Electric vehicle real-time charging decision method, system, medium, equipment and terminal

Abstract

The present invention belongs to the technical field of electric vehicle charging decision-making, and discloses a real-time charging decision-making method, system, medium, equipment and terminal for electric vehicles. It comprehensively analyzes arrival time, remaining power and psychological safety power factors to construct a charging demand model for electric vehicle owners under different peak and valley electricity prices; in view of the psychological perception differences of vehicle owners on arrival time and remaining power, it establishes an improved cumulative prospect theory heterogeneous reference point model based on risk preference; and uses the cumulative prospect income of the vehicle as an evaluation indicator to evaluate charging decisions. The real-time charging decision-making model for electric vehicles of the present invention can well describe the charging decision-making behavior of irrational vehicle owners and provide ideas for optimizing the charging behavior of large-scale electric vehicles. Under different risk attitudes, the psychological perceived value of vehicle owners changes. The heterogeneous reference point model for analyzing the risk preferences of heterogeneous vehicle owners of the present invention expands the original cumulative prospect theory's characterization of risk attitudes and improves sensitivity.

Description

Technical Field

[0001] The present invention belongs to the technical field of electric vehicle charging decision-making, and in particular relates to a real-time charging decision-making method, system, medium, equipment and terminal for electric vehicles. Background Art

[0002] With the massive growth in the number of electric vehicles, electric vehicle charging behavior and methods have become a hot topic of research. Different groups of electric vehicle users exhibit distinct charging behavior characteristics. For example, due to the unique nature of their work, electric commercial vehicle owners need to actively monitor their vehicle's charge status to avoid driving with low batteries and experiencing range anxiety. Existing research on electric commercial vehicle owners' charging behavior has focused on modeling objective factors such as arrival time and remaining battery life, and has often assumed that owners are completely rational. However, in reality, owners are not completely rational; their decision-making behavior is determined by a combination of objective factors and subjective perceptions. Expected Utility Theory (EUT) assumes that decision makers are completely rational. Cumulative Prospect Theory (CPT) addresses the EUT's lack of rationality by comprehensively analyzing factors such as risk preferences and psychologically perceived value (PPV) when making decisions.

[0003] In terms of establishing charging demand models, the literature (Deng Changmian, Zhang Yong. Research on data-driven electric taxi charging station planning method [J]. Forest Engineering, 2020, 36(03): 77-85.) and (Tian Sheng, Zeng Lili. Research on electric taxi fast charging behavior based on improved kernel density estimation [J]. Transportation Systems Engineering and Information, 2021, 21(04): 221-229.) both analyzed the impact of objective factors such as charging time and remaining power on charging demand modeling, but did not analyze the impact of car owners' subjective perception, and could not accurately reflect the car owners' actual charging needs.

[0004] In terms of establishing charging decision models, the literature (Huang Songyu. Data-driven modeling and analysis of electric vehicle charging behavior and charging demand [D]. Zhejiang University, 2020.) models and analyzes the charging behavior of electric vehicles from the perspective of probabilistic modeling, and uses the characteristic variables of charging behavior to characterize its charging regularity; the literature (Ye Wenhao, Lu Fengyi, Long Xuemei, et al. A charging recommendation strategy for electric taxis based on analysis of time and electricity prices [J / OL]. Electrical Measurement and Instrumentation: 1-8 [2022-03-3].) comprehensively analyzes the impact of taxi time cost and electricity cost on the charging behavior of car owners, but does not analyze the incomplete rationality of car owners in making decisions.

[0005] In terms of CPT reference point setting, the literature (Guo Fubin. Research on travel mode selection behavior of travelers based on cumulative prospect theory [D]. Xiangtan University, 2020.) sets waiting time as the reference point and finds that people's travel mode choices will change with different reference points, but does not analyze the relationship between people's risk attitude and reference point setting; the literature (Xu Junxiang, Zhang Jin. Emergency logistics path selection model and case based on cumulative prospect theory [J / OL]. Journal of Safety and Environment: 1-12 [2022-03-03].) sets reference points for transportation time and road section risk respectively, constructs an emergency logistics path selection model based on dual reference points, and provides improvement ideas for reference point setting.

[0006] Through the above analysis, the problems and defects of the existing technology are as follows:

[0007] (1) Current analyses of electric vehicle charging decision-making behavior often assume that vehicle owners are completely rational. However, in reality, vehicle owners are not completely rational, and their decision-making behavior is influenced by multiple factors. This rational assumption is insufficient.

[0008] (2) The existing charging demand model does not analyze the influence of the car owner’s subjective perception and cannot accurately reflect the car owner’s actual charging demand.

[0009] (3) The existing technology does not analyze the incomplete rationality of the car owner when making decisions, and does not analyze the relationship between people's risk attitude and reference point setting. Summary of the Invention

[0010] In response to the problems existing in the prior art, the present invention provides a method, system, medium, device and terminal for real-time charging decision-making of electric vehicles, and in particular relates to a method for constructing a real-time charging decision model for electric vehicles based on improved cumulative prospect theory.

[0011] The present invention is implemented as follows: a real-time charging decision method for electric vehicles, the real-time charging decision method for electric vehicles comprising:

[0012] Firstly, a comprehensive analysis of arrival time, remaining power and psychological safety power factors is conducted to construct a charging demand model for electric commercial vehicle owners under different peak and valley electricity prices. Secondly, based on the psychological perception differences of commercial vehicle owners towards arrival time and remaining power, an improved cumulative prospect theory heterogeneous reference point model based on risk preference is established. Finally, the cumulative prospect benefit of commercial vehicles is used as an evaluation indicator to assess charging decisions.

[0013] Furthermore, the electric vehicle real-time charging decision-making method includes the following steps:

[0014] Step 1: Building a charging demand model: This combines objective factors with the owner's subjective perception, comprehensively analyzing factors such as PSP, arrival time, and remaining battery life. This model establishes a charging demand model for operating vehicle owners under different peak and valley electricity prices to determine the set of alternative charging decision options and the benefit function.

[0015] Step 2: Constructing an irrational decision model: Based on the irrationality of car owners, the CPT is applied to the modeling of car owners' charging decisions, establishing the ACPT-CDM. The cumulative prospect theory divides the uncertain decision-making process into an editing phase and an evaluation phase. In the editing phase, a reference point is determined and the results are converted into gains or losses relative to the reference point. In the evaluation phase, the value function and weight function are used to analyze and evaluate the car owner's charging decision and select the optimal charging solution.

[0016] Step 3: Construct a heterogeneous reference point model based on risk preference: During the editing stage of the ACPT-CDM, based on the set of alternative options determined by the charging demand model, analyze the differences in PPV among electric vehicle owners, combine personal risk attitudes with psychological perception reference points, improve the reference points of the cumulative prospect theory, and establish a heterogeneous reference point model based on risk preference.

[0017] Furthermore, the construction of the charging demand model in step 1 includes:

[0018] The initial state of charge B is a key factor affecting the charging time of electric vehicles, as shown in the following formula:

[0019]

[0020] Where C is the rated capacity of the electric vehicle battery; B is the initial state of charge of the electric vehicle when it arrives at the station (i.e., the remaining power); B lea The battery state of charge when the electric vehicle is fully charged; P EV is the charging power of the electric vehicle; θ char is the charging efficiency of electric vehicles; T char is the total charging time.

[0021] The start charging time t of an operating vehicle depends on many factors, including the type of vehicle and the owner's personal behavior. Previous studies approximated the distribution of the end time of the vehicle's last trip to a normal distribution through maximum likelihood estimation and regarded it as the start charging time, expressed as formula (2):

[0022]

[0023] Among them, μ s , σ s are the expectation and standard deviation of t, respectively.

[0024] Furthermore, the construction of the charging demand model in step 1 further includes:

[0025] Classification is done by setting three charging states:

[0026] State 1: Arriving during peak hours, and the charging time is less than the remaining time during the peak period, the owner has two options: charge during the peak period, or wait until the normal period to charge during the normal period;

[0027] State 2: Arriving during peak hours, and the charging time is greater than the remaining time during the peak period, the owner has two options: charge during the off-peak period or wait until the off-peak period to charge during the off-peak period;

[0028] State 3: Arriving during normal hours, the owner charges directly during normal hours.

