A big data-based vehicle-network interaction multi-objective optimization scheduling method

Abstract

The embodiment of the application provides a kind of based on big data's vehicle network interaction multi-objective optimization scheduling method, comprising, step S1, with electric vehicle charging and discharging power, energy storage system power, thermal power unit output, new energy output and grid interaction power as decision variable, with time-of-use electricity price as incentive variable, establish multi-objective optimization scheduling model;Step S2, with regional power grid economy optimal, load curve peak-valley difference minimum, new energy consumption rate highest as optimization goal, based on the multi-objective optimization scheduling model constructed, the electric vehicle charging and discharging is optimized and scheduled, realizes the peak load shifting and new energy consumption of power system.This embodiment of the application overcomes the defects that the practicality of existing scheduling method is insufficient and the adaptability to uncertainty is weak, and provides an optimization scheduling model that can be directly deployed in power grid dispatching system.

Description

Technical Field

[0001] This invention relates to the fields of vehicle networking and smart grid technology, and in particular to a multi-objective optimization scheduling method for vehicle-grid interaction based on big data. Background Technology

[0002] With the continuous growth of the number of new energy vehicles, the large-scale random access of electric vehicles to the power grid has a significant impact on its stable operation. Existing scheduling methods generally suffer from the following problems: 1) Most scheduling methods are single-objective and cannot simultaneously meet the needs of economic operation, peak shaving and valley filling and new energy consumption; 2) Insufficient consideration of the uncertainties in wind power, photovoltaic output and conventional loads leads to poor adaptability of the dispatching scheme in actual operation; 3) There is a lack of implementation methods that can directly output deterministic scheduling strategies, which limits engineering applications.

[0003] Therefore, it is necessary to construct a multi-objective optimization scheduling method that can comprehensively optimize power grid operation while ensuring users' vehicle usage needs. Summary of the Invention

[0004] In view of the above-mentioned problems in the prior art, the present invention provides a multi-objective optimization scheduling method for vehicle-to-grid interaction based on big data, so as to solve the technical problems in the prior art, such as difficulty in simultaneously meeting the needs of economic operation, peak shaving and valley filling and new energy consumption, and poor adaptability and engineering practicality.

[0005] This invention provides a multi-objective optimization scheduling method for vehicle-to-network interaction based on big data, comprising the following steps: Step S1: Using electric vehicle charging and discharging power, energy storage system power, thermal power unit output, new energy output and grid interaction power as decision variables, and time-of-use electricity price as incentive parameter, establish a multi-objective optimization scheduling model; Step S2: With the optimization objectives of achieving optimal regional power grid economy, minimizing the peak-to-valley difference in the load curve, and maximizing the renewable energy absorption rate, the charging and discharging of electric vehicles is optimized based on the constructed multi-objective optimization scheduling model to realize peak shaving and valley filling and renewable energy absorption in the power system.

[0006] In one embodiment, step S1 includes, Step S11: Construct an electric vehicle characteristic model, which includes an electric vehicle load spatiotemporal distribution characteristic model, an electric vehicle charging and discharging characteristic model, and an electric vehicle scheduling potential model. Step S12: Construct a multi-objective optimization scheduling model that considers vehicle-to-network interaction; Step S13: Construct a multi-objective optimization scheduling model that considers uncertainty and vehicle-to-network interaction.

[0007] In one embodiment, the spatiotemporal distribution characteristic model of electric vehicle load includes a first trip time model, a parking duration model, an entry / exit time model, and a daily mileage model. The spatiotemporal distribution characteristic model of electric vehicle load is obtained by fitting electric vehicle travel data sample set and related environmental factor data provided by the corresponding regional aggregator through statistical methods and mathematical modeling techniques.

[0008] In one embodiment, the electric vehicle charge / discharge characteristic model is based on a planned charging power model and an initial state of charge (SOC) model. The planned charging power model is constructed as follows:

[0009] in, for t Always plan the charging power. The user's expected battery level when leaving the station. This is the current battery level. For battery capacity, These are the user's expected departure time and the current time, respectively. The initial SOC model is constructed as follows:

[0010] in, SOC init This represents the initial state of charge of the electric vehicle upon entering the station. SOC q For the expected charge of electric vehicles, d r This refers to the daily mileage of electric vehicles, in km. d e This refers to the rated driving range of an electric vehicle, expressed in kilometers.

