converged, and proceed to Step 6. If the two matrices are not exactly the same, return to Step 2 and continue iterating. This includes the following steps:

[0102] Step 5.1: Update the user behavior matrix, setting the transition probability threshold as follows: , for the Electric vehicles, screening Time period set ,like ,Pick This is the start period for new charging. = Update the behavior matrix for the new end segment. :

[0103]

[0104] In the formula: The transition probability threshold (dimensionless); , These represent the start and end points of a new charge; This is the updated behavior matrix. If... ,but , This is the current behavior matrix; For the initial start charging time, For the updated start charging time, For the new end of charging time;

[0105] Step 5.2: Calculate the updated electric vehicle cluster load:

[0106]

[0107] In the formula: For the updated version Total load of electric vehicle cluster during the time period The updated charging and discharging behavior of the i-th EV during the t-th time period;

[0108] Step 5.3: Compare the behavior matrices before and after the change. If they are the same, convergence is considered complete, proceed to step six. If they are different, iterate... but Return to step two and continue iterating, where For the number of iterations, This is the maximum number of iterations set.

[0109] Step Six: Based on the user behavior model generated in Step Five, generate the load curve of the electric vehicle cluster, and then use the converged optimal behavior matrix. The final load of the electric vehicle cluster is calculated using equation (1.20):

[0110]

[0111] In the formula: For the final number Total load of electric vehicle cluster during the time period (unit: kW); This is the optimal behavior matrix after convergence. This represents the charging and discharging behavior of the i-th EV within time period t.

[0112] It is worth noting that this invention is the first to introduce the Weber-Fechner psychological effect quantification model into the optimization process of electric vehicle charging service fees. This method constructs a behavioral response model that can characterize users' true decision-making psychology by establishing the logarithmic relationship between the urgency of user charging demand and the initial electricity volume, and the piecewise functional relationship between user price sensitivity and price difference. Based on this, and driven by this behavioral model, a multi-objective optimization problem is constructed that simultaneously minimizes the distribution network load difference, total user cost, photovoltaic curtailment, and maximizes energy storage arbitrage revenue. A closed-loop iterative solution framework is designed to optimize service fees, predict user responses, and update system states, thus systematically integrating multiple factors such as user psychology, system operation, and market regulation into the mathematical model. The method of this invention, through the psychological effect model, makes the predicted user charging time shifts closer to actual behavior, improving the effectiveness of price signals in guiding load shifts. Furthermore, this invention, through a multi-objective collaborative optimization model, can simultaneously reduce the peak-valley difference of the power grid, reduce user charging expenses, increase local photovoltaic absorption rate, and increase energy storage revenue, thus achieving a balance of interests among multiple parties. In addition, the closed-loop iterative mechanism of this invention ensures that the optimized service fee strategy can still achieve the system objectives after the user responds, improving the stability and applicability of the strategy in practical applications.

[0113] The above specific embodiments are merely several optional embodiments of the present invention. Based on the technical solutions of the present invention and the relevant teachings of the above embodiments, those skilled in the art can make various alternative improvements and combinations to the above specific embodiments.

Claims

1. A method for optimizing electric vehicle charging service fees by incorporating user psychological effects, characterized in that, Includes the following steps: S1. Establish a historical data model of electric vehicle charging characteristics and obtain regional residential load and photovoltaic-storage model; S2. Based on the data from step S1, generate an initial electric vehicle user behavior matrix and construct an electric vehicle cluster load curve; S3. Define the multi-objective model and constraint model for regional service fee optimization, and use the NSGA-II algorithm to solve for the current optimal aggregator service fee sequence. S4. Based on the optimal aggregator service fee sequence, calculate the transition probability of each user in each time slot according to the Weber-Fechner effect, and generate a transition probability matrix; S5. Update the user behavior matrix based on the transition probability matrix and compare it with the behavior matrix of the previous iteration. If the two are consistent, the model is determined to have converged and proceed to the next step. Otherwise, return to step S2 and continue the iteration; S6. Based on the converged user behavior matrix, generate the optimized electric vehicle cluster load curve.

2. The method for optimizing electric vehicle charging service fees by incorporating user psychological effects as described in claim 1, characterized in that, Step S1 specifically includes: S1.1 Obtaining data under disordered charging conditions Charging characteristic sequence of electric vehicles ,in, , Indicates the first Charging characteristic sequence of electric vehicles Indicates the first The time when an electric vehicle begins charging. Indicates the first The battery level of an electric vehicle when it begins charging. Indicates the first The time it takes for an electric vehicle to finish charging; S1.2 Obtaining the regional feature sequence ,in, This indicates the residential electricity load sequence for the area. Represents the residential load sequence. This represents the photovoltaic power output sequence.

3. The method for optimizing electric vehicle charging service fees by incorporating user psychological effects as described in claim 1, characterized in that, Step S2 specifically includes: S2.1 Initialize the user behavior matrix Matrix elements Indicates the first electric vehicles in the first Charging status during a given period; S2.2, Based on behavior matrix and charging power of a single electric vehicle Calculate the initial cluster load ; S2.3 Calculate the initial total charging cost for users. .

4. The method for optimizing electric vehicle charging service fees by incorporating user psychological effects as described in claim 1, characterized in that, Step S3 specifically includes: S3.1 Define the optimization variable vector This includes service fees for each time period and energy storage charging and discharging power; S3.2 Constructing a multi-objective optimization function ,in, Due to poor load on the distribution network, Total cost for electric vehicle users This refers to the amount of solar power curtailment. For energy storage arbitrage profits, This represents the transpose of a matrix. S3.3, Set constraints on energy storage charging and discharging power and SOC evolution; S3.4 Solve using the NSGA-II algorithm and output the optimal service fee sequence. .

5. The method for optimizing electric vehicle charging service fees by incorporating user psychological effects as described in claim 4, characterized in that, Step S4 includes: S4.1, Calculate the first... Actual price paid by users during the time period , This indicates that the power grid is in The price of electricity at that time; S4.2 Calculate the probability of necessity for user switching based on the Weber-Fechner effect. ; S4.3 Calculating suitability probability based on price difference ; S4.4, Weighted fusion yields the overall transition probability. Construct the transition probability matrix.

6. The method for optimizing electric vehicle charging service fees by incorporating user psychological effects according to claim 5, characterized in that, The probability of necessity The calculation formula is: ； in, , This is the difference between the current time period and the original charging time period. , For necessity probability parameters; For the first Initial SOC of an electric vehicle; The time difference between the initial start time of the connection, in hours; , This represents the normalized range of the necessity index.

7. The method for optimizing electric vehicle charging service fees by incorporating user psychological effects according to claim 5, characterized in that, The calculation of the suitability probability includes three user psychological states: normal state, positive state, and anxious state, which correspond to... , , The final suitability probability is its mean.

8. The method for optimizing electric vehicle charging service fees by incorporating user psychological effects according to claim 1, characterized in that, Step S5 specifically includes: S5.1, Based on the transition probability threshold Update user behavior matrix ; S5.2 Calculate the updated cluster load ; S5.3 Determine whether the behavior matrix has converged. If it has not converged and the maximum number of iterations has not been reached, return to step two to continue iterating.

9. The method for optimizing electric vehicle charging service fees by incorporating user psychological effects according to claim 1, characterized in that, The final electric vehicle cluster load calculation formula in step six is ​​as follows: ； in, For the final number Total load of electric vehicle cluster during the time period This is the optimal behavior matrix after convergence. Charging power for a single electric vehicle.

10. An electronic device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the method as described in any one of claims 1-9.

Images