Electric vehicle V2G regulation capability detection method

Abstract

An electric vehicle V2G regulation capability detection method comprises the following steps: constructing a vehicle network interaction (V2G) flexibility regulation model considering regulation inertia, considering electric vehicle user energy consumption behavior habits and electric vehicle (EV) and charging facility characteristic constraint conditions, and fully considering subsequent influences on a power system after V2G behavior regulation is finished; user charging and discharging behaviors are dynamically simulated by combining global operation characteristics of a power system, and V2G adjustable resources are quantitatively calculated. According to the model, a load side adjustable resource calculation method and a power supply and power grid integrated optimization planning (GTEP) calculation method are further mutually embedded and iterated, and quantitative evaluation of the influence of V2G on power supply installation and 8760-hour energy storage output of a power system is realized while the calculation accuracy of the adjustable resources of the system is further optimized.

Description

Technical Field

[0001] This invention relates to a technology in the field of power grid control, specifically a method for detecting the V2G regulation capability of electric vehicles. Background Technology

[0002] Electric vehicles (EVs) have long parking times and are equipped with energy storage batteries, enabling bidirectional regulation and rapid response. While parked and connected to the grid, EVs can utilize idle batteries to support grid power supply during peak load periods and recharge during off-peak periods to replenish electricity consumed during travel and discharge. This flexible "source-load" role interaction provides short-term flexibility to the grid, achieving V2G (Vehicle-to-Grid) regulation. However, existing regulation technologies, after EVs discharge during peak periods, may lead to a surge in charging during subsequent peak periods due to travel demand constraints, causing new peak loads and deviating from system requirements. This situation is not fully considered in current research. Furthermore, existing V2G models do not adequately integrate with power system models, failing to fully account for the high proportion of renewable energy in new power systems, thus hindering the analysis of the mutual influence between V2G and the power system's source-grid-storage components. Summary of the Invention

[0003] To address the aforementioned shortcomings of existing technologies, this invention proposes a method for detecting the V2G (Vehicle-to-Grid) regulation capability of electric vehicles. This method constructs a vehicle-to-grid (V2G) flexibility regulation model that considers regulation inertia. It takes into account the energy consumption habits of electric vehicle users, the characteristics and constraints of electric vehicles (EVs) and charging facilities, and fully considers the subsequent impact on the power system after V2G behavior regulation ends. The model dynamically simulates user charging and discharging behavior and quantifies V2G adjustable resources by combining the global operating characteristics of the power system. Furthermore, the model embeds and iterates the load-side adjustable resource calculation method with the power grid integrated optimization planning (GTEP) method, further optimizing the accuracy of system adjustable resource calculation while achieving a quantitative assessment of the impact of V2G on the installed power generation capacity and 8760-hour energy storage output of the power system.

[0004] This invention is achieved through the following technical solution:

[0005] This invention relates to a method for detecting the V2G regulation capability of electric vehicles, comprising:

[0006] Step 1: Construct a quantitative analysis model that includes an electric vehicle load sub-model and a generation and transmission extension planning (GTEP) sub-model for high-proportion renewable energy grid connection;

[0007] Step 2: By adding new constraints to the quantitative analysis model, the electric vehicle load sub-model and the GTEP sub-model can interact and iterate to achieve integrated source-grid-load-storage collaborative simulation, thereby obtaining the load-side flexibility adjustment potential and the changes in installed capacity, power generation, and cost parameters before and after adjustment.

[0008] The data interaction and iteration mentioned above include:

[0009] A) Using the electric vehicle load sub-model, the probability distribution of the daily grid connection time, grid connection duration, daily mileage, and battery capacity parameters of each EV in the EV cluster is obtained based on the Monte Carlo algorithm. Considering the automatic charging of EVs after grid connection, the total load curve of electric vehicles for 8760 hours when EVs do not participate in V2G is simulated. After adding it with the other social loads, the total social load curve for 8760 hours is obtained, which is used as the input of the GTEP model. The power grid planning results are generated using the GTEP sub-model to obtain the power supply, energy storage capacity and 8760-hour output, as well as the power flow. The simulation results of the GTEP sub-model are output to the electric vehicle load sub-model.