[0029] In the charging status analysis, the relationship between arrival time, remaining power, charging time and charging status is comprehensively analyzed; where (0, t0) is the peak period, (t0, t max ) is the normal period.

[0030] When the remaining power is the minimum power B min , charge it to the maximum capacity B max Requires T max , this is the longest charging time, and when the remaining power is B max When the owner arrives at time 0, the maximum charging cut-off time is T max , the minimum charging cut-off time is 0 straight line Indicates the corresponding relationship between the remaining power and the charging cut-off time when time 0 is reached. Corresponding car owner t0-T max The correspondence between the remaining power when the time arrives and the charging cut-off time.

[0031] Set time t0 as the start time of the normal period, so time t0 is used as the critical point of charging cut-off time. When the time t∈(0,t0-T max )hour, Towards Moving, the straight line cluster forms a parallelogram area a0b0c0d0 under the limitation of the remaining power range. In the area, the charging decision is state 1, charging during peak hours or normal hours; when reaching time t∈(t0-T max ,t0), Towards Move to form a parallelogram area c0d0e0f0, in which state 1 and state 2 exist at the same time, and division is performed when a mixture of charging states occurs; l mn is a dynamically changing straight line segment, representing the charging cut-off curve under different arrival times. and The equations are:

[0032]

[0033] l mn With straight line t end =t0 There is an intersection point K, with point K as the dividing point. The decisions of the car owners above point K are in state one, and the decisions of the car owners below point K are in state two.

[0034] Converting the charging cut-off time to the arrival time, the equation of the straight line segment is:

[0035]

[0036] When the car owner M When the time arrives, the remaining power is higher than B M State 1, when the remaining power is lower than B M It is state 2; similarly, at any moment, the two mixed states in the time period are separated by a straight line segment.

[0037] Modeling is done for the vehicle owner to avoid the evening peak; when the vehicle owner arrives late, the vehicle owner charges the battery to B max It takes a long time, and most car owners charge the battery to the psychological safety level B saf Then leave the charging station.

[0038] The equation of the line segment is:

[0039]

[0040] Car owner t N For example, if the remaining power is higher than B N At t max Charge the remaining power to B saf ; If the remaining power is lower than B N , the owner charges to B saf Some peak operating periods will be missed.

[0041] Three charging schemes are set for the three charging states:

[0042] Charging only during peak hours: The owner arrives at a peak hour, and the charging time is less than the remaining time during the peak hour;

[0043] Charging only during normal hours: The owner arrives during peak hours, and the charging time is less than the remaining time during peak hours, or the owner arrives during normal hours;

[0044] Only charging during inter-peak periods: The owner arrives during peak hours, but the charging time is greater than the remaining time during the peak period.

[0045] The owner's income is the measurement indicator. The income function of electric vehicles operating under the three charging schemes is defined as follows:

[0046] Y = E-J1-J2;

[0047] Among them, Y is the actual profit, E is the total operating profit for the whole day, J1 is the charging fee, and J2 is the customer loss.

[0048] (1) The benefits of operating vehicle owners choosing to charge during peak hours:

[0049]

[0050] (2) The benefits of operating vehicle owners choosing to charge during normal hours are:

[0051]

[0052] (3) The third benefit of operating vehicle owners choosing to charge across different time periods:

[0053]

[0054] Among them, p1 is the passenger flow during peak hours, and p2 is the passenger flow during normal hours; T1 is the charging time during peak hours, and T2 is the charging time during normal hours; q1 is the average customer price during peak hours, and q2 is the average customer price during normal hours; S1 is the electricity price during peak hours, and S2 is the electricity price during normal hours.

[0055] Furthermore, the construction of the risk preference-based heterogeneous reference point model in step 3 includes:

[0056] (1) Heterogeneous reference point model based on risk preference

[0057] In the EUT model, vehicle owners make charging decisions by comparing the expected value of benefits from various charging decisions, with drivers favoring those with higher expected values. During the CPT calculation process, we analyze irrational factors. The choice of reference point influences the calculation of the cumulative prospect value, and we refine the reference point during the editing phase.

[0058] 1) Psychological Perception Reference Point

[0059] Assume I0 is the psychological reference point for heterogeneous car owners to choose charging solutions, and the expected profit value of the car owner is:

[0060] I0=EU=∑(Y i ,P i );

[0061] Among them, U(x) is the expected utility function, Y i represents the benefit of each charging scheme; P i Indicates the corresponding probability; i=1,2,3.

[0062] The PPV deviation is ΔI. When the total profit I(x) gained in one day is greater than the expected profit I0, the car owner obtains a positive PPV ΔI. When the total profit I(x) gained in one day is less than the expected profit I0, the car owner obtains a negative PPV ΔI.

[0063] ΔI=I(x)-I0.

[0064] (2) Constructing a psychological perception reference point model based on risk preference

[0065] μ (0<μ<1) is the risk preference factor. In the decision-making process of car owners, if 0<μ<0.5, the car owner is risk-averse; if 0.5<μ<1, the car owner is risk-seeking; and μ﹦0.5 is risk-neutral.

[0066] When determining the reference point, a heterogeneous reference point model based on risk preferences was established for different PPVs of car owners. The maximum and minimum values of psychological security were used as threshold points for each region. A risk factor μ was also introduced. The heterogeneous reference point was obtained by multiplying the risk factor μ with the difference between the maximum and minimum values and adding the minimum value. When decision makers hold different risk preferences, the reference point also fluctuates within the aforementioned extreme value range.

[0067] I=I min +(I max -I min )μ.

[0068] Furthermore, the construction of the risk preference-based heterogeneous reference point model in step 3 further includes prospect value calculation, which includes:

[0069] (1) Value Function

[0070]

[0071] Among them, α and β represent the owner's sensitivity to risk, λ is the loss aversion coefficient, and α﹦β﹦0.88, λ﹦2.25.

[0072] (2) Weight function

[0073] When car owners face the benefits:

[0074]

[0075] When a car owner faces losses:

[0076]

[0077] Among them, γ﹦0.61 and δ﹦0.69.

[0078] CPT decision weight function πi + and π i － The definition is as follows:

[0079]

[0080] Among them, π ij + is a positive cumulative decision weight function, that is, the cumulative decision weight function when the decision maker faces "benefits"; π ij － is a negative cumulative decision weight function, that is, the cumulative decision weight function when the decision maker faces "loss"; n is the possible result when the owner feels the benefit under each charging scheme; m is the possible result when the owner feels the loss under each charging scheme. + and ω － is a strictly increasing function that satisfies:

[0081] ω + (0) = ω - (0)=0;

[0082] ω + (1) = ω - (1) = 1;

[0083] (3) Calculation of cumulative prospect value

[0084]

[0085] Among them, CPV j + is the positive cumulative prospect value under scenario j, CPV j － is the negative cumulative prospect value under scenario j; CPV j is the actual comprehensive cumulative prospect value, j = 1, 2, 3, CPV1, CPV2 and CPV3 are the comprehensive cumulative prospect values of plan 1, plan 2 and plan 3 respectively; when making charging decisions, electric vehicle owners usually tend to choose the charging plan with the largest comprehensive cumulative prospect value as the optimal charging option.

[0086] Another object of the present invention is to provide an electric vehicle real-time charging decision system applying the electric vehicle real-time charging decision method, the electric vehicle real-time charging decision system comprising:

[0087] A charging demand model building module is used to comprehensively analyze arrival time, remaining power, and psychological safety power factors to build a charging demand model for electric vehicle owners under different peak and valley electricity prices;

[0088] A heterogeneous reference point model building module is used to establish a heterogeneous reference point model based on risk preference and improved cumulative prospect theory, taking into account the differences in the psychological perceptions of vehicle owners regarding arrival time and remaining battery power.