[0011] In one embodiment, the electric vehicle scheduling potential model is constructed as follows:

[0012]

[0013]

[0014] in, for t The upper limit of the dispatchable charging and discharging load at time +1 for t The lower limit of the dispatchable charging and discharging load at time +1. for t +1 The electric vehicle charging load that must be increased during scheduling time. , These represent the increased electric vehicle charging load due to charging scheduling only and the decreased electric vehicle charging load due to discharging scheduling only, respectively. , They are respectively t The increase and decrease in electric vehicle charging load at time +1 for t +1 refers to the electric vehicle charging load power increased according to scheduling needs at each scheduling time.

[0015] In one embodiment, the optimal economic objective of the regional power grid is to minimize the total operating cost of the regional power grid, which includes the cost of electricity purchase, thermal power generation, renewable energy generation, and energy storage charging and discharging. The objective of minimizing the peak-valley difference of the load curve is to minimize the variance of the load curve and reduce the peak-valley difference. The objective of maximizing the renewable energy absorption rate is to minimize the amount of wind and solar power curtailed.

[0016] In one embodiment, the constraints that the multi-objective optimization scheduling model considering vehicle-to-grid interaction should satisfy include power balance constraints, electric vehicle scheduling potential constraints, upper and lower power limits of each device, climbing power constraints of thermal power units, and state of charge constraints of energy storage devices.

[0017] In one embodiment, the multi-objective optimization scheduling model considering uncertainty and vehicle-to-grid interaction uses interval numbers to represent the random fluctuation range of uncertainty factors, including wind power generation, photovoltaic power generation, and conventional electrical load. The multi-objective optimization scheduling model considering uncertainty and vehicle-to-grid interaction includes an outer optimization model and an inner optimization model. The outer optimization model is used to determine the feasible region of decision variables, and the inner optimization model is used to solve for the upper and lower bounds of each objective function within the interval.

[0018] In one embodiment, the outer optimization model is solved using an improved non-dominated sorting genetic algorithm, which introduces interval confidence and interval overlap to determine whether an individual meets the constraints and compares individuals with the same ordinal value to calculate the crowding distance of the individual.

[0019] Compared with existing technologies, the beneficial effects of the multi-objective optimization scheduling method for vehicle-grid interaction based on big data provided by the embodiments of the present invention are as follows: The embodiments of the present invention use the charging and discharging power of electric vehicles, the power of energy storage systems, the output of thermal power units, the output of new energy sources, and the power of grid interaction as decision variables, and use time-of-use electricity prices as incentive parameters to establish a multi-objective optimization scheduling model. By introducing interval numbers to model uncertain factors, an interval multi-objective optimization scheduling model is constructed, and an improved NSGA-II algorithm is used to solve the problem to obtain a Pareto solution set. Finally, a unique executable scheduling scheme is output through the analytic hierarchy process. This overcomes the shortcomings of existing scheduling methods in terms of practicality and weak adaptability to uncertainty, and provides an optimized scheduling model that can be directly deployed in the power grid scheduling system. Attached Figure Description

[0020] Figure 1 This is a schematic diagram of a multi-objective optimization scheduling method for vehicle-to-network interaction based on big data, provided in an embodiment of the present invention. Detailed Implementation

[0021] To enable those skilled in the art to better understand the technical solution of the present invention, the present invention will be described in detail below with reference to the accompanying drawings and specific embodiments.

[0022] Various embodiments and features of this application are described herein with reference to the accompanying drawings.

[0023] These and other features of this application will become apparent from the following description of preferred forms of embodiments given as non-limiting examples, with reference to the accompanying drawings.

[0024] It should also be understood that although this application has been described with reference to some specific examples, those skilled in the art can certainly implement many other equivalent forms of this application, which have the features described in the claims and are therefore all within the scope of protection defined herein.

[0025] The above and other aspects, features and advantages of this application will become more apparent when taken in conjunction with the accompanying drawings and in view of the following detailed description.

[0026] Specific embodiments of this application are described below with reference to the accompanying drawings; however, it should be understood that the claimed embodiments are merely examples of this application, which can be implemented in various ways. Well-known and / or repeated functions and structures are not described in detail to ascertain the true intent based on the user's historical operations, and to avoid unnecessary or redundant details that would obscure this application. Therefore, the specific structural and functional details claimed herein are not intended to be limiting, but merely serve as the basis and representative basis for the claims to teach those skilled in the art to use this application in various ways with substantially any suitable detailed structure.

