An electric vehicle daily power tracking optimization method and system

Abstract

The application discloses an electric vehicle daily power tracking optimization method and system, relates to the electric vehicle regulation and control technical field, and the method comprises the following steps: constructing an initial correlation fluctuation relationship between discharge willingness and the feasible region of an electric vehicle according to historical electric vehicle discharge response data and historical discharge compensation data based on the Weber-Fechner law; quantifying the degree of uncertainty in the initial correlation fluctuation relationship based on the information gap decision theory, and constructing a dynamic discharge willingness fluctuation relationship; constructing an optimization target based on the maximum tracking degree of a day-ahead plan, constructing a daily rolling optimization model according to the dynamic discharge willingness fluctuation relationship and the optimization target; according to user optimization requirements, calling a robust optimization model or an opportunity optimization model, acquiring current power grid operation data and a day-ahead plan, inputting the data into the robust optimization model or the opportunity optimization model, and outputting an electric vehicle dispatching strategy. The application has the beneficial effects that the regulation and control accuracy and reliability of the electric vehicle are improved.

Description

Technical Field

[0001] This application relates to the field of electric vehicle control technology, and in particular to a method and system for optimizing intraday power tracking of electric vehicles. Background Technology

[0002] In existing electric vehicle (EV) aggregation and control technologies, aggregators typically aggregate EV resources eligible for V2G (Vehicle-to-Grid) operation and formulate a total power plan during the day-ahead phase. During the intraday phase, they track the day-ahead plan based on ultra-short-term forecasts and deaggregate the total power into specific control commands for each EV for execution. However, influenced by users' subjective intentions (such as subsidy price sensitivity and changes in travel demand), the actual discharge response rate of EVs (i.e., the proportion of users willing to participate in grid regulation) often deviates significantly between the day-ahead and intraday phases. This uncertainty may cause intraday real-time control commands to fail (e.g., the planned discharge amount cannot be achieved), thereby affecting the power balance of the grid.

[0003] To address this issue, many related technologies employ deterministic models or traditional uncertainty handling methods: deterministic models assume fixed user response behavior and do not consider actual fluctuations in intention, resulting in poor adaptability of optimization strategies in real-world scenarios; stochastic optimization methods require pre-setting specific probability distributions of response rates, but user behavior is highly random and difficult to model accurately, easily leading to overly risky decisions; robust optimization, while ensuring the feasibility of decisions within uncertain sets, sacrifices economy to cover worst-case scenarios, resulting in excessively high control costs and limiting the control accuracy and reliability of intraday de-aggregation models. Summary of the Invention

[0004] This application addresses the issues of poor accuracy and reliability in electric vehicle aggregation control in related technologies by providing an intraday power point tracking optimization method and system for electric vehicles. It utilizes the Weber-Fechner law and information gap decision theory to quantify the dynamic fluctuations and uncertainties in the discharge response intentions of electric vehicle users. Based on benefit-cost robustness deviation constraints and benefit-cost opportunity deviation constraints, it ensures that the control strategy not only meets the scheduling needs of electric vehicle aggregators but also adapts to the dynamic changes in the discharge response intentions of electric vehicle users, thereby improving the accuracy and reliability of electric vehicle control.

[0005] To achieve the aforementioned technical objectives, this application provides a technical solution: an intraday power point tracking optimization method for electric vehicles, comprising the following steps: constructing an initial correlation fluctuation relationship between discharge intention and the feasible region of electric vehicles based on Weber-Fechner's law, historical electric vehicle discharge response data, and historical discharge compensation data; quantifying the uncertainty in the initial correlation fluctuation relationship based on information gap decision theory, and constructing a dynamic discharge intention fluctuation relationship; constructing an optimization objective based on the maximum tracking degree of the day-ahead plan, and constructing an intraday rolling optimization model based on the dynamic discharge intention fluctuation relationship and the optimization objective; constructing a robust optimization model based on benefit-cost robust deviation constraints and the intraday rolling optimization model; constructing an opportunistic optimization model based on benefit-cost opportunity deviation constraints and the intraday rolling optimization model; retrieving the robust optimization model or the opportunistic optimization model according to user optimization requirements, obtaining current grid operation data and the day-ahead plan, inputting them into the robust optimization model or the opportunistic optimization model, and outputting an electric vehicle scheduling strategy.

[0006] Furthermore, the initial correlation fluctuation relationship between discharge intention and the feasible region of electric vehicles based on the Weber-Fechner law and historical electric vehicle discharge response data and historical discharge compensation data includes: constructing a first correlation fluctuation relationship between discharge response rate and discharge compensation price based on the Weber-Fechner law; constructing the feasible region of electric vehicles based on the virtual battery model, electric vehicle hardware parameters, user behavior data, and the first correlation fluctuation relationship; and constructing an initial correlation fluctuation relationship using the first correlation fluctuation relationship and the feasible region of electric vehicles.

[0007] Furthermore, the step of quantifying the uncertainty in the initial correlation fluctuation relationship based on information gap decision theory and constructing a dynamic discharge intention fluctuation relationship includes: calculating the uncertainty of the discharge response rate based on the predicted value of the discharge response rate and the upper and lower limits of the discharge response rate in the initial correlation fluctuation relationship using the envelope model in information gap decision theory; and constructing a dynamic discharge intention fluctuation relationship using the uncertainty and the initial correlation fluctuation relationship.

[0008] Furthermore, the optimization target is constructed based on the maximum tracking degree of the daily plan, and the intraday rolling optimization model is constructed according to the dynamic discharge intention fluctuation relationship and the optimization target, including: obtaining the tracking degree by the difference between the planned power of the electric vehicle aggregator in the daily stage and the total charging and discharging power of the electric vehicle.