[0089] The charging decision evaluation module is used to evaluate charging decisions based on the cumulative prospective benefits of operating vehicles.

[0090] Another object of the present invention is to provide a computer device, comprising a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor performs the following steps:

[0091] A comprehensive analysis of arrival time, remaining power and psychological safety power factors is conducted to construct a charging demand model for electric commercial vehicle owners under different peak and valley electricity prices. Based on the differences in psychological perception of arrival time and remaining power among commercial vehicle owners, an improved cumulative prospect theory heterogeneous reference point model based on risk preference is established. Charging decisions are evaluated using the cumulative prospect benefit of commercial vehicles as an evaluation indicator.

[0092] Another object of the present invention is to provide a computer-readable storage medium storing a computer program, wherein when the computer program is executed by a processor, the processor performs the following steps:

[0093] A comprehensive analysis of arrival time, remaining power and psychological safety power factors is conducted to construct a charging demand model for electric commercial vehicle owners under different peak and valley electricity prices. Based on the differences in psychological perception of arrival time and remaining power among commercial vehicle owners, an improved cumulative prospect theory heterogeneous reference point model based on risk preference is established. Charging decisions are evaluated using the cumulative prospect benefit of commercial vehicles as an evaluation indicator.

[0094] Another object of the present invention is to provide an information data processing terminal, which is used to implement the electric vehicle real-time charging decision system.

[0095] In combination with the above technical solutions and the technical problems solved, please analyze the advantages and positive effects of the technical solutions to be protected by the present invention from the following aspects:

[0096] First, in view of the technical problems existing in the above-mentioned prior art and the difficulty of solving these problems, this paper closely combines the technical solutions to be protected by the present invention and the results and data during the research and development process, and analyzes in detail and in depth how the technical solutions of the present invention solve the technical problems and some creative technical effects brought about by solving the problems. The specific description is as follows:

[0097] To accurately describe the irrational charging decision-making behavior of electric commercial vehicle owners, this paper establishes an electric commercial vehicle charging decision-making model based on improved cumulative prospect theory. First, a comprehensive analysis of factors such as arrival time, remaining battery charge, and psychological safety level is conducted to construct a charging demand model for electric commercial vehicle owners under different peak and valley electricity prices. Second, based on the differences in the psychological perceptions of arrival time and remaining battery charge among commercial vehicle owners, a heterogeneous reference point model based on improved cumulative prospect theory is established based on risk preference. Finally, charging decisions are evaluated using the cumulative prospect benefit of commercial vehicles as an evaluation metric. Case results show that, given a certain psychological safety level, owners with higher perceived psychological value will charge early, while those with lower perceived psychological value will tend to delay charging. This model, which combines risk attitude and psychological perceived value, can more accurately reflect the irrational decision-making behavior of commercial vehicle owners and has important theoretical and practical significance.

[0098] Second, considering the technical solution as a whole or from the perspective of the product, the technical effects and advantages of the technical solution to be protected by the present invention are described in detail as follows:

[0099] (1) The real-time charging decision model for electric vehicles based on the improved cumulative prospect theory proposed in this paper can well describe the charging decision-making behavior of irrational car owners and provide certain ideas for optimizing the charging behavior of large-scale electric vehicles.

[0100] (2) When the psychological safety level is low, the choice of charging in advance by car owners with higher psychological perceived value will shift forward overall; when the psychological safety level is high, the choice of postponing charging by car owners with lower psychological perceived value will shift backward overall. The case results provided by this invention show that combining risk attitude and psychological perceived value can more accurately reflect the irrational decision-making behavior of operating vehicle owners.

[0101] (3) The daytime charging behavior of operating vehicle owners is concentrated in the low-operation period, which is consistent with the empirical conclusion that charging decisions should avoid the peak period. The case results provided by this invention better reflect the proactive decision-making characteristics of decision makers to seek benefits and avoid harm.

[0102] Third, as auxiliary evidence for the inventiveness of the claims of the present invention, it is also reflected in the following important aspects:

[0103] (1) The expected benefits and commercial value of the technical solution of the present invention after transformation are:

[0104] The present invention can provide owners of electric commercial vehicles with a more accurate charging behavior reference by simulating their charging behavior, thereby enabling owners to increase their profits.

[0105] (2) The technical solution of the present invention fills the technical gap in the industry at home and abroad:

[0106] A. Existing research on the charging needs of electric commercial vehicle owners has focused primarily on objective factors such as arrival time and remaining battery power. This paper combines these objective factors with the owner's subjective perception, comprehensively considering factors such as psychological safety level, arrival time, and remaining battery power, to establish a charging demand model for commercial vehicle owners under different peak and valley electricity prices.

[0107] B. Existing research on the charging behavior of electric vehicle owners often assumes complete rationality of decision makers. This paper considers the irrationality of vehicle owners and applies cumulative prospect theory to modeling their charging decisions. This paper establishes an improved cumulative prospect theory-based electric vehicle charging decision model (ACPT-CDM). Cumulative prospect theory divides the uncertain decision-making process into editing and evaluation phases. In the editing phase, a reference point is determined and the results are converted into gains or losses relative to the reference point. In the evaluation phase, a value function and a weight function are used to analyze and evaluate the vehicle owner's charging decisions and select the optimal charging solution.

[0108] C. Existing research on cumulative prospect theory has ignored the impact of individual risk preferences on the setting of reference points. During the editing phase of the ACPT-CDM, this paper, based on the set of alternative solutions determined by the charging demand model, considers the differences in the PPV of electric vehicle owners, combines individual risk preferences with reference points, improves the reference points of cumulative prospect theory, and establishes a heterogeneous reference point model that considers risk preferences. BRIEF DESCRIPTION OF THE DRAWINGS

[0109] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the embodiments of the present invention. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0110] Figure 1 This is a flow chart of a real-time charging decision-making method for electric vehicles provided by an embodiment of the present invention;

[0111] Figure 2 This is a structural block diagram of the electric vehicle real-time charging decision system provided by an embodiment of the present invention;

[0112] Figure 3 This is a block diagram of an electric vehicle charging decision model based on the improved cumulative prospect theory provided by an embodiment of the present invention;

[0113] Figure 4 This is a charging status analysis diagram provided by an embodiment of the present invention;

[0114] Figure 5 This is a charging state division diagram provided by an embodiment of the present invention;

[0115] Figure 6 This is a schematic diagram of a psychological safety power model provided by an embodiment of the present invention;

[0116] Figure 7 is a schematic diagram of a psychological perception reference point model provided by an embodiment of the present invention;

[0117] Figure 8 This is a schematic diagram of an example provided by an embodiment of the present invention;

[0118] FIG9( a ) is a schematic diagram of a vehicle owner's charging decision under the AUT model provided by an embodiment of the present invention;

[0119] FIG9( b ) is a schematic diagram of a vehicle owner's charging decision under the CPT model provided by an embodiment of the present invention;

[0120] FIG9( c ) is a schematic diagram of an ACPT-CDM vehicle owner's charging decision-making process according to an embodiment of the present invention;

[0121] Figure 10(a) 、 10(b) 10(c) is a heterogeneous PSP decision diagram provided by an embodiment of the present invention;

[0122] FIG10( a ) is a schematic diagram of a charging decision of a car owner with a PSP of 0.3 provided by an embodiment of the present invention;

[0123] FIG10( b ) is a schematic diagram of a charging decision of a car owner with a PSP of 0.5 provided by an embodiment of the present invention;

[0124] FIG10( c ) is a schematic diagram of a charging decision of a car owner with a PSP of 0.7 provided by an embodiment of the present invention;

[0125] FIG11( a ) is a schematic diagram of a charging decision-making process for a risk-averse vehicle owner according to an embodiment of the present invention;

[0126] FIG11( b ) is a schematic diagram of a robust charging decision-making process for a vehicle owner according to an embodiment of the present invention;

[0127] FIG11( c ) is a schematic diagram of a charging decision-making process of a risk-seeking car owner according to an embodiment of the present invention;

[0128] FIG11( d ) is a schematic diagram of a charging decision of a car owner using the CPT model provided by an embodiment of the present invention;

[0129] In the figure: 1. Charging demand model construction module; 2. Heterogeneous reference point model construction module; 3. Charging decision evaluation module. DETAILED DESCRIPTION

[0130] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0131] In order to solve the problems existing in the prior art, the present invention provides a method, system, medium, device and terminal for real-time charging decision-making of electric vehicles. The present invention is described in detail below with reference to the accompanying drawings.