[0027] This specification may use the phrases “in one embodiment,” “in another embodiment,” “in yet another embodiment,” or “in other embodiments,” all of which may refer to one or more of the same or different embodiments according to this application.

[0028] The principles and features of the present invention are described below with reference to the accompanying drawings. The embodiments described are for illustrative purposes only and are not intended to limit the scope of the invention. The following description, in conjunction with... Figure 1 The preferred embodiments of the present invention will be described in further detail below: like Figure 1 As shown, this embodiment of the invention provides a multi-objective optimization scheduling method for vehicle-to-network interaction based on big data, including the following steps: Step S1: Using electric vehicle charging and discharging power, energy storage system power, thermal power unit output, new energy output and grid interaction power as decision variables, and time-of-use electricity price as incentive parameter, establish a multi-objective optimization scheduling model; Step S2: With the optimization objectives of achieving optimal regional power grid economy, minimizing the peak-to-valley difference in the load curve, and maximizing the renewable energy absorption rate, the charging and discharging of electric vehicles is optimized based on the constructed multi-objective optimization scheduling model to realize peak shaving and valley filling and renewable energy absorption in the power system.

[0029] In one embodiment, step S1 includes, Step S11: Construct an electric vehicle characteristic model, which includes an electric vehicle load spatiotemporal distribution characteristic model, an electric vehicle charging and discharging characteristic model, and an electric vehicle scheduling potential model. Step S12: Construct a multi-objective optimization scheduling model that considers vehicle-to-network interaction; Step S13: Construct a multi-objective optimization scheduling model that considers uncertainty and vehicle-to-network interaction.

[0030] In one embodiment, the spatiotemporal distribution characteristic model of electric vehicle load includes a first trip time model, a parking duration model, an entry / exit time model, and a daily mileage model. The spatiotemporal distribution characteristic model of electric vehicle load is obtained by fitting electric vehicle travel data sample set and related environmental factor data provided by the corresponding regional aggregator through statistical methods and mathematical modeling techniques.

[0031] In one embodiment, the electric vehicle charge / discharge characteristic model is based on a planned charging power model and an initial state of charge (SOC) model. The planned charging power model is constructed as follows:

[0032] in, for t Always plan the charging power. The user's expected battery level when leaving the station. This is the current battery level. For battery capacity, These are the user's expected departure time and the current time, respectively. The initial SOC model is constructed as follows:

[0033] in, SOC init This represents the initial state of charge of the electric vehicle upon entering the station. SOC q For the expected charge of electric vehicles, d r This refers to the daily mileage of electric vehicles, in km. d e This refers to the rated driving range of an electric vehicle, expressed in kilometers.

[0034] In one embodiment, the electric vehicle scheduling potential model is constructed as follows:

[0035]

[0036]

[0037] in, for t The upper limit of the dispatchable charging and discharging load at time +1 for t The lower limit of the dispatchable charging and discharging load at time +1. for t +1 The electric vehicle charging load that must be increased during scheduling time. , These represent the increased electric vehicle charging load due to charging scheduling only and the decreased electric vehicle charging load due to discharging scheduling only, respectively. , They are respectively t The increase and decrease in electric vehicle charging load at time +1 for t +1 refers to the electric vehicle charging load power increased according to scheduling needs at each scheduling time.

[0038] In one embodiment, the optimal economic objective of the regional power grid is to minimize the total operating cost of the regional power grid, which includes the cost of electricity purchase, thermal power generation, renewable energy generation, and energy storage charging and discharging. The objective of minimizing the peak-valley difference of the load curve is to minimize the variance of the load curve and reduce the peak-valley difference. The objective of maximizing the renewable energy absorption rate is to minimize the amount of wind and solar power curtailed.

[0039] In one embodiment, the constraints that the multi-objective optimization scheduling model considering vehicle-to-grid interaction should satisfy include power balance constraints, electric vehicle scheduling potential constraints, upper and lower power limits of each device, climbing power constraints of thermal power units, and state of charge constraints of energy storage devices.