[0010] B) Based on the total 8760-hour social load curve obtained in the previous step, and combined with the wind and solar power output curves output by the GTEP submodel, the 8760-hour net social load curve can be obtained by subtracting them. After setting a threshold upper limit for the net load (EVs discharge when the net load exceeds the threshold upper limit), the net load can be used as a V2G regulation signal and returned to the electric vehicle load submodel. Considering the charging and travel demand of electric vehicles, battery capacity limits, system regulation targets, and the constraint that concentrated charging after electric vehicle discharge does not cause new load peaks, the Monte Carlo algorithm is used to discharge electric vehicles during periods of excessive net load and guide electric vehicles to charge during periods of low net load. The electric vehicle load submodel further simulates the 8760-hour total load curve of electric vehicles after participating in V2G regulation.

[0011] C) The electric vehicle load sub-model outputs the simulation results to the GTEP sub-model again. The total 8760-hour load curve of electric vehicles participating in V2G is added to the remaining total societal load to obtain the total 8760-hour societal load curve, which is then used as input to the GTEP model. The GTEP model is used to generate the power grid planning results after V2G participation in regulation, obtaining the power source, energy storage capacity, and 8760-hour output, as well as the power flow. Finally, the 8760-hour load curves of electric vehicles, the total 8760-hour load curve of the societal system, the power source, energy storage capacity, and 8760-hour output, and the power flow are obtained before and after V2G regulation, enabling the calculation of V2G regulation capacity and the analysis of its impact on the power system.

[0012] Technical effect

[0013] This invention considers inertial regulation constraints, meaning that after an electric vehicle participates in V2G and discharges, subsequent concentrated charging does not cause new load peaks. EVs meeting the constraints participate in V2G normally, while EVs not meeting the inertial constraints stop V2G discharge regulation and participate in regulation only through a compromise of orderly charging. 2. The electric vehicle load sub-model receives GTEP sub-models for mutual iteration. The electric vehicle load sub-model receives other load and wind / solar output data from the GTEP sub-model to generate 8760 hours of net load data as a basis for regulation, used to calculate the V2G regulation potential of electric vehicles. The GTEP sub-model receives the characteristic curves before and after 8760 hours of load regulation provided by the electric vehicle load sub-model, used to calculate the changes in installed capacity, power generation, and cost parameters before and after regulation.

[0014] Compared with existing technologies, this invention, through inertial adjustment constraints, fully considers the subsequent impact on the power system after the V2G behavior adjustment ends. While considering the charging needs of electric vehicles and effectively avoiding the negative impact of V2G on the power system, it maximizes the utilization of EV flexibility adjustment capabilities, meeting the dual practical needs of electric vehicle users and the power system. 2. By iterating between the electric vehicle load sub-model and the GTEP sub-model, the model strengthens the internal coupling between electric vehicles and the power system. This enables a quantitative assessment of the impact of V2G on various power generation parameters, including power system energy storage capacity and 8760-hour energy storage output, as well as power generation and cost parameters, thus clarifying the adjustment value of V2G on the power system. Attached Figure Description

[0015] Figure 1 This is a schematic diagram / flowchart of the invention.

[0016] Figure 2 A schematic diagram of the overall framework of the EV flexibility adjustment model;

[0017] Figure 3 This is a schematic diagram of a load curve-based operation simulation framework.

[0018] Figure 4 Flowchart of the regulation strategy considering V2G regulation inertia constraints;

[0019] Figures 5-7 This is a schematic diagram illustrating the effect of an example. Detailed Implementation

[0020] like Figure 1 As shown, this embodiment relates to a method for detecting the V2G regulation capability of an electric vehicle, including:

[0021] 1) Construct a quantitative analysis model that includes an electric vehicle load sub-model and a generation and transmission extension planning (GTEP) sub-model for high-proportion renewable energy grid connection. The electric vehicle load sub-model generates system regulation demand data based on the net load and energy storage output data in the GTEP sub-model to calculate the load-side flexibility regulation potential. The GTEP sub-model provides load regulation characteristics before and after the electric vehicle load sub-model, and performs power balance analysis calculations with the premise of ensuring power reliability and the lowest comprehensive cost per kilowatt-hour as the optimization objective.