[0132] 1. Explanatory Examples In order to enable those skilled in the art to fully understand how to implement the present invention, this section provides an illustrative example that expands upon the technical solutions of the claims.

[0133] like Figure 1 As shown, the electric vehicle real-time charging decision-making method provided by the embodiment of the present invention includes the following steps:

[0134] S101: Comprehensively analyze arrival time, remaining power, and psychological safety power factors to build a charging demand model for electric vehicle owners under different peak and valley electricity prices;

[0135] S102: Based on the differences in the psychological perceptions of vehicle owners regarding arrival time and remaining battery power, a modified cumulative prospect theory heterogeneous reference point model based on risk preference is established;

[0136] S103, using the cumulative prospective benefits of the operating vehicles as an evaluation indicator, evaluate the charging decision.

[0137] like Figure 2 As shown, the electric vehicle real-time charging decision system provided by the embodiment of the present invention includes:

[0138] Charging demand model construction module 1 is used to comprehensively analyze arrival time, remaining power, and psychological safety power factors to build a charging demand model for electric vehicle owners under different peak and valley electricity prices;

[0139] Heterogeneous reference point model construction module 2 is used to establish a heterogeneous reference point model based on the improved cumulative prospect theory of risk preference, taking into account the differences in the psychological perception of vehicle owners regarding arrival time and remaining battery power.

[0140] The charging decision evaluation module 3 is used to evaluate the charging decision using the cumulative prospective benefits of the operating vehicles as an evaluation indicator.

[0141] The technical solution of the present invention is further described below with reference to specific embodiments. 1. Summary of the Invention

[0143] When describing the charging decision-making behavior of electric vehicle owners, current technologies have the following problems: when modeling charging demand, they mostly analyze objective factors while ignoring the subjective perception of the owners; when modeling charging demand, they mostly assume that the owners are completely rational while ignoring their irrationality; and existing CPT-related research ignores the impact of individual risk preferences on reference points. To address these problems, the present invention establishes a Charging Decision Model of Operational EV Based on Adapted Cumulated Prospect Theory (ACPT-CDM), the structural block diagram of which is shown in the figure below. Figure 3 shown.

[0144] (1) Charging demand model

[0145] By combining objective factors with the subjective perception of vehicle owners, comprehensively analyzing factors such as PSP, arrival time and remaining power, a charging demand model for operating vehicle owners under different peak and valley electricity prices is established to determine the alternative plan set and benefit function for charging decisions.

[0146] (2) Irrational decision-making model

[0147] By analyzing the irrationality of car owners, the CPT was applied to modeling their charging decisions, resulting in the ACPT-CDM. Cumulative prospect theory divides the uncertain decision-making process into two phases: editing and evaluation. In the editing phase, a reference point is determined and the results are converted into gains or losses relative to that reference point. In the evaluation phase, value and weight functions are used to analyze and evaluate the car owner's charging decisions and select the optimal charging solution.

[0148] (3) Heterogeneous reference point model based on risk preference

[0149] During the editing stage of the ACPT-CDM, based on the set of alternative options determined by the charging demand model, the differences in the PPV of electric commercial vehicle owners were analyzed, and the reference points of the cumulative prospect theory were improved by combining personal risk attitudes with psychological perception reference points. A heterogeneous reference point model based on risk preference was established.

[0150] 2. Charging demand model

[0151] Electric commercial vehicles operate for extended periods each day, but due to battery capacity limitations, they require at least two charges per day: one at night and at least one during the day. This article only analyzes the charging of vehicles operated during the day.

[0152] The initial state of charge B is a key factor affecting the charging time of electric vehicles, as shown in the following formula:

[0153]

[0154] Where C is the rated capacity of the electric vehicle battery; B is the initial state of charge of the electric vehicle when it arrives at the station (i.e., the remaining power); B lea The battery state of charge when the electric vehicle is fully charged; P EV is the charging power of the electric vehicle; θ char is the charging efficiency of electric vehicles; T char is the total charging time.

[0155] The start charging time t of an operating vehicle depends on many factors, including the type of vehicle and the owner's personal behavior. Previous studies approximated the distribution of the end time of the vehicle's last trip to a normal distribution through maximum likelihood estimation and regarded it as the start charging time, expressed as formula (2):

[0156]

[0157] 2.1 Charging state classification

[0158] In general, the probability of an electric vehicle being charged at a specific moment may correspond to different combinations of charging conditions, such as remaining power, arrival time, and the owner's psychological safety power (PSP). Therefore, the charging option at a specific moment is not unique. The present invention classifies them by setting three charging states.

[0159] State 1: Arriving during peak hours, and the charging time is less than the remaining time during the peak period. In this case, the owner has two options: charge during the peak period, or wait until the normal period to charge during the normal period.

[0160] State 2: Arriving during peak hours, and the charging time is greater than the remaining time during the peak hours. In this case, the owner has two options: charge during the off-peak hours or wait until the off-peak hours to charge during the off-peak hours.

[0161] State 3: Arriving during normal hours. At this time, the owner can charge directly during normal hours.

[0162] This paper de-labels car owners, thus dividing charging decisions into those for charging at the current moment (e.g., arriving at a peak hour, resulting in a decision to charge at peak time) and those for charging at a future time (e.g., arriving at a peak hour, resulting in a decision to charge at a normal time). Based on the above analysis, for any given moment, the charging behavior calculated in this paper comes from car owners who choose to charge directly at the current moment. Car owners who arrive but do not choose to charge at the current moment are included in the model calculation at a future moment, so the arrival time is equivalent to the start time of charging.

[0163] like Figure 4As shown in the figure, the charging status analysis diagram comprehensively analyzes the relationship between arrival time, remaining power, charging time and charging status. Among them, (0, t0) is the peak period, (t0, t max ) is the normal period.

[0164] Assume that the remaining power is the minimum power B min , charge it to the maximum capacity B max Requires T max , this is the longest charging time, and when the remaining power is B max When the owner arrives at time 0, the maximum charging cut-off time is T max , the minimum charging cut-off time is 0, such as Figure 4 As shown, the straight line Indicates the corresponding relationship between the remaining power and the charging cut-off time when time 0 is reached. Corresponding car owner t0-T max The correspondence between the remaining power when the time arrives and the charging cut-off time.

[0165] Set time t0 as the start time of the normal period, so it is used as the critical point of charging cut-off time. Figure 4 It can be seen that when the arrival time t∈(0,t0-T max )hour, Towards Move, at this time the straight line cluster forms a parallelogram area a0b0c0d0 under the limitation of the remaining power range. In this area, the charging decision is state 1, that is, charging during peak hours or normal hours. When the arrival time t∈(t0-T max ,t0), Towards Move to form a parallelogram area c0d0e0f0, but in this area there are both state 1 and state 2, and the charging state is mixed, so it needs to be divided. Figure 4 As shown, l mn is a dynamically changing straight line segment, representing the charging cut-off curve under different arrival times. and The equations are:

[0166]

[0167] l mn With straight line t end =t0 There is an intersection point K, with point K as the dividing point. The decisions of the car owners above point K are in state one, and the decisions of the car owners below point K are in state two.

[0168] Convert the charging deadline time to the arrival time, and we can get Figure 5 .

[0169] Depend on Figure 5 The equation of the straight line segment is:

[0170]

[0171] Taking point M as an example, the owner t M When the time arrives, the remaining power is higher than B M State 1, when the remaining power is lower than B M It is state 2. Similarly, at any time, the straight line segment can be used to effectively separate the two mixed states in the period.