[0040] In one embodiment, the multi-objective optimization scheduling model considering uncertainty and vehicle-to-grid interaction uses interval numbers to represent the random fluctuation range of uncertainty factors, including wind power generation, photovoltaic power generation, and conventional electrical load. The multi-objective optimization scheduling model considering uncertainty and vehicle-to-grid interaction includes an outer optimization model and an inner optimization model. The outer optimization model is used to determine the feasible region of decision variables, and the inner optimization model is used to solve for the upper and lower bounds of each objective function within the interval.

[0041] In one embodiment, the outer optimization model is solved using an improved non-dominated sorting genetic algorithm, which introduces interval confidence and interval overlap to determine whether an individual meets the constraints and compares individuals with the same ordinal value to calculate the crowding distance of the individual.

[0042] The present invention relates to a V2G multi-objective optimization scheduling model applicable to regional power grid dispatching systems, which takes into account economic efficiency, peak shaving and valley filling and renewable energy absorption capacity.

[0043] This invention establishes a multi-objective optimization scheduling model using electric vehicle charging and discharging power, energy storage system power, thermal power unit output, renewable energy output, and grid interaction power as decision variables, and time-of-use pricing as an incentive parameter. The objectives include: 1) The regional power grid operates at its most economical speed; 2) The load curve has the smallest peak-to-valley difference; 3) It has the highest rate of new energy consumption.

[0044] To address the uncertainty in wind and solar power output and load, this invention introduces interval numbers to model the uncertainty factors, constructs an interval multi-objective optimization scheduling model, and uses an improved NSGA-II algorithm to solve it, obtaining a Pareto solution set. Finally, a unique executable scheduling scheme is output through the analytic hierarchy process.

[0045] The embodiments of the present invention can be widely applied to regional power distribution network dispatching systems to achieve peak shaving and valley filling, as well as efficient absorption of new energy sources.

[0046] In one embodiment, the present invention specifically includes the following steps: 1.1 Electric Vehicle Characteristic Modeling 1.1.1 Spatiotemporal Distribution Characteristics Model of Electric Vehicle Load The spatiotemporal distribution model of electric vehicle load includes: first trip time model, parking duration model, station entry / exit time model, and daily mileage model. This model is derived from a sample set of electric vehicle travel data and related environmental factor data provided by aggregators in the region, and is fitted using statistical methods and mathematical modeling techniques.

[0047] 1) First trip time The distribution of first trip time follows a Burr-type XII distribution with the following probability density function:

[0048] In the formula, α is the scale parameter, c1 is the shape parameter 1, and c2 is the shape parameter 2. We take α = 6.1, c1 = 9.5, and c2 = 0.53.

[0049] 2) Parking duration The parking duration is approximately estimated to follow a log-normal distribution, with a mean of μ. p =2.8, variance σ p =0.95, then the probability density function is:

[0050] 3) Arrival / Departure Time Assuming that the electric vehicle connects to the charging station immediately upon entering the station, the time when the electric vehicle begins charging or discharging can be approximated by the time it enters the station, taking the mean μ of a normal distribution. s =17.5, variance σ s =4.1, then the probability density function of the electric vehicle's grid access time is:

[0051] 4) Daily mileage Data fitting reveals that the daily mileage of electric vehicles follows a log-normal distribution, with the probability density function being:

[0052] In the formula, the mean μ of the normal distribution is taken. r =3.4, variance σ r =0.98.

[0053] 1.1.2 Electric Vehicle Charging and Discharging Characteristic Model 1) Planned charging power Electric vehicles can only participate in dispatching if they meet the condition of reaching the desired state of charge when the user leaves the station. (Definition) t Always plan charging power as follows:

[0054] In the formula, The user's expected battery level when leaving the station. This is the current battery level. For battery capacity, These are the user's expected departure time and the current time, respectively.

[0055] 2) Initial SOC Electric vehicle initial entry SOC Related to daily mileage, given the rated mileage of an electric vehicle, the daily power consumption is the ratio of the daily mileage to the rated mileage. The initial charge of the electric vehicle upon entering the station is the difference between the expected charge and the aforementioned ratio. The expression is:

[0056] In the formula, SOC init This represents the initial state of charge of the electric vehicle upon entering the station. SOC q For the expected charge of electric vehicles, d r The daily mileage (km) of electric vehicles. d e Rated driving range (km) for electric vehicles.