[0009] Furthermore, the construction of the first correlation fluctuation relationship between discharge response rate and discharge compensation price based on the Weber-Fechner law includes: constructing a dynamic fluctuation relationship of maximum discharge response rate based on historical maximum discharge response rate data corresponding to each discharge compensation price; constructing a dynamic fluctuation relationship of minimum discharge response rate based on historical minimum discharge response rate data corresponding to each discharge compensation price; constructing a dynamic fluctuation relationship of benchmark response rate based on historical average discharge response rate data corresponding to each discharge compensation price; and constructing the first correlation fluctuation relationship using the dynamic fluctuation relationship of maximum discharge response rate, the dynamic fluctuation relationship of minimum discharge response rate, and the dynamic fluctuation relationship of benchmark response rate.

[0010] Furthermore, the calculation of the uncertainty of the discharge response rate based on the predicted discharge response rate and the upper and lower limits of the discharge response rate in the envelope model of the information gap decision theory includes: obtaining the maximum value, minimum value, and predicted value of the discharge response rate under the corresponding discharge compensation price according to the dynamic fluctuation relationship of the maximum discharge response rate, the dynamic fluctuation relationship of the minimum discharge response rate, and the dynamic fluctuation relationship of the benchmark response rate; and obtaining the uncertainty under the discharge compensation price according to the relative fluctuation ratio of the predicted discharge response rate value with respect to the maximum and minimum discharge response rate values.

[0011] Furthermore, the construction of the robust optimization model based on the benefit-cost robust deviation constraint and the intraday rolling optimization model includes: taking the maximization of uncertainty as the robust optimization objective, constructing the optimization layer of the robust optimization model with the optimization objective of the intraday rolling optimization model and the robust optimization objective, and constructing the constraint layer of the robust optimization model with the constraint conditions of the intraday rolling optimization model and the benefit-cost robust deviation constraint.

[0012] Furthermore, the construction of the opportunity optimization model based on the benefit-cost-opportunity deviation constraint and the intraday rolling optimization model includes: using the minimization of uncertainty as the opportunity optimization objective, constructing the optimization layer of the opportunity optimization model with the optimization objective of the intraday rolling optimization model and the opportunity optimization objective, and constructing the constraint layer of the opportunity optimization model with the constraints of the intraday rolling optimization model and the benefit-cost-opportunity deviation constraint.

[0013] Furthermore, the benefit-cost robustness deviation constraint is the maximum increase in cost ratio of the benefit-cost of the scheduling strategy relative to the benefit-cost corresponding to the predicted discharge response rate; the benefit-cost opportunity deviation constraint is the maximum decrease in cost ratio of the benefit-cost of the scheduling strategy relative to the benefit-cost corresponding to the predicted discharge response rate.

[0014] Another technical solution provided in this application is an intraday power tracking optimization system for electric vehicles, used to implement the method described above, comprising: a correlation analysis unit, used to construct an initial correlation fluctuation relationship between discharge intention and the feasible region of electric vehicles based on historical electric vehicle discharge response data and historical discharge compensation data according to the Weber-Fechner law, and to quantify the uncertainty in the initial correlation fluctuation relationship based on information gap decision theory, and to construct a dynamic discharge intention fluctuation relationship; an intraday optimization unit, used to construct an optimization objective based on the maximum tracking degree of the daily plan, and to construct an intraday rolling optimization model according to the dynamic discharge intention fluctuation relationship and the optimization objective; a robust optimization unit, used to construct a robust optimization model according to the benefit-cost robust deviation constraint and the intraday rolling optimization model; an opportunity optimization unit, used to construct an opportunity optimization model according to the benefit-cost opportunity deviation constraint and the intraday rolling optimization model; and a scheduling unit, used to retrieve the robust optimization model or the opportunity optimization model according to the user's optimization requirements, obtain the current power grid operation data and the daily plan, input them into the robust optimization model or the opportunity optimization model, and output an electric vehicle scheduling strategy.

[0015] The beneficial effects of this application are as follows: 1. First, by combining the Weber-Fechner law with historical data, an initial correlation fluctuation relationship between user discharge intention and the feasible region of electric vehicles is constructed, quantifying the changes in user sensitivity to compensation prices. Then, through information gap decision theory, the uncertainty in the initial correlation fluctuation relationship is quantified, controlling the impact of uncertainty caused by random changes in user intention in electric vehicle discharge behavior on actual regulation. This ensures that while incorporating users' subjective discharge intention into the regulation strategy to enhance its adaptability, the regulation's ability to resist interference from the uncertainty of subjective intention is guaranteed, improving the accuracy and stability of regulation. Furthermore, an optimization objective is constructed based on the maximum tracking degree of the day-ahead plan, and a dual-model synergy is implemented based on different risk preferences to ensure that the scheduling strategy adapts to the actual needs of users and the grid operation requirements.

[0016] 2. By calculating the relative fluctuation ratio of the predicted discharge response rate to the extreme value, the discrete fluctuation range under different discharge compensation prices is converted into uncertainty. This ensures that the uncertainty covers the possible range of deviation between the actual value of the user's discharge response rate and the predicted value, thereby ensuring that the scheduling strategy can adapt to changes in the user's discharge intention within the possible range and ensuring the anti-interference capability and robustness of the scheduling strategy. Attached Figure Description

[0017] Figure 1 This is a schematic flowchart illustrating one embodiment of the intraday power tracking optimization method for electric vehicles according to this application.