[0172] Figure 6 For the psychological safety power model, the model is built for the operating vehicle owner to avoid the evening peak. When the owner arrives late, the owner charges the battery to B max It takes a long time. In order to avoid missing the large passenger flow during the evening peak, most car owners charge their batteries to a psychological safety level. Most car owners charge their batteries to a psychological safety level B. saf , and it will leave the charging station.

[0173] Depend on Figure 6 The equation of the line segment is:

[0174]

[0175] Car owner t N For example, if the remaining power is higher than B N At t max Charge the remaining power to B saf ; If the remaining power is lower than B N , the owner charges to B saf Some peak operating periods will be missed.

[0176] Modeling is done for the vehicle owner to avoid the evening peak; when the vehicle owner arrives late, the vehicle owner charges the battery to B max It takes a long time, and most car owners charge the battery to the psychological safety level B saf Then leave the charging station.

[0177] 2.2 Charging Solution

[0178] For the three charging states mentioned above, the present invention provides three charging schemes:

[0179] Charging only during peak hours: The owner arrives at a peak hour, and the charging time is less than the remaining time during the peak hour;

[0180] Charging only during normal hours: The owner arrives during peak hours, and the charging time is less than the remaining time during peak hours, or the owner arrives during normal hours;

[0181] Only charging during inter-peak periods: The owner arrives during peak hours, but the charging time is greater than the remaining time during the peak period.

[0182] The owner's income is the measurement indicator of this invention. For the three charging schemes proposed in this invention, the income function of the electric commercial vehicle can be defined as follows without loss of generality:

[0183] Y=E-J1-J2(7)

[0184] Among them, Y is the actual profit, E is the total profit of the whole day operation, J1 is the charging fee, and J2 is the customer loss.

[0185] (1) The benefits of operating vehicle owners who choose to charge during peak hours (benefit 1):

[0186]

[0187] (2) The benefits of operating vehicle owners who choose to charge during normal hours (benefit 1):

[0188]

[0189] (3) The benefits of operating vehicle owners who choose to charge across different time periods (benefit three):

[0190]

[0191] Among them, p1 is the passenger flow during peak hours, and p2 is the passenger flow during normal hours; T1 is the charging time during peak hours, and T2 is the charging time during normal hours; q1 is the average customer price during peak hours, and q2 is the average customer price during normal hours; S1 is the electricity price during peak hours, and S2 is the electricity price during normal hours.

[0192] 3. Charging decision model based on improved cumulative prospect theory

[0193] This invention sets a reference point for the editing phase and describes the evaluation phase. First, the reference point is refined, followed by the calculation of the prospect value. Unlike the EUT, the CPT is applicable to risky and uncertain decisions and can explain irrational situations that violate the EUT.

[0194] 3.1 Heterogeneous Reference Point Model Based on Risk Preference

[0195] In the EUT model, vehicle operators make charging decisions by comparing the expected value of benefits from various charging options, with drivers favoring those with higher expected values. However, the CPT calculation process analyzes irrational factors. The choice of reference point influences the calculation of the cumulative prospect value and also has a significant impact on travel decisions.

[0196] (1) Psychological Perception Reference Point

[0197] Let I0 be the psychological reference point for heterogeneous car owners to choose charging solutions, that is, the car owner’s expected profit value. Among them, U(x) is the expected utility function, Y i represents the benefit of each charging scheme; P i Indicates the corresponding probability; i=1,2,3.

[0198] I0=EU=∑(U i ,P i ) (11)

[0199] The psychological perceived value (PPV) deviation is ΔI. When the total profit I(x) gained in one day is greater than the expected profit I0, the car owner obtains a positive psychological perception ΔI; when the total profit I(x) gained in one day is less than the expected profit I0, the car owner obtains a negative psychological perception ΔI.

[0200] ΔI=I(x)-I0 (12)

[0201] Figure 7 This is a model diagram of psychological perception reference points.

[0202] Since the arrival time is (t0-T max ,t0) interval has mixed charging states, so under this reference point model, the interval division reference Figure 5 State analysis diagram.

[0203] In view of the fact that different electric car owners have different perceptions of their current situation and different sensitivities to the remaining battery power and arrival time, the present invention analyzes the heterogeneity of reference points, divides the intervals according to the different charging states, and sets reference points for each of the three areas to achieve a more accurate assessment of the owner's future decision-making. Ideally, the earlier the owner arrives, the more charging options they have, and the stronger their psychological security. Conversely, the later the arrival time, the fewer charging options they have, and the worse their psychological security. Similarly, the more remaining power, the stronger the sense of psychological security; the less remaining power, the worse the sense of psychological security. Therefore, in interval 1, point a1 is the maximum point and point d1 is the minimum point; the same applies to intervals 2 and 3.

[0204] (2) Psychological Perception Reference Point Model Based on Risk Preference

[0205] μ (0<μ<1) is the risk preference factor. In the decision-making process of car owners, if 0<μ<0.5, the car owner is risk-averse; if 0.5<μ<1, the car owner is risk-seeking; and μ﹦0.5 is risk-neutral.

[0206] When determining the reference point, a heterogeneous reference point model based on risk preference was established for different PPVs of car owners (see Figure 7 ), this model can improve the sensitivity of ACPT-CDM under different PPVs.

[0207] The maximum and minimum values of psychological security are used as the threshold points of each area. At the same time, the risk factor μ is introduced. By multiplying the risk factor μ with the difference between the maximum and minimum values and adding the minimum value, we can finally obtain a heterogeneous reference point. That is, when decision makers hold different risk preferences, their reference points also fluctuate in the aforementioned extreme value range.

[0208] I=I min +(I max -I min )μ(13)

[0209] 3.2 Foreground value calculation

[0210] (1) Value Function

[0211] The classic value function model is:

[0212]

[0213] Where α and β represent the owner's sensitivity to risk, and λ is the loss aversion coefficient. Analysis of a large amount of experimental data shows that α = β = 0.88 and λ = 2.25 are more appropriate.

[0214] (2) Weight function

[0215] The present invention adopts the weight function proposed by Yversky and Kahneman.

[0216] When car owners face the benefits:

[0217]

[0218] When a car owner faces losses:

[0219]

[0220] Among them, γ﹦0.61 and δ﹦0.69, which are consistent with the empirical data.

[0221] CPT decision weight function π i + and π i － It can be defined as follows:

[0222]

[0223] Among them, π ij + is a positive cumulative decision weight function, that is, the cumulative decision weight function when the decision maker faces "benefits"; π ij －is a negative cumulative decision weight function, that is, the cumulative decision weight function when the decision maker faces "loss"; n is the possible result when the owner feels the benefit under each charging scheme; m is the possible result when the owner feels the loss under each charging scheme. + and ω － is a strictly increasing function that satisfies:

[0224] ω + (0) = ω - (0)=0(19)

[0225] ω + (1) = ω - (1) = 1(20)

[0226] (3) Calculation of cumulative prospect value

[0227]

[0228] Among them, CPV j + Represents the positive cumulative prospect value under scenario j, CPV j － Represents the negative cumulative prospect value under scenario j. CPV j represents the actual comprehensive cumulative prospect value, j = 1, 2, 3, CPV1, CPV2 and CPV3 represent the comprehensive cumulative prospect values of plan 1, plan 2 and plan 3 respectively. When making charging decisions, electric vehicle owners usually tend to choose the charging plan with the largest comprehensive cumulative prospect value as the optimal charging option.

[0229] 4. The real-time charging decision model for electric vehicles based on the improved cumulative prospect theory proposed in this invention can well describe the charging decision-making behavior of irrational car owners and provide certain ideas for optimizing the charging behavior of large-scale electric vehicles.

[0230] When the psychological safety level is low, drivers with higher perceived psychological value tend to choose to charge early. When the psychological safety level is high, drivers with lower perceived psychological value tend to choose to postpone charging. This case study suggests that combining risk attitudes with perceived psychological value can more accurately reflect the irrational decision-making behavior of commercial vehicle owners.