[0022] like Figure 2 As shown, the GTEP sub-model takes power generation capacity, grid capacity, power generation and transmission cost efficiency data, and renewable energy resource endowment and policy objectives as inputs. It comprehensively considers power supply and demand balance, unit operation, power transmission, and unit construction constraints, using the lowest overall cost per kilowatt-hour as the optimization objective. It employs a large-scale mixed integer programming algorithm (CPLEX) to optimize and output the power generation structure for the target year and the generation structure, power flow changes, fuel consumption, and carbon emission parameters for 8760 hours. Specifically: Among them: power investment cost Power grid investment costs System operating costs .

[0023] The optimization objective assumes that the operating cost of intermittent renewable energy and energy storage is 0, and does not consider the penalty for renewable energy curtailment.

[0024] like Figure 3 As shown, to reflect the system fluctuation characteristics under the scenario of large-scale renewable energy integration, the GTEP sub-model adopts load time-series curves, taking the time-series variation characteristics of load levels as an important indicator, laying the foundation for introducing power supply peak shaving and ramping flexibility balance constraints. The load time-series curves are divided by day, and several typical scenarios are selected using a scenario reduction algorithm, which is then combined into an annual load time-series curve. Based on the load time-series curves, the model adds a conventional thermal power ramping constraint set and a unit start-up and shutdown behavior constraint set to the linear operation simulation constraint set, thereby solving the problem that traditional operation simulation methods cannot consider ramping and peak-shaving power system operation constraints, clarifying the positioning of thermal power units as flexible adjustment power sources in the new power system, specifically: .

[0025] The constraints of the GTEP sub-model include:

[0026] Maximum capacity constraints for conventional thermal power units: ;

[0027] Maximum capacity constraints for intermittent renewable energy units: ;

[0028] Maximum feasible capacity constraints for concentrated solar power (CSP) generator units: ;

[0029] Maximum capacity constraints for energy storage units: ;

[0030] Power investment budget constraints: ;

[0031] Power grid investment budget constraints: ;

[0032] Constraints on the proportion of renewable energy generation: ;

[0033] The constraint on the proportion of renewable energy generation also corresponds to the renewable energy quota system requirements in energy policy: ;

[0034] Generation-load balance constraints: Among them: the load power of the node bus cannot exceed the power consumption of the node. ;

[0035] Power flow constraints of existing lines: , , ;

[0036] Current constraints to be addressed: , ;

[0037] Energy output constraints of intermittent renewable energy sources: ;

[0038] Simplified operational constraints for solar thermal power plants: , , , , , , , , , ;

[0039] Operating constraints of energy storage devices: , , , , , ;

[0040] Operating constraints of conventional thermal power units: , , , , , , ;

[0041] System backup constraints: Where: t is the t-th time period, and its value ranges from 1 to N. T ; g represents the g-th conventional thermal power unit, with a value ranging from 1 to N. G ; i represents the i-th type of conventional thermal power unit, with a value from 1 to N. I w represents the w-th wind farm or photovoltaic power station, with a value ranging from 1 to N. W c represents the c-th solar thermal power plant, with values ​​ranging from 1 to N. C b represents the b-th energy storage device, with a value ranging from 1 to N. B n is the nth bus node, with values ​​ranging from 1 to N. N ; l represents the l-th transmission line, with a value ranging from 1 to N. L The start and end nodes are l (+) and l (−) ; s represents the s-th running scenario, with values ​​ranging from 1 to N. S The nth bus node is the set of conventional thermal power units. For the nth bus node, it is a set of wind farms or photovoltaic power stations; For the nth bus node, the set of concentrated solar power plants; For the nth bus node, the set of energy storage devices; For the i-th type of conventional thermal power, there is a set of conventional thermal power units. / This is a collection of existing lines / a collection of lines to be built; The annualized investment cost for conventional thermal power units is expressed in $ / kW-yr. Start-up and shutdown costs for conventional thermal power units, in k$; Variable operating costs for conventional thermal power units, in $ / kW; The