[0057] 1.1.3 Electric Vehicle Dispatch Potential Model The arrival and departure times of electric vehicles follow a normal probability distribution, and the distribution parameters are obtained through statistics. Large-scale electric vehicle deployment... t The upper and lower limits of schedulable power at time +1 are calculated as follows:

[0058]

[0059]

[0060] In the formula, for t The upper limit of the dispatchable charging and discharging load at time +1 for t The lower limit of the dispatchable charging and discharging load at time +1. for t +1 The EV load (electric vehicle charging load) that must be added at the scheduling time. , These represent the EV load increase due to charging scheduling only and the EV load decrease due to discharging scheduling only, respectively. , They are respectively tThe increase and decrease of EV load at time +1 for t +1 refers to the EV load power added during scheduling based on scheduling needs.

[0061] 1.2 Multi-objective optimization scheduling model considering vehicle-to-network interaction 1.2.1 Parameters to be optimized For a given scheduling period, record all scheduling time slots in the model. θ T The decision variables are x s .for t Scheduling period, decision variables x s As shown in Table 1-1.

[0062] Table 1-1 Decision Variables x s

[0063] For a given scheduling period, record all scheduling time slots in the model. θ T The state variables are x c .for t Scheduling period, decision variables x c As shown in Table 1-2.

[0064] Table 1-2 State Variables x c

[0065] 1.2.2 Objective Function 1) Economic objective: Minimize the total operating cost of the regional power grid, including electricity purchase cost, thermal power cost, new energy power generation cost, and energy storage charging and discharging cost.

[0066]

[0067]

[0068] In the formula, C ag For the operating costs of the regional power grid, C grid To reduce the cost of exchanging power with the external power grid, C w For the cost of wind power generation, C pv For the cost of photovoltaic power generation, C fFor the cost of thermal power generation, C b The charging and discharging costs of energy storage devices, Price per unit of power exchanged between the regional power grid and the external power grid (RMB / kW).

[0069] 2) Peak shaving and valley filling objective: Minimize the variance of the load curve and reduce the peak-valley difference.

[0070]

[0071] In the formula, V load The variance of the load curve, The power of the electrical load within the grid (kW) This represents the average electrical load (kW).

[0072] 3) New energy consumption target: Minimize the amount of wind and solar power curtailment.

[0073]

[0074] In the formula, P x The amount of wind and solar power curtailed at a certain moment (kW) Let t be the maximum power generation of the photovoltaic unit (kW). for t Maximum generating capacity of wind turbine at any given time (kW).

[0075] 1.2.3 Constraints for t ∈ θ T Each variable should satisfy the following constraints.

[0076] 1) Power balance constraints:

[0077] In the formula, In order to exchange power (kW) with the external power grid, For conventional electrical load (kW), The charging and discharging power (kW) of electric vehicles The power generation capacity (kW) of the wind turbine unit. The power generation capacity (kW) of the photovoltaic unit. The power generation capacity (kW) of thermal power units. The charging and discharging power (kW) of the energy storage device.

[0078] 2) Constraints on the dispatch potential of electric vehicles:

[0079] In the formula, for t Lower limit power (kW) for electric vehicle charging and discharging scheduling during specific time periods. for t Maximum power (kW) for electric vehicle charging and discharging during specific time periods.

[0080] 3) Power upper and lower limits constraints for each device:

[0081] In the formula, To determine the lower limit (kW) of power exchange with the external power grid, The upper limit (kW) of power exchanged with the external power grid, The lower limit of charging and discharging power (kW) for electric vehicles. The upper limit of charging and discharging power (kW) for electric vehicles. This refers to the lower limit of the battery's charge and discharge power (kW). The upper limit of battery charging and discharging power (kW), This represents the lower limit of thermal power generation capacity (kW). This is the upper limit for thermal power generation capacity (kW). This represents the upper limit of wind power generation capacity (kW). This represents the upper limit of photovoltaic power generation (kW).

[0082] 4) Power constraints for thermal power units climbing slopes:

[0083] In the formula, P down_max The power limit for thermal power units climbing downhill (kW) P up_max Power limit (kW) for thermal power units climbing uphill.

[0084] 5) State of charge constraints for energy storage devices:

[0085] In the formula, SOC T This represents the final value for the state-of-charge scheduling of energy storage devices. SOC init This represents the initial state of charge (SOC) of the energy storage device. SOC b_min This is the state-of-charge limit for electrical energy storage devices. SOC b_max This represents the upper limit of the state of charge for energy storage devices.