[0231] The daytime charging behavior of operating vehicle owners is concentrated in the low-operation period, which is consistent with the empirical conclusion that charging decisions should avoid the peak operation period. The case results well reflect the proactive decision-making characteristics of decision makers to seek benefits and avoid harm.

[0232] Subsequent analysis will focus on the impact of electric vehicle charging and discharging on the power grid under time-of-use electricity prices.

[0233] 2. Application Examples: In order to demonstrate the creativity and technical value of the technical solution of the present invention, this section provides application examples of the claimed technical solution on specific products or related technologies.

[0234] The effectiveness and accuracy of this model are verified through analysis and research on the charging behavior of electric commercial vehicles in a certain city.

[0235] 1. The peak-to-flat electricity price multiplier is set to three times, and the charging decision preferences of risk-seeking car owners in the EUT, CPT and ACPT-CDM models are modeled and analyzed under the conditions of remaining power B~N(0.5,0.25) and arrival time t~N(5,9).

[0236] 2. The peak-to-flat electricity price multiplier is set to three times, and the psychological safety power in ACPT-CDM is set to 0.3, 0.5, and 0.7 respectively. The charging decision preferences of conservative car owners are compared under the conditions of remaining power B~N(0.5,0.25) and arrival time t~N(5,9).

[0237] 3. The peak-to-flat electricity price multiplier is set to three times, and the risk preference factors are set to 0.2, 0.5, and 0.8 in the ACPT-CDM model, respectively. The charging decision preferences of vehicle owners in the model are analyzed under the conditions of Bsaf=0.5, remaining power B~N(0.5,0.25), and arrival time t~N(5,9). In addition, the charging decision behavior of operating vehicle owners in the CPT model is modeled and analyzed under the same conditions.

[0238] 3. Evidence of the effects of the embodiments: The embodiments of the present invention have achieved some positive effects during the development or use process, and indeed have great advantages over the existing technology. The following content describes them with reference to the data, charts, etc. of the experimental process.

[0239] 1. Case Analysis

[0240] 1.1 Case parameter settings

[0241] According to the peak and valley electricity price periods in Wuhan, 9:00-14:00 is the peak period, and 14:00-17:00 is the off-peak period. In addition, the peak operating hours for taxis in Wuhan are 7:00-10:00 and 17:00-20:00. Obviously, operating vehicle owners will try to avoid charging during peak electricity price and operating periods. The present invention sets the research time as 9:00-17:00. The trial value method is used to determine the time parameters. Based on the simulation conclusions and market research results, the paper finally determines that the mean of the normal distribution is 14:00 and the standard deviation is 3.

[0242] For the time and power data in the present invention, the per-unit value data is used, that is, 0.9 represents 90% of the power, and 9:00 is recorded as 1. The present invention assumes that the remaining power obeys the normal distribution, that is, B~N(0.5,0.25), and 0.1 <B<B saf , when the power reaches 0.9, charging is completed; the arrival time is set to obey the normal distribution, that is, t~N(5,9), and t∈(0,8).

[0243] According to the survey, BYDE6 electric vehicle is a common operating vehicle model in Wuhan. Its battery capacity is 45kW·h, and the charging power of Wuhan’s general charging pile is 35kW. The charging efficiency is θ Char Table 1 shows the parameters for peak and normal periods. The normal electricity price and average customer unit price are based on the Wuhan price.

[0244] Table 1 Parameter settings

[0245]

[0246] After the parameters are introduced, we can get Figure 8 It takes 1.083 hours to charge the remaining power from 0.1 to 0.9, and 0.545 hours to charge from 0.1 to 0.5; time 5 is the end time of the peak period.

[0247] This paper designs three case scenarios to validate the model's effectiveness, comparing and analyzing the charging decisions of drivers at different arrival times. Drivers' charging decisions during normal arrival times are relatively fixed and serve as auxiliary decision-making tools. The case studies focus on drivers' decisions during peak arrival times. Table 2 shows a table of psychological extreme points, used for calculating heterogeneous reference points in the ACPT-CDM model. In the CPT model, the expected value of return is used as the reference point.

[0248] Table 2 Psychological extreme point table

[0249]

[0250] 1.2 Case Analysis

[0251] (1) Model comparison case

[0252] Under triple electricity prices, risk-seeking car owners compare their charging decision preferences under the EUT, CPT, and ACPT-CDM models with the remaining power B~N(0.5,0.25) and arrival time t~N(5,9).

[0253] As shown in Table 3, in interval 1, where the arrival time is early, and interval 2, where the remaining battery power is high, the driver's decision is the same in the EUT, CPT, and ACPT-CDM models. They all choose Option 2 (charging during the normal period). In interval 3, where the arrival time is late and the remaining battery power is low, the driver's charging decision does not change in the EUT model. However, in the CPT model, the driver's decision changes at time 4.7, and Option 3 (charging across different periods) is chosen. In the ACPT-CDM model, the driver's decision changes at time 4.4, and Option 3 is chosen.

[0254] Table 3 Model comparison results

[0255]

[0256] Figure 9(a) is a schematic diagram of the owner's charging decision under the AUT model, Figure 9(b) is a schematic diagram of the owner's charging decision under the CPT model, and Figure 9(c) is a schematic diagram of the owner's charging decision under the ACPT-CDM model.

[0257] The decision-making results show that charging decisions differ under the three models when arrival time is late and remaining power is relatively low. In ACPT-CDM, the later the arrival time, the greater the driver's psychological anxiety. Even in the face of loss, the driver will choose to charge promptly. The AUT model, however, shows that the rational driver's charging decision is always option 2, charging during normal hours. This shows that the AUT model assumes that the driver is rational and ignores the influence of psychological factors. CPT and ACPT-CDM analyze the driver's bounded rationality, but CPT does not analyze the driver's PPV. The ACPT-CDM model incorporates the driver's PPV and risk attitude into its analysis, which can better reflect people's irrational decision-making behavior.

[0258] (2) Sensitivity analysis under different psychological safety levels

[0259] The peak-to-flat electricity price multiplier is set to three times. Under the conditions of remaining power B~N(0.5,0.25) and arrival time t~N(5,9), conservative car owners set the PSP under ACPT-CDM to 0.3, 0.5, and 0.7 respectively, and compare the charging decision behaviors of operating car owners.

[0260] Table 4 shows that when the driver's arrival time falls within intervals 1 and 2, all three PSP drivers choose Option 2 (on-peak charging). When the driver arrives in interval 3, the charging time for commercial vehicle owners is concentrated around 1:30 PM. With late arrival times and low remaining battery life, the decision of the driver with a PSP of 0.3 changes at 4:80 PM, choosing Option 3 (inter-hour charging). The decision of the driver with a PSP of 0.5 changes at 4:60 PM, choosing Option 3. The decision of the driver with a PSP of 0.7 changes at 4:20 PM, choosing Option 3.

[0261] Sensitivity analysis results show that arrival time, remaining charge, and PSP all have significant impacts on drivers' charging decisions. A higher PSP correlates with a weaker sense of security. When drivers arrive late and their remaining charge falls far below their PSP, they experience battery anxiety and rush to recharge. Due to high electricity prices during peak hours, drivers analyze charging costs and desire to access electricity earlier, so they choose to charge early. However, due to low remaining charge, they may need to charge across different time periods. A lower PSP correlates with a stronger sense of security. When remaining charge falls below their PSP, drivers consider charging costs and postpone charging due to high electricity prices during peak hours.

[0262] Table 4 Decision-making results of different psychological safety power levels

[0263]

[0264] Heterogeneous PSP decision diagram is as follows Figure 10(a) 、 10(b) , as shown in Figures 10(c); wherein, Figure 10(a) is a schematic diagram of the charging decision of the car owner with a PSP of 0.3, Figure 10(b) is a schematic diagram of the charging decision of the car owner with a PSP of 0.5, and Figure 10(c) is a schematic diagram of the charging decision of the car owner with a PSP of 0.7.