[0018] Figure 2 This is a schematic diagram illustrating the user discharge response rate quantification under one embodiment of the intraday power tracking optimization method for electric vehicles according to this application. Detailed Implementation

[0019] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description of this application is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely one preferred embodiment of this application and are only used to explain this application. They do not limit the scope of protection of this application. All other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0020] like Figure 1 As shown, a method for optimizing intraday power point tracking in electric vehicles includes the following steps: Based on the Weber-Fechner law, a correlation fluctuation relationship between discharge intention and the initial feasible region of electric vehicles is constructed using historical electric vehicle discharge response data and historical discharge compensation data. Based on the information gap decision theory, the degree of uncertainty in the initial correlation fluctuation relationship is quantified, and a dynamic discharge intention fluctuation relationship is constructed. An optimization objective is constructed based on the maximum tracking level of the daily plan, and an intraday rolling optimization model is constructed based on the dynamic discharge intention fluctuation relationship and the optimization objective. A robust optimization model is constructed based on the benefit-cost robust deviation constraint and the intraday rolling optimization model; An opportunity optimization model is constructed based on the benefit-cost-opportunity deviation constraint and the intraday rolling optimization model. Based on the user's optimization requirements, a robust optimization model or an opportunistic optimization model is invoked to obtain the current power grid operation data and the day-ahead plan, which are then input into the robust optimization model or opportunistic optimization model to output the electric vehicle dispatch strategy.

[0021] In this embodiment, the initial correlation fluctuation relationship between user discharge intention and the feasible region of electric vehicles is first constructed using the Weber-Fechner law combined with historical data, quantifying the user's sensitivity to changes in compensation prices. Then, information gap decision theory is used to quantify the uncertainty in the initial correlation fluctuation relationship, controlling the impact of uncertainty caused by random changes in user intention in electric vehicle discharge behavior on actual regulation. This ensures that while incorporating users' subjective discharge intention into the regulation strategy to enhance its adaptability, the regulation's ability to resist interference from the uncertainty of subjective intention is guaranteed, improving the accuracy and stability of regulation. Furthermore, an optimization objective is constructed based on the maximum tracking degree of the day-ahead plan, and a dual-model synergy is implemented based on different risk preferences to ensure that the scheduling strategy adapts to the actual needs of users and the grid operation requirements.

[0022] Specifically, based on the Weber-Fechner law, the initial correlation fluctuation relationship between discharge intention and the feasible region of electric vehicles is constructed using historical electric vehicle discharge response data and historical discharge compensation data. This includes: The first correlation fluctuation relationship between discharge response rate and discharge compensation price is constructed based on the Weber-Fechner law. Based on the virtual battery model, the feasible domain of electric vehicles is constructed according to the hardware parameters of electric vehicles, user behavior data, and the first correlation fluctuation relationship. An initial correlation fluctuation relationship is constructed based on the first correlation fluctuation relationship and the feasible region of electric vehicles.

[0023] Electric vehicles (EVs) share functional similarities with energy storage systems as flexible resources, but their adaptability is easily constrained by multiple uncertainties on the user side, including variables such as initial grid-connected power levels, planned off-grid energy storage demand, and actual grid-connected duration. It is worth noting that EVs also possess significant social attributes; the owner's subjective willingness to participate has a considerable impact on their adaptability.

[0024] As a key intermediary connecting end-users and the regional power grid, electric vehicle aggregators (EVAs) incentivize sufficient EVs to participate in discharge services through economic compensation mechanisms. The Weber-Fechner law reveals a non-linear mapping relationship between psychological perception and physical stimulus. Therefore, based on the Weber-Fechner law, the first correlation fluctuation relationship can be constructed by quantifying the change in user response rate with discharge compensation price as follows: ; in, This indicates the price of discharge compensation for electric vehicles. The discharge compensation price is indicated as Time The discharge response rate of an electric vehicle Indicates the first User response coefficient for electric vehicles Indicates the first The user response constant of a vehicle.

[0025] It is understood that in this embodiment, the discharge compensation price is directly retrieved data and is not adjusted.

[0026] like Figure 2 As shown, the first correlation fluctuation relationship between discharge responsivity and discharge compensation price, constructed based on the Weber-Fechner law, includes: The first correlation fluctuation relationship between discharge responsivity and discharge compensation price, based on the Weber-Fechner law, includes: A dynamic fluctuation relationship of the maximum discharge response rate is constructed based on the historical maximum discharge response rate data corresponding to each discharge compensation price. A dynamic fluctuation relationship of minimum discharge response rate is constructed based on historical minimum discharge response rate data corresponding to each discharge compensation price. A dynamic fluctuation relationship of the benchmark response rate is constructed based on the historical average discharge response rate data corresponding to each discharge compensation price. The first correlation fluctuation relationship is constructed based on the dynamic fluctuation relationship of the maximum discharge response rate, the dynamic fluctuation relationship of the minimum discharge response rate, and the dynamic fluctuation relationship of the reference response rate.

[0027] By utilizing historical maximum, minimum, and average discharge response rate data corresponding to various discharge compensation prices, a segmented relationship of discharge response rates is constructed. This avoids the mutual influence of mixed samples of different response rates. Based on obtaining the majority of user response patterns, some extreme characteristics of user responses are retained, so that the first correlation fluctuation relationship covers user response behavior and improves the accuracy of predicting user response intentions.

[0028] When a user's willingness to discharge changes, the feasible region of an electric vehicle also changes. A virtual battery model is used to package the electric vehicle into a virtual, large battery, facilitating overall calculations. Based on the virtual battery model, the feasible region of the electric vehicle is constructed according to the electric vehicle's hardware parameters, user behavior data, and the first correlation fluctuation relationship, including: Build a virtual battery model for electric vehicles; Construct charging pile constraints based on the charging pile parameters of electric vehicles; Battery constraints are constructed based on the physical parameters of the electric vehicle's battery. Time and energy constraints are constructed based on historical user behavior data; Construct intention constraints based on the first associated fluctuation relationship; The feasible domain of electric vehicles is constructed using constraints such as charging piles, batteries, time, energy, and willingness.