[0086] 1.2.4 Optimize scheduling strategy The optimized scheduling strategy of this invention uses time-of-use pricing as an incentive parameter. By acquiring parameters such as the initial charge of the electric vehicle, battery parameters, and expected departure time, grid dispatch instructions are calculated. The electric vehicle receives and executes the grid dispatch instructions to achieve goals such as assisting the grid in reducing operating costs, peak shaving and valley filling, and facilitating the integration of renewable energy.

[0087] 1.3 Vehicle-to-Network Interaction Optimization Scheduling Model Considering Uncertainty Building upon the multi-objective optimization scheduling model shown in Section 1.2, an uncertainty factor description is added to form a vehicle-network interaction optimization scheduling model that considers uncertainty. This section uses interval numbers to represent the value range of the uncertainty factors. These interval numbers can be described by obtaining the upper and lower limits, or the midpoint and width. Compared to other uncertainty representation methods, obtaining the values ​​of the interval numbers is relatively easy.

[0088] 1.3.1 Uncertainty Factors 1) Wind power generation Using interval numbers [ P W This describes the range of random fluctuations in the power output of wind power generation.

[0089]

[0090]

[0091] In the formula, for t Forecasted wind power generation at any given time (kW) for t The fluctuation range of wind power generation at any given time (kW), This represents the maximum fluctuation range.

[0092] 2) Photovoltaic power generation Using interval numbers [ P pv This describes the range of random fluctuations in the power output of photovoltaic power generation.

[0093]

[0094]

[0095]

[0096]

[0097] In the formula, for t Forecasted photovoltaic power generation at any given time (kW) for t Lower limit of photovoltaic power generation forecast (kW) at any given time. for t The upper limit of the predicted photovoltaic power generation capacity at any given time (kW). for t The fluctuation range of photovoltaic power generation at any given time (kW), for t Predicted solar radiation intensity at any given time (W / m²) for t Lower limit of solar radiation intensity prediction at any given time (W / m²) for t Upper limit of predicted solar radiation intensity (W / m²) at any given time. for t The fluctuation range of solar radiation intensity at any given time (W / m²), This represents the uncertainty fluctuation coefficient of photovoltaic power generation. This represents the maximum fluctuation range (kW) of photovoltaic power generation. This represents the uncertainty fluctuation coefficient of solar radiation intensity. This represents the maximum fluctuation range (kW) of solar radiation intensity.

[0098] 3) Conventional electrical load Using interval numbers [ This describes the range of uncertain random fluctuations in conventional electrical load.

[0099]

[0100]

[0101] In the formula, for t Lower limit of predicted conventional electrical load power (kW) at any given time. for t Upper limit of predicted conventional electrical load power (kW) at any given time. for t Forecast value of conventional electrical load power (kW) at any given time. for t The fluctuation range of normal electrical load power at any given time (kW), This represents the uncertainty fluctuation coefficient of conventional electrical load. This represents the maximum fluctuation range (kW) of conventional electrical load power.

[0102] 1.3.2 Inter-regional Multi-objective Optimization Scheduling Model The interval multi-objective optimization scheduling model consists of two parts: an outer optimization model and an inner optimization model. The outer optimization model is used to determine the feasible region of the decision variables, while the inner optimization model is used to solve for the upper and lower bounds of each objective function within the interval.

[0103] 1) Outer layer optimization model

[0104] The constraints of the outer optimization model are basically the same as those in Section 1.2.3, except that the constraints involving interval numbers are as follows: t ∈ θ T :

[0105]

[0106]

[0107] 2) Inner layer optimization model The inner optimization model is based on the decision variables determined by the outer optimization model. x s The optimization is performed within a certain range to achieve the optimization objective. The decision variables of the inner optimization model are the various uncertain factors. By optimizing the inner optimization model, the interval boundaries of each objective function value are found.

[0108] The objective functions are as follows:

[0109]

[0110]

[0111] 1.3.3 Solving the Interval Multi-Objective Optimization Scheduling Model 1.3.3.1 Solving the outer layer optimization model This invention employs an improved NSGA-II algorithm to solve the problem of solving the outer optimization model. By introducing interval confidence and interval overlap, it determines whether an individual meets the constraint conditions and compares individuals with the same ordinal value to calculate the individual crowding distance.