[0265] (3) Cases of heterogeneous risk preferences

[0266] The peak-to-flat electricity price multiplier is set to three times, the owner's psychological safety power is 0.5, and under the conditions of remaining power B~N(0.5,0.25) and arrival time t~N(5,9), the risk factor μ is set to 0.2, 0.5, and 0.8 respectively in ACPT-CDM. The charging decision-making behaviors of operating vehicle owners with three different risk attitudes under this model and under the CPT model are compared.

[0267] Table 5 shows that when the driver's arrival time falls within intervals 1 and 2, both models choose Option 2 (on-peak charging). When the driver arrives in interval 3, under ACPT-CDM, the risk-averse driver (μ = 0.2) changes his decision at 4:80, choosing Option 3 (inter-hour charging); the conservative driver (μ = 0.5) changes his decision at 4:60, choosing Option 3; and the risk-seeking driver (μ = 0.8) changes his decision at 4:40, choosing Option 3. Under CPT, however, the driver consistently chooses Option 3 at 4:60, resulting in a blunted decision-making pattern. However, the ACPT-CDM model significantly expands the characterization of risk attitudes and improves sensitivity.

[0268] Table 5 Decision-making results of heterogeneous risk preferences

[0269]

[0270] Figure 11(a) is a schematic diagram of the charging decision of a risk-averse car owner, Figure 11(b) is a schematic diagram of the charging decision of a conservative car owner, Figure 11(c) is a schematic diagram of the charging decision of a risk-seeking car owner, and Figure 11(d) is a schematic diagram of the charging decision of a CPT model car owner.

[0271] As can be seen from Figures 11(a), 11(b), 11(c), and 11(d), when the psychological safety level is 0.5, risk-seeking car owners have a higher psychological perceived value. They are willing to take certain risks to obtain higher returns, so they will choose to charge in advance during the peak period with higher risks, even if the price will soon reach the normal period with lower prices; risk-averse car owners are more sensitive to losses and have a lower psychological perceived value. They tend to avoid risks, so they will postpone charging as much as possible if the remaining power can support it; conservative car owners, on the one hand, worry about the losses of charging during peak periods, and on the other hand, expect higher returns. They are more willing to find a relatively moderate time point to choose cross-period charging.

[0272] Take Zhang San as an example (see Figure 8 ), Zhang San arrives at 4:4, and the remaining power is 0.3. Assuming he is a risk-seeking car owner, since the remaining power is low at this time, but his psychological perceived value is high, he will choose to charge in time and start the next stage of orders earlier with more sufficient power; assuming he is a conservative car owner with low psychological perceived value, he will make a comprehensive assessment of the electricity price and power before choosing to charge; assuming he is a risk-averse car owner, he pays more attention to losses. At this time, when the remaining power is not particularly low, the high electricity price during peak hours will make him charge later.

[0273] In the actual operation of electric vehicles, heterogeneous owners have different psychological safety levels. Therefore, changes in psychological safety levels will affect the decisions of owners with different risk preferences. When the psychological safety level is low, risk-seeking owners have a higher perceived value, so their overall choice to charge early shifts earlier. When the psychological safety level is high, risk-averse owners have a lower perceived value, so their overall choice to postpone charging shifts later.

[0274] The case results show that under different psychological safety levels, car owners with different risk preferences will have different charging decisions. Therefore, when analyzing car owners' charging decision-making behavior, the influence of risk attitude and psychological perceived value should be comprehensively analyzed.

[0275] It should be noted that the embodiments of the present invention can be implemented by hardware, software, or a combination of software and hardware. The hardware portion can be implemented using dedicated logic; the software portion can be stored in a memory and executed by an appropriate instruction execution system, such as a microprocessor or dedicated design hardware. Those skilled in the art will appreciate that the above-mentioned devices and methods can be implemented using computer-executable instructions and / or contained in processor control code, for example, such as a carrier medium such as a disk, CD or DVD-ROM, a programmable memory such as a read-only memory (firmware), or a data carrier such as an optical or electronic signal carrier. The devices and modules of the present invention can be implemented by hardware circuits such as very large-scale integrated circuits or gate arrays, semiconductors such as logic chips, transistors, or programmable hardware devices such as field programmable gate arrays, programmable logic devices, etc., can also be implemented by software executed by various types of processors, or can be implemented by a combination of the above-mentioned hardware circuits and software, such as firmware.

[0276] The above description is only a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any modifications, equivalent substitutions and improvements made by any technician familiar with this technical field within the technical scope disclosed by the present invention and within the spirit and principles of the present invention should be covered by the scope of protection of the present invention.

Claims

1. A real-time charging decision method for electric vehicles, characterized in that: The electric vehicle real-time charging decision-making method includes: First, a comprehensive analysis of arrival time, remaining battery life, and psychological safety level factors was conducted to construct a charging demand model for electric vehicle owners under different peak and valley electricity prices. Second, a heterogeneous reference point model based on risk preference was established to address the differences in the psychological perceptions of arrival time and remaining battery life among electric vehicle owners. Finally, charging decisions were evaluated using the cumulative prospective benefits of electric vehicles as an evaluation indicator. The electric vehicle real-time charging decision-making method comprises the following steps: Step 1: Building a charging demand model: This combines objective factors with the owner's subjective perception, comprehensively analyzing the owner's psychological safety power (PSP), arrival time, and remaining power. This model is used to establish a charging demand model for operating vehicle owners under different peak and valley electricity prices, thereby determining a set of alternative charging decision options and a profit function. Step 2: Constructing an irrational decision model: Based on the irrationality of vehicle owners, we apply Cumulative Prospect Theory (CPT) to modeling vehicle owners' charging decisions. We then establish an improved ACPT-CDM model for electric vehicle charging decisions. The ACPT divides the uncertain decision-making process into an editing phase and an evaluation phase. In the editing phase, a reference point is determined, and the results are converted into gains or losses relative to the reference point. In the evaluation phase, a value function and a weight function are used to analyze and evaluate vehicle owners' charging decisions, ultimately selecting the optimal charging solution. Step 3: Constructing a Heterogeneous Reference Point Model Based on Risk Preference: During the ACPT-CDM editing phase, based on the set of alternative options determined by the charging demand model, we analyzed the differences in the PPV (Psychologically Perceived Value) of electric vehicle owners. By combining individual risk attitudes with psychologically perceived reference points, we improved the reference point of the cumulative prospect theory and established a heterogeneous reference point model based on risk preference. The construction of the heterogeneous reference point model based on risk preference in step 3 includes: (1) Heterogeneous reference point model based on risk preference Under the expected utility theory (EUT) model, vehicle owners make charging decisions by comparing the expected value of benefits under various charging decisions. They tend to favor charging decisions with larger expected values. During the CPT calculation process, irrational factors are analyzed. The choice of reference point affects the calculation of the cumulative prospect value, and the reference point in the editing stage is improved. 1) Psychological Perception Reference Point Assume I0 is the psychological reference point for heterogeneous car owners to choose charging solutions, and the expected profit value of the car owner is: I0=EU=∑(Y i ,P i ); Among them, U(x) is the expected utility function, Y i represents the benefit of each charging scheme; P i Indicates the corresponding probability; i = 1, 2, 3; The PPV deviation is ΔI. When the total profit I(x) gained in one day is greater than the expected profit I0, the car owner obtains a positive psychological perception ΔI. When the total profit I(x) gained in one day is less than the expected profit I0, the car owner obtains a negative psychological perception ΔI. ΔI=I(x)-I0; (2) Constructing a psychological perception reference point model based on risk preference μ (0 < μ < 1) is the risk preference factor. In the decision-making process of the car owner, if 0 < μ < 0.5, the car owner is risk-averse; if 0.5 < μ < 1, the car owner is risk-seeking; and if μ = 0.5, the car owner is risk-neutral. When determining the reference point, a heterogeneous reference point model based on risk preferences was established for different PPVs of car owners. The maximum and minimum values of psychological security were used as threshold points for each region. A risk factor μ was also introduced. The heterogeneous reference point was obtained by multiplying the risk factor μ with the difference between the maximum and minimum values and adding the minimum value. When decision makers hold different risk preferences, the reference point also fluctuates within the extreme value range. I=I min +(I max -AND min )μ。 2. The real-time charging decision-making method for electric vehicles according to claim 1, characterized in that: The construction of the charging demand model in step 1 includes: The initial state of charge B is a key factor affecting the charging time of electric vehicles, as shown in the following formula: Where C is the rated capacity of the electric vehicle battery; B is the initial state of charge of the electric vehicle when it arrives at the station (i.e., the remaining power); B lea The battery state of charge when the electric vehicle is fully charged; P EV is the charging power of the electric vehicle; θ char is the charging efficiency of electric vehicles; T char is the total charging time; The start charging time t of an operating vehicle depends on many factors, including the type of vehicle and the driver's personal behavior. Previous studies approximated the distribution of the end time of the vehicle's last trip to a normal distribution through maximum likelihood estimation and regarded it as the start charging time as follows: Among them, μ s , σ s are the expectation and standard deviation of t, respectively.