[0029] Constructing the feasible domain of electric vehicles based on their hardware parameters, user behavior data, and the first correlation fluctuation relationship can ensure that the regulation of electric vehicles does not deviate from the actual charging and discharging conditions of electric vehicles, thereby improving the accuracy of electric vehicle regulation.

[0030] In this embodiment, the feasible region of the electric vehicle is constructed using constraints such as charging piles, batteries, time, energy, and willingness: ; ; ; ; ; ; ; in, Indicates the first electric vehicles The maximum permissible energy at any given moment; Indicates the first Electric vehicles at the time of grid connection The initial energy; Indicates the first The maximum charging / discharging power of a vehicle; Indicates the first The moment when the electric vehicle connects to the grid; Indicates a time scale; Indicates the first The rated maximum capacity of the battery of an electric vehicle; Indicates the first After the electric vehicles are connected to the grid Minimum energy intermediate constraint at time step; Indicates the first The discharge response rate of an electric vehicle; Indicates the first The minimum charging / discharging power of an electric vehicle; Indicates the first The minimum rated capacity of the battery for an electric vehicle; Indicates the first After the electric vehicle was taken off the grid Minimum energy intermediate constraint at time step; Indicates the first The expected remaining energy of an electric vehicle when it leaves the grid; Indicates the first The time when an electric vehicle leaves the network; Indicates the first electric vehicles The minimum allowable energy at any given moment; Indicates the first electric vehicles Maximum permissible power at any given time; Indicates the first electric vehicles The maximum permissible energy at any given moment; Indicates the first electric vehicles The minimum permissible energy; Indicates the first electric vehicles Minimum allowable power at any given time; Indicates the first electric vehicles The minimum allowable energy at any given moment; Indicates the first electric vehicles Maximum permissible energy; Indicates the first electric vehicles Time's up The amount of energy change between moments; Indicates the first electric vehicles The state of charge at any given moment; Indicates the first electric vehicles The state of charge at any given moment.

[0031] Then, the feasible domain of electric vehicles is associated with the first correlation fluctuation relationship to obtain the changes in the feasible domain of electric vehicles as user intentions change, and an initial correlation fluctuation relationship is constructed.

[0032] Traditional Weber-Fechner consumer psychology models characterize the discharge control response rate of electric vehicle (EV) users as a deterministic parameter. However, in actual response processes, users' response levels to specific discharge subsidy prices exhibit significant fluctuations. Information Gap Decision Theory (IGDT) is employed to quantify this uncertainty. Compared to stochastic optimization methods, IGDT demonstrates significant advantages in handling non-probabilistic uncertainty optimization control problems. Furthermore, this theory considers the expected benefits of system operation during the optimization process, achieving a synergistic optimization of efficiency and robustness compared to traditional robust optimization methods that only emphasize robustness.

[0033] In nonprobabilistic uncertainty description methods within information gap decision theory, the weighted mean square error model characterizes uncertainty by analyzing the prediction error generation mechanism, while the ellipsoidal model represents uncertainty based on the differences between variables. However, both types of models rely on historical data of uncertain parameters to calculate the covariance matrix, leading to significant data dependence and computational complexity in the uncertainty quantification process.

[0034] Therefore, in this embodiment, the envelope model is used as the uncertainty characterization method. It eliminates the need to calculate the covariance matrix and defines the uncertainty set solely through the predicted response rate and the upper and lower limits of the piecewise model. This solves the problem of limited response data and unknown probability distribution among electric vehicle users in practice. Based on information gap decision theory, the uncertainty in the initial correlation fluctuation relationship is quantified, and the dynamic discharge intention fluctuation relationship is constructed, including: The uncertainty of the discharge response rate is calculated using the envelope model in the information gap decision theory based on the predicted value of the discharge response rate in the initial correlation fluctuation relationship and the upper and lower limits of the discharge response rate. A dynamic discharge intention fluctuation relationship is constructed based on uncertainty and initial correlation fluctuation relationship.

[0035] In this embodiment, the uncertainty of the discharge response rate is calculated using the envelope model in the information gap decision theory. The range with a probability distribution reflects the user's actual response behavior, reducing the probability that the scheduling strategy planning cannot be executed due to the dynamic influence of actual user intentions. This ensures that the scheduling strategy can be affected by changes in user intentions to a certain extent, thus guaranteeing the executability of the scheduling strategy.

[0036] Specifically, the uncertainty of the discharge response rate, calculated using the envelope model in information gap decision theory based on the predicted discharge response rate value and the upper and lower limits of the discharge response rate in the initial correlation fluctuation relationship, includes: Based on the dynamic fluctuation relationship of the maximum discharge response rate, the dynamic fluctuation relationship of the minimum discharge response rate, and the dynamic fluctuation relationship of the benchmark response rate, the maximum discharge response rate, the minimum discharge response rate, and the predicted discharge response rate under the corresponding discharge compensation price are obtained. The uncertainty under the discharge compensation price is obtained by measuring the relative fluctuation ratio of the predicted discharge response rate to the maximum and minimum discharge response rates.

[0037] By calculating the relative fluctuation ratio of the predicted discharge response rate to the extreme value, the discrete fluctuation range under different discharge compensation prices is converted into uncertainty. This ensures that the uncertainty covers the possible range of deviation between the actual value of the user's discharge response rate and the predicted value, thereby ensuring that the scheduling strategy can adapt to changes in the user's discharge intention within the possible range and ensuring the anti-interference capability and robustness of the scheduling strategy.