[0112] For interval numbers q 1=[ a 1, b 1], q 2=[ a 2, b 2]∈ I ( R ), interval width w 1= w ( q 1), w 2= w ( q 2) The interval confidence level is expressed as follows:

[0113] Recorded as q 1≥ q Interval confidence level of 2.

[0114] For individuals x It satisfies the first j The interval confidence level of each constraint is ,but:

[0115] Correspondingly, individuals x Degree of violation of this constraint L j for:

[0116] For some interval equality constraints, in the form of... h k ( x , c )= b k To facilitate the solution, these equality constraints can be relaxed, that is, they can be transformed into two inequality constraints:

[0117]

[0118] Set a confidence threshold That is, when the confidence level of x with respect to a certain constraint is greater than or equal to this threshold, it is said that... x A solution is considered feasible if it is not feasible otherwise. Dominance ranking is determined by comparing the dominance relationships between feasible solutions and between infeasible solutions. For evolutionary individuals with the same ordinal value... x 1, x 2， The intersection interval of their objective function values ​​can be expressed as: f i ( x 1, c )∩ f i ( x 2, c ),but x 1, x The interval overlap of 2 is expressed as:

[0119] x 1 and x The distance of 2 is:

[0120] In the formula m ( f i ( x 1, c ))—— Interval f i ( x 1, c )midpoint; V ( x 1)—— x The volume of the hyperbody for the objective function 1.

[0121] Assumption x 2 and x 3 is x An evolved individual of 1, and at a distance x 1. Recently, then x The crowding distance of 1 can be expressed as:

[0122] 1.3.3.2 Solving the Inner Layer Optimization Model (1) Economic objective: For the economic objective function, since the power generation cost of wind power and photovoltaic power is significantly lower than the power generation cost of thermal power, the discharge cost of energy storage and the cost of purchasing electricity from the external grid by the regional power grid, the economic objective function value reaches its minimum value when the power generation of wind power and photovoltaic power are both at the upper limit of the fluctuation range and the conventional power load is at the lower limit of the fluctuation range; conversely, the economic objective function value reaches its maximum value when the power generation of wind power and photovoltaic power are both at the lower limit of the fluctuation range and the conventional power load is at the upper limit of the fluctuation range.

[0123] (2) Peak shaving and valley filling objective: For the peak shaving and valley filling objective function, since thermal power generation and energy storage systems are controllable, their power generation can be flexibly adjusted. They can provide additional power during peak load and reduce the load during valley load. Therefore, when the power generation of wind power and photovoltaic power are both at the lower limit of the fluctuation range, and the conventional power load is at the lower limit of the fluctuation range, the thermal power generation and energy storage system have the greatest effect, and the peak shaving and valley filling objective function value reaches the minimum value; conversely, when the power generation of wind power and photovoltaic power are both at the upper limit of the fluctuation range, and the conventional power load is at the upper limit of the fluctuation range, the peak shaving and valley filling objective function value reaches the maximum value.

[0124] (3) New Energy Consumption Target: Obviously, the grid-connected power of wind power and photovoltaic power plays an important role in the new energy consumption target function. When the grid-connected power of wind power and photovoltaic power are both at the upper limit of the fluctuation range, and the conventional power load is at the upper limit of the fluctuation range, the new energy consumption target function value reaches the minimum value; conversely, when the grid-connected power of wind power and photovoltaic power are both at the lower limit of the fluctuation range, and the conventional power load is at the lower limit of the fluctuation range, the new energy consumption target function value reaches the maximum value.

[0125] The above embodiments are merely exemplary embodiments of the present invention and are not intended to limit the present invention. The scope of protection of the present invention is defined by the claims. Those skilled in the art can make various modifications or equivalent substitutions to the present invention within its spirit and scope of protection, and such modifications or equivalent substitutions should also be considered to fall within the scope of protection of the present invention.

Claims

1. A multi-objective optimization scheduling method for vehicle-to-grid interaction based on big data, characterized in that, Includes the following steps: Step S1: Using electric vehicle charging and discharging power, energy storage system power, thermal power unit output, new energy output and grid interaction power as decision variables, and time-of-use electricity price as incentive parameter, establish a multi-objective optimization scheduling model; Step S2: With the optimization objectives of achieving optimal regional power grid economy, minimizing the peak-to-valley difference in the load curve, and maximizing the renewable energy absorption rate, the charging and discharging of electric vehicles is optimized based on the constructed multi-objective optimization scheduling model to realize peak shaving and valley filling and renewable energy absorption in the power system.