3. The real-time charging decision-making method for electric vehicles according to claim 1, characterized in that: The construction of the charging demand model in step 1 further includes: Classification is done by setting three charging states: State 1: Arriving during peak hours, and the charging time is less than the remaining time during the peak period, the owner has two options: charge during the peak period, or wait until the normal period to charge during the normal period; State 2: Arriving during peak hours, and the charging time is greater than the remaining time during the peak period, the owner has two options: charge during the off-peak period or wait until the off-peak period to charge during the off-peak period; State 3: Arriving during normal hours, the owner charges directly during normal hours; In the charging status analysis, the relationship between arrival time, remaining power, charging time and charging status is comprehensively analyzed; where (0, t0) is the peak period, (t0, t max ) is a normal period; When the remaining power is the minimum power B min , charge it to the maximum capacity B max Requires T max , this is the longest charging time, and when the remaining power is B max When the owner arrives at time 0, the maximum charging cut-off time is T max , the minimum charging cut-off time is 0 straight line Indicates the corresponding relationship between the remaining power and the charging cut-off time when time 0 is reached; similarly, Corresponding car owner t0-T max The corresponding relationship between the remaining power when the time arrives and the charging cut-off time; Assuming time t0 is the start time of the normal period, time t0 is taken as the critical point of charging cut-off time; when the time t∈(0,t0-T max )hour, Towards Moving, the straight line cluster forms a parallelogram area a0b0c0d0 under the limitation of the remaining power range. In the area, the charging decision is state 1, charging during peak hours or normal hours; when reaching time t∈(t0-T max ,t0), Towards Move to form a parallelogram area c0d0e0f0, in which state 1 and state 2 exist at the same time, and division is performed when a mixture of charging states occurs; l mn is a dynamically changing straight line segment, representing the charging cut-off curve under different arrival times. and The equations are: l mn With straight line t end = t0 has an intersection point K, with point K as the dividing point. The decisions of the car owners above point K are in state 1, and the decisions of the car owners below point K are in state 2; Converting the charging cut-off time to the arrival time, the equation of the straight line segment is: When the car owner M When the time arrives, the remaining power is higher than B M State 1, when the remaining power is lower than B M When is state 2; similarly, at any time, the two mixed states in the time period are separated by a straight line segment; Modeling is done for the vehicle owner to avoid the evening peak; when the vehicle owner arrives late, the vehicle owner charges the battery to B max It takes a long time, and most car owners charge the battery to the psychological safety level B saf Then leave the charging station; The equation of the line segment is: Car owner t N For example, if the remaining power is higher than B N At t max Charge the remaining power to B saf ; If the remaining power is lower than B N , the owner charges to B saf Some peak operating periods will be missed; Three charging schemes are set for the three charging states: Charging only during peak hours: The owner arrives at a peak hour, and the charging time is less than the remaining time during the peak hour; Charging only during normal times: The owner arrives during peak hours, and the charging time is less than the remaining time during peak hours, or the owner arrives during normal times; Charging only during off-peak hours: The owner arrives during peak hours, but the charging time is greater than the remaining time during the peak hours; The owner's income is the measurement indicator. The income function of electric vehicles operating under the three charging schemes is defined as follows: Y = E-J1-J2; Among them, Y is the actual revenue, E is the total operating revenue for the whole day, J1 is the charging fee, and J2 is the customer loss; (1) The benefits of operating vehicle owners choosing to charge during peak hours: Y1=p1*T1*q1+p2*T2*q2-T char *p1*q1 -(T char *P EV *θ char ) / 1000*S1; (2) The benefits of operating vehicle owners choosing to charge during normal hours are: <h2 style=";text-align:left;direction:ltr">Y2 = p1 * T1 * q1 + p2 * T2 * q2 - T<h2 style=";text-align:left;direction:ltr"> char <h2 style=";text-align:left;direction:ltr"> *p2*q2 -(T char *P EV *θ char ) / 1000*S2; (3) The third benefit of operating vehicle owners choosing to charge across different time periods: Y3=p1*T1*q1+p2*T2*q2-T1*p1*q1 -(T1*P EV *θ char ) / 1000*S1-T2*p2*q2 -(T2*P EV *i char ) / 1000*S1; Among them, p1 is the passenger flow during peak hours, and p2 is the passenger flow during normal hours; T1 is the charging time during peak hours, and T2 is the charging time during normal hours; q1 is the average customer price during peak hours, and q2 is the average customer price during normal hours; S1 is the electricity price during peak hours, and S2 is the electricity price during normal hours.

4. The real-time charging decision-making method for electric vehicles according to claim 1, characterized in that: The construction of the risk preference-based heterogeneous reference point model in step 3 further includes prospect value calculation, which includes: (1) Value Function Among them, α and β represent the owner's sensitivity to risk, λ is the loss aversion coefficient, and α﹦β﹦0.88, λ﹦2.25; (2) Weight function When car owners face the benefits: When a car owner faces losses: Among them, γ﹦0.61, δ﹦0.69; CPT decision weight function π i + and π i － The definition is as follows: Among them, π ij + is a positive cumulative decision weight function, that is, the cumulative decision weight function when the decision maker faces "benefits"; π ij － is a negative cumulative decision weight function, that is, the cumulative decision weight function when the decision maker faces "loss"; n is the possible result when the owner feels the benefit under each charging scheme; m is the possible result when the owner feels the loss under each charging scheme; ω + and ω － is a strictly increasing function that satisfies: oh + (0)=ω - (0)=0; oh + (1)=ω - (1)=1; (3) Calculation of cumulative prospect value Among them, CPV j + is the positive cumulative prospect value under scenario j, CPV j － is the negative cumulative prospect value under scenario j; CPV j is the actual comprehensive cumulative prospect value, j = 1, 2, 3, CPV1, CPV2 and CPV3 are the comprehensive cumulative prospect values of plan 1, plan 2 and plan 3 respectively; when making charging decisions, electric vehicle owners usually tend to choose the charging plan with the largest comprehensive cumulative prospect value as the optimal charging option.

5. An electric vehicle real-time charging decision system using the electric vehicle real-time charging decision method according to any one of claims 1 to 4, characterized in that: The electric vehicle real-time charging decision system includes: A charging demand model building module is used to comprehensively analyze arrival time, remaining power, and psychological safety power factors to build a charging demand model for electric vehicle owners under different peak and valley electricity prices; A heterogeneous reference point model building module is used to establish a heterogeneous reference point model based on risk preference and improved cumulative prospect theory, taking into account the differences in the psychological perceptions of vehicle owners regarding arrival time and remaining battery power. The charging decision evaluation module is used to evaluate charging decisions based on the cumulative prospective benefits of operating vehicles.

6. A computer device, characterized in that: The computer device includes a memory and a processor, the memory stores a computer program, and when the computer program is executed by the processor, the processor executes the steps of the electric vehicle real-time charging decision method according to any one of claims 1 to 4.

7. A computer-readable storage medium storing a computer program, wherein when the computer program is executed by a processor, the processor executes the steps of the electric vehicle real-time charging decision-making method according to any one of claims 1 to 4.

Images