[0038] Specifically, the dynamic discharge intention fluctuation relationship is constructed based on uncertainty and initial correlation fluctuation relationship as follows: ; ; ; in, Indicates the first The maximum discharge response rate of an electric vehicle; Indicates the first The maximum user response coefficient for a single electric vehicle; This represents the minimum discharge compensation price at which the discharge response rate reaches its maximum value. This represents the saturation discharge compensation price when the discharge response rate is saturated. Indicates the first The maximum user response constant of a single electric vehicle; Indicates the first The minimum discharge response rate of an electric vehicle; Indicates the first Minimum user response coefficient for a single electric vehicle; This represents the minimum discharge compensation price at which the discharge response rate reaches its minimum value. Indicates the first Minimum user response constant for a vehicle; Represents the uncertainty of the discharge response rate; Indicates the first Predicted discharge response rate of electric vehicles; Indicates the first The discharge response rate of an electric vehicle is uncertain.

[0039] In this embodiment, a dynamic discharge intention fluctuation relationship is constructed using the uncertainty set of the electric vehicle discharge response rate.

[0040] Based on the maximum tracking level of the daily plan, an optimization objective is constructed. An intraday rolling optimization model is then built according to the dynamic discharge intention fluctuation relationship and the optimization objective, including: The tracking level is obtained by measuring the difference between the planned power output of electric vehicle aggregators at a given time and the total charging and discharging power of electric vehicles.

[0041] Specifically, the intraday rolling optimization model is as follows: ; st ; ; ; ; ; in, This indicates the control time domain for intraday rolling optimization; This indicates the planned power output of electric vehicle aggregators at the current stage; This represents the total number of electric vehicles; Indicates the first electric vehicles on a time scale The charging and discharging power; Indicates the first electric vehicles The charging and discharging power at any given moment; Indicates the first electric vehicles The charging power at any given time; This indicates the EV charging / discharging indicator, with 1 representing charging and 0 representing discharging. Indicates the first electric vehicles Discharge power at any given moment; Indicates the first electric vehicles Battery level at any given moment; Indicates the first electric vehicles Battery level at any given moment; Indicates the first electric vehicles The battery charge correction amount at any given time; st represents a constraint.

[0042] At the intraday operational level, due to the unpredictability of electric vehicle users' discharge response intentions in future periods, the optimal solution for global charging and discharging behavior is difficult to obtain directly. Intraday rolling optimization focuses on maximizing local benefits within the control time domain, with the goal of ensuring consistency between the optimization process and the day-ahead scheduling plan. Therefore, the difference between the planned power of the electric vehicle aggregator in the day-ahead stage and the total charging and discharging power of the electric vehicles is calculated, and the minimum difference is used as the maximum tracking degree to ensure that the final electric vehicle scheduling meets the day-ahead plan requirements.

[0043] In this embodiment, based on the differentiated risk response strategies of electric vehicle aggregators and their user optimization needs, robust optimization models that emphasize risk avoidance and opportunity capture are constructed respectively, based on benefit-cost robust deviation constraints and benefit-cost opportunity deviation constraints.

[0044] In some cases, robust optimization models are constructed based on benefit-cost robust bias constraints and intraday rolling optimization models, including: Using the maximization of uncertainty as the robust optimization objective, the optimization layer of the robust optimization model is constructed using the optimization objective of the intraday rolling optimization model and the robust optimization objective. The constraint layer of the robust optimization model is constructed using the constraints of the intraday rolling optimization model and the benefit-cost robust deviation constraint.

[0045] In this scenario, maximizing uncertainty is used as the outer optimization objective of the robust optimization model, while maximizing the tracking accuracy of the current day's schedule is used as the inner optimization objective, thus constructing a two-layer robust optimization model. During the optimization process, it is ensured that the scheduling strategy satisfies the feasible region of the electric vehicle, the cost-benefit robustness deviation constraints, and the upper and lower limits of the discharge response rate.

[0046] The benefit-cost robustness deviation constraint can be set directly based on the expected cost in the user's optimization requirements. In this case, if the benefit cost under the scheduling strategy meets the expected cost, the optimal scheduling strategy is considered to satisfy the benefit-cost robustness deviation constraint. Alternatively, the benefit-cost robustness deviation constraint can be set based on the expected deviation in the user's optimization requirements. In this case, the benefit cost corresponding to the discharge response rate of the scheduling strategy and the benefit cost corresponding to the predicted discharge response rate are calculated simultaneously. The deviation between the two is calculated to obtain the robust deviation. If the robust deviation is less than the expected deviation, the scheduling strategy is considered to satisfy the benefit-cost robustness deviation constraint.

[0047] In this embodiment, the maximum uncertainty under the discharge compensation price is obtained based on the maximum relative fluctuation ratio of the predicted discharge response rate value to the maximum discharge response rate value and the minimum discharge response rate value.

[0048] The relative fluctuation ratios of the predicted discharge response rate relative to the maximum discharge response rate and the relative fluctuation ratios of the predicted discharge response rate relative to the minimum discharge response rate are calculated separately. The maximum value of the two relative fluctuation ratios is taken as the maximum uncertainty. In the process of solving the robust optimization model, the maximum uncertainty is used as the initial value for iterative solution.