2. The multi-objective optimization scheduling method for vehicle-to-network interaction based on big data as described in claim 1, characterized in that: Step S1 includes, Step S11: Construct an electric vehicle characteristic model, which includes an electric vehicle load spatiotemporal distribution characteristic model, an electric vehicle charging and discharging characteristic model, and an electric vehicle scheduling potential model. Step S12: Construct a multi-objective optimization scheduling model that considers vehicle-to-network interaction; Step S13: Construct a multi-objective optimization scheduling model that considers uncertainty and vehicle-to-network interaction.

3. The multi-objective optimization scheduling method for vehicle-to-network interaction based on big data as described in claim 2, characterized in that: The electric vehicle load spatiotemporal distribution characteristic model includes an initial trip time model, a parking duration model, an entry / exit time model, and a daily mileage model. The electric vehicle load spatiotemporal distribution characteristic model is obtained by fitting electric vehicle travel data sample sets and related environmental factor data provided by the corresponding regional aggregator through statistical methods and mathematical modeling techniques.

4. The multi-objective optimization scheduling method for vehicle-to-network interaction based on big data as described in claim 2, characterized in that: The electric vehicle charging and discharging characteristic model is based on the planned charging power model and the initial SOC model. The planned charging power model is constructed as follows: in, for t Always plan the charging power. The user's expected battery level when leaving the station. This is the current battery level. For battery capacity, These are the user's expected departure time and the current time, respectively. The initial SOC model is constructed as follows: in, SOC init This represents the initial state of charge of the electric vehicle upon entering the station. SOC q For the expected charge of electric vehicles, d r This refers to the daily mileage of electric vehicles, in km. d e This refers to the rated driving range of an electric vehicle, expressed in kilometers.

5. The multi-objective optimization scheduling method for vehicle-to-network interaction based on big data according to claim 2, characterized in that: The electric vehicle scheduling potential model is constructed as follows: in, for t The upper limit of the dispatchable charging and discharging load at time +1 for t The lower limit of the dispatchable charging and discharging load at time +1. for t +1 The electric vehicle charging load that must be increased during scheduling time. , These represent the increased electric vehicle charging load due to charging scheduling only and the decreased electric vehicle charging load due to discharging scheduling only, respectively. , They are respectively t The increase and decrease in electric vehicle charging load at time +1 for t +1 refers to the electric vehicle charging load power increased according to scheduling needs at each scheduling time.

6. The multi-objective optimization scheduling method for vehicle-to-network interaction based on big data according to claim 2, characterized in that: The optimal economic objective of the regional power grid is to minimize the total operating cost of the regional power grid, which includes the cost of electricity purchase, thermal power generation, renewable energy generation, and energy storage charging and discharging. The objective of minimizing the peak-valley difference of the load curve is to minimize the variance of the load curve and reduce the peak-valley difference. The objective of maximizing the renewable energy absorption rate is to minimize the amount of wind and solar power curtailed.

7. The multi-objective optimization scheduling method for vehicle-to-network interaction based on big data according to claim 2, characterized in that: The constraints that the multi-objective optimization scheduling model considering vehicle-to-grid interaction should satisfy include power balance constraints, electric vehicle scheduling potential constraints, upper and lower power limits of each device, climbing power constraints of thermal power units, and state of charge constraints of energy storage devices.

8. The multi-objective optimization scheduling method for vehicle-to-network interaction based on big data according to claim 7, characterized in that: The multi-objective optimization scheduling model considering uncertainty and vehicle-to-grid interaction uses interval numbers to represent the random fluctuation range of uncertainty factors, including wind power generation, photovoltaic power generation, and conventional electrical load. The multi-objective optimization scheduling model considering uncertainty and vehicle-to-grid interaction includes two parts: an outer optimization model and an inner optimization model. The outer optimization model is used to determine the feasible region of decision variables, and the inner optimization model is used to solve for the upper and lower bounds of each objective function within the interval.

9. The multi-objective optimization scheduling method for vehicle-to-network interaction based on big data as described in claim 8, characterized in that: The outer optimization model is solved using an improved non-dominated sorting genetic algorithm. By introducing interval confidence and interval overlap, it determines whether an individual meets the constraints and compares individuals with the same ordinal value to calculate the crowding distance of the individual.

Images