[0049] Based on the benefit-cost-opportunity deviation constraint and the intraday rolling optimization model, the opportunity optimization model is constructed as follows: Minimizing uncertainty is taken as the opportunity optimization objective. The optimization layer of the opportunity optimization model is constructed using the optimization objective of the intraday rolling optimization model and the opportunity optimization objective. The constraint layer of the opportunity optimization model is constructed using the constraints of the intraday rolling optimization model and the benefit-cost-opportunity deviation constraint.

[0050] At this point, minimizing uncertainty is used as the outer optimization objective of the opportunity optimization model, and maximizing the tracking degree of the daily plan is used as the inner optimization objective, thus constructing a two-layer optimization model for the opportunity optimization model. During the optimization solution process, it is ensured that the scheduling strategy satisfies the feasible region of electric vehicles, the benefit-cost-opportunity deviation constraints, and the upper and lower limits of the discharge response rate.

[0051] The benefit-cost opportunity deviation constraint can be set directly based on the expected cost in the user's optimization requirements. In this case, if the benefit cost under the scheduling strategy meets the expected cost, the scheduling strategy is considered to satisfy the benefit-cost opportunity deviation constraint. Alternatively, the benefit-cost opportunity deviation constraint can be set based on the expected deviation in the user's optimization requirements. In this case, the benefit cost corresponding to the discharge response rate of the scheduling strategy and the benefit cost corresponding to the predicted discharge response rate are calculated simultaneously, and the deviation between the two is calculated to obtain the opportunity deviation. If the robust deviation is less than the expected deviation, the scheduling strategy is considered to satisfy the benefit-cost opportunity deviation constraint.

[0052] In this embodiment, the minimum uncertainty under the discharge compensation price is obtained based on the minimum relative fluctuation ratio of the predicted discharge response rate value to the maximum and minimum discharge response rates.

[0053] The relative fluctuation ratios of the predicted discharge response rate relative to the maximum discharge response rate and the relative fluctuation ratios of the predicted discharge response rate relative to the minimum discharge response rate are calculated separately. The minimum of the two is taken as the minimum uncertainty. In the process of solving the opportunity optimization model, the minimum uncertainty is used as the initial value for iterative solution.

[0054] In other cases, the benefit-cost robustness deviation constraint and the benefit-cost opportunity deviation constraint are the proportion of increase and decrease in benefit-cost in the user's optimization requirements. The proportion of increase in benefit-cost constrains the maximum increase in the benefit-cost of the scheduling strategy relative to the benefit-cost corresponding to the predicted discharge response rate. The proportion of decrease in benefit-cost constrains the maximum decrease in the benefit-cost of the scheduling strategy relative to the benefit-cost corresponding to the predicted discharge response rate. This avoids over-controlling risk or over-pursuing cost, and ensures that the scheduling strategy has a certain degree of risk control and cost pursuit.

[0055] Therefore, based on user optimization requirements, a robust optimization model or an opportunistic optimization model is invoked to obtain current power grid operation data and day-ahead plans. These are then input into the robust optimization model or opportunistic optimization model to output electric vehicle dispatch strategies, including: The robust optimization model is retrieved based on the benefit-cost robustness deviation constraints in the user's optimization requirements; The opportunity optimization model is retrieved based on the benefit-cost-opportunity deviation constraint in the user's optimization requirements.

[0056] Based on the current power grid operation data, obtain the current electric vehicle hardware parameters and current discharge compensation data. Based on the day-ahead plan, obtain the planned power for the day-ahead stage. Based on the benefit-cost robustness deviation constraint or benefit-cost opportunity deviation constraint in the user's optimization requirements, call up the corresponding robust optimization model or opportunity optimization model to output the scheduling strategy.

[0057] In another scenario, based on user optimization requirements, a robust optimization model or opportunistic optimization model is invoked to obtain current grid operation data and day-ahead plans. This data is then input into the robust optimization model or opportunistic optimization model to output an electric vehicle dispatch strategy. Other examples include: Obtain historical discharge compensation data and historical scheduling data of users, and obtain the demand fluctuation relationship between the historical scheduling correction ratio and the change range of historical scheduling cost correction of users. Based on the robust optimization model, the first scheduling strategy and the first scheduling cost deviation are output according to the current power grid operation data and the day-ahead plan. Based on the opportunity optimization model, a second dispatching strategy and the second dispatching cost deviation are output according to the current power grid operation data and the day-ahead plan. Based on the demand fluctuation relationship, obtain the first scheduling correction ratio corresponding to the first scheduling cost deviation and the second scheduling correction ratio corresponding to the second scheduling cost deviation, and output the first scheduling strategy and / or the second scheduling strategy based on the comparison result of the first scheduling correction ratio and the second scheduling correction ratio.

[0058] It is understandable that the users here refer to electric vehicle aggregators. Considering that some electric vehicle aggregators cannot determine the benefit-cost robustness deviation constraint or the benefit-cost opportunity deviation constraint, a demand fluctuation relationship is constructed based on the historical discharge compensation data and scheduling data of the electric vehicle aggregators, relating the magnitude of scheduling cost correction changes to the user's scheduling correction ratio. This reflects the risk tolerance of the electric vehicle aggregators. Based on this, cost deviations under robust optimization and opportunity optimization are calculated separately and matched with the demand fluctuation relationship to select the first and second scheduling strategies. The time point at which the electric vehicle aggregator corrects its scheduling strategy is selected, and the scheduling costs corresponding to the scheduling strategies before and after correction are obtained. The difference between the scheduling costs before and after correction is taken as the magnitude of the scheduling cost correction. The cost caused by the scheduling change is also included in the scheduling cost corresponding to the corrected scheduling strategy.

[0059] Specifically, the output of the first scheduling policy and / or the second scheduling policy based on the comparison result of the first scheduling correction ratio and the second scheduling correction ratio includes: If the first scheduling correction ratio is greater than the second scheduling correction ratio, then the first scheduling strategy is output. If the second scheduling correction ratio is greater than the first scheduling correction ratio, then the second scheduling strategy is output. If the first scheduling correction ratio and the second scheduling correction ratio are the same, then the first scheduling policy and the second scheduling policy are output synchronously.

[0060] In this embodiment, the historical scheduling correction ratio is obtained based on the proportion of scheduling corrections corresponding to the change in scheduling cost correction in historical data to the total number of scheduling corrections. The first scheduling correction ratio is obtained by matching the change in scheduling cost correction in the relationship between the first scheduling cost deviation and demand fluctuation. The second scheduling correction ratio is obtained by matching the change in scheduling cost correction in the relationship between the second scheduling cost deviation and demand fluctuation.

[0061] As a second embodiment of this application, an intraday power tracking optimization system for electric vehicles includes: The correlation analysis unit is used to construct the initial correlation fluctuation relationship between discharge intention and the feasible domain of electric vehicles based on the Weber-Fechner law, historical electric vehicle discharge response data and historical discharge compensation data, and to quantify the uncertainty in the initial correlation fluctuation relationship based on the information gap decision theory to construct the dynamic discharge intention fluctuation relationship. The intraday optimization unit is used to construct optimization targets based on the maximum tracking degree of the daily plan, and to construct an intraday rolling optimization model based on the dynamic discharge intention fluctuation relationship and the optimization targets. Robust optimization unit, used to construct robust optimization model based on benefit-cost robust deviation constraints and intraday rolling optimization model; The opportunity optimization unit is used to construct an opportunity optimization model based on the benefit-cost-opportunity deviation constraint and the intraday rolling optimization model. The scheduling unit is used to retrieve the robust optimization model or the opportunistic optimization model according to the user's optimization requirements, obtain the current power grid operation data and the day-ahead plan, and input them into the robust optimization model or the opportunistic optimization model to output the electric vehicle scheduling strategy.

[0062] In this embodiment, the correlation analysis unit and the scheduling unit are connected to the electric vehicle aggregator's data storage system to obtain historical data. The intraday optimization unit is connected to the correlation analysis unit, the robust optimization unit, and the opportunity optimization unit, and the robust optimization unit and the opportunity optimization unit are connected to the scheduling unit.

[0063] The specific embodiments described above are preferred embodiments of the intraday power tracking optimization method and system for electric vehicles of this application, and are not intended to limit the specific scope of this application. The scope of this application includes but is not limited to the specific embodiments described above. All equivalent changes made in accordance with the shape and structure of this application are within the protection scope of this application.

Claims

1. A method for optimizing intraday power point tracking in electric vehicles, characterized in that: Includes the following steps: Based on the Weber-Fechner law, a correlation fluctuation relationship between discharge intention and the initial feasible region of electric vehicles is constructed using historical electric vehicle discharge response data and historical discharge compensation data. Based on the information gap decision theory, the degree of uncertainty in the initial correlation fluctuation relationship is quantified, and a dynamic discharge intention fluctuation relationship is constructed. An optimization objective is constructed based on the maximum tracking level of the daily plan, and an intraday rolling optimization model is constructed based on the dynamic discharge intention fluctuation relationship and the optimization objective. A robust optimization model is constructed based on the benefit-cost robust deviation constraint and the intraday rolling optimization model; An opportunity optimization model is constructed based on the benefit-cost-opportunity deviation constraint and the intraday rolling optimization model. Based on the user's optimization requirements, the robust optimization model or the opportunistic optimization model is invoked to obtain the current power grid operation data and the day-ahead plan, and input into the robust optimization model or the opportunistic optimization model to output the electric vehicle dispatch strategy; Based on user optimization requirements, robust optimization models or opportunistic optimization models are retrieved to obtain current power grid operation data and day-ahead plans. These data are then input into the robust optimization models or opportunistic optimization models to output electric vehicle dispatch strategies. Other aspects include: Obtain historical discharge compensation data and historical scheduling data of users, and obtain the demand fluctuation relationship between the historical scheduling correction ratio and the change range of historical scheduling cost correction of users. Based on the robust optimization model, the first scheduling strategy and the first scheduling cost deviation are output according to the current power grid operation data and the day-ahead plan. Based on the opportunity optimization model, a second dispatching strategy and the second dispatching cost deviation are output according to the current power grid operation data and the day-ahead plan. Based on the demand fluctuation relationship, obtain the first scheduling correction ratio corresponding to the first scheduling cost deviation and the second scheduling correction ratio corresponding to the second scheduling cost deviation, and output the first scheduling strategy and / or the second scheduling strategy based on the comparison result of the first scheduling correction ratio and the second scheduling correction ratio. The first scheduling policy and / or the second scheduling policy are output based on the comparison result of the first scheduling adjustment ratio and the second scheduling adjustment ratio, including: If the first scheduling correction ratio is greater than the second scheduling correction ratio, then the first scheduling strategy is output. If the second scheduling correction ratio is greater than the first scheduling correction ratio, then the second scheduling strategy is output. If the first scheduling correction ratio and the second scheduling correction ratio are the same, then the first scheduling policy and the second scheduling policy are output synchronously.

2. The method for optimizing intraday power point tracking of electric vehicles as described in claim 1, characterized in that: The initial correlation fluctuation relationship between discharge intention and the feasible region of electric vehicles, constructed based on the Weber-Fechner law and historical electric vehicle discharge response data and historical discharge compensation data, includes: Based on the Weber-Fechner law, a first correlation fluctuation relationship is constructed between discharge response rate and discharge compensation price; Based on the virtual battery model, the feasible domain of electric vehicles is constructed according to the hardware parameters of electric vehicles, user behavior data, and the first correlation fluctuation relationship. An initial correlation fluctuation relationship is constructed based on the first correlation fluctuation relationship and the feasible region of electric vehicles.

3. The method for optimizing intraday power point tracking of electric vehicles as described in claim 2, characterized in that: The method for quantifying the uncertainty in the initial correlation fluctuation relationship based on information gap decision theory and constructing the dynamic discharge intention fluctuation relationship includes: The uncertainty of the discharge response rate is calculated using the envelope model in the information gap decision theory based on the predicted value of the discharge response rate in the initial correlation fluctuation relationship and the upper and lower limits of the discharge response rate. A dynamic discharge intention fluctuation relationship is constructed based on uncertainty and initial correlation fluctuation relationship.

4. The method for optimizing intraday power point tracking of electric vehicles as described in claim 1, characterized in that: The optimization objective is constructed based on the maximum tracking degree of the daily plan, and the intraday rolling optimization model is constructed based on the dynamic discharge intention fluctuation relationship and the optimization objective, including: The tracking level is obtained by measuring the difference between the planned power output of electric vehicle aggregators at a given time and the total charging and discharging power of electric vehicles.

5. The method for optimizing intraday power point tracking of electric vehicles as described in claim 3, characterized in that: The first correlation fluctuation relationship between discharge response rate and discharge compensation price constructed based on the Weber-Fechner law includes: A dynamic fluctuation relationship of the maximum discharge response rate is constructed based on the historical maximum discharge response rate data corresponding to each discharge compensation price. A dynamic fluctuation relationship of minimum discharge response rate is constructed based on historical minimum discharge response rate data corresponding to each discharge compensation price. A dynamic fluctuation relationship of the benchmark response rate is constructed based on the historical average discharge response rate data corresponding to each discharge compensation price. The first correlation fluctuation relationship is constructed based on the dynamic fluctuation relationship of the maximum discharge response rate, the dynamic fluctuation relationship of the minimum discharge response rate, and the dynamic fluctuation relationship of the reference response rate.

6. The method for optimizing intraday power point tracking of electric vehicles as described in claim 5, characterized in that: The uncertainty in calculating the discharge response rate using the envelope model in information gap decision theory based on the predicted discharge response rate value and the upper and lower limits of the discharge response rate in the initial correlation fluctuation relationship includes: Based on the dynamic fluctuation relationship of the maximum discharge response rate, the dynamic fluctuation relationship of the minimum discharge response rate, and the dynamic fluctuation relationship of the benchmark response rate, the maximum discharge response rate, the minimum discharge response rate, and the predicted discharge response rate under the corresponding discharge compensation price are obtained. The uncertainty under the discharge compensation price is obtained by measuring the relative fluctuation ratio of the predicted discharge response rate to the maximum and minimum discharge response rates.

7. The method for optimizing intraday power point tracking of electric vehicles as described in claim 6, characterized in that: The construction of the robust optimization model based on the benefit-cost robust deviation constraint and the intraday rolling optimization model includes: Using the maximization of uncertainty as the robust optimization objective, the optimization layer of the robust optimization model is constructed using the optimization objective of the intraday rolling optimization model and the robust optimization objective. The constraint layer of the robust optimization model is constructed using the constraints of the intraday rolling optimization model and the benefit-cost robust deviation constraint.

8. The method for optimizing intraday power point tracking of electric vehicles as described in claim 6, characterized in that: The construction of the opportunity optimization model based on the benefit-cost-opportunity deviation constraint and the intraday rolling optimization model includes: Minimizing uncertainty is taken as the opportunity optimization objective. The optimization layer of the opportunity optimization model is constructed using the optimization objective of the intraday rolling optimization model and the opportunity optimization objective. The constraint layer of the opportunity optimization model is constructed using the constraints of the intraday rolling optimization model and the benefit-cost-opportunity deviation constraint.

9. The method for optimizing intraday power point tracking of an electric vehicle as described in claim 6, characterized in that: The benefit-cost robustness deviation constraint is the maximum increase in cost ratio of the benefit-cost of the scheduling strategy relative to the benefit-cost corresponding to the predicted discharge response rate. The benefit-cost opportunity deviation constraint is the maximum cost reduction ratio of the benefit-cost of the scheduling strategy relative to the benefit-cost corresponding to the predicted discharge response rate.

10. An intraday power point tracking optimization system for electric vehicles, used to implement the method as described in any one of claims 1 to 9, characterized in that: include: The correlation analysis unit is used to construct the initial correlation fluctuation relationship between discharge intention and the feasible region of electric vehicles based on the Weber-Fechner law, historical electric vehicle discharge response data and historical discharge compensation data, and to quantify the uncertainty in the initial correlation fluctuation relationship based on the information gap decision theory, and construct the dynamic discharge intention fluctuation relationship. The intraday optimization unit is used to construct optimization targets based on the maximum tracking degree of the daily plan, and to construct an intraday rolling optimization model based on the dynamic discharge intention fluctuation relationship and the optimization targets. Robust optimization unit, used to construct robust optimization model based on benefit-cost robust deviation constraints and intraday rolling optimization model; The opportunity optimization unit is used to construct an opportunity optimization model based on the benefit-cost-opportunity deviation constraint and the intraday rolling optimization model. The scheduling unit is used to retrieve the robust optimization model or the opportunistic optimization model according to the user's optimization requirements, obtain the current power grid operation data and the day-ahead plan, and input them into the robust optimization model or the opportunistic optimization model to output the electric vehicle scheduling strategy.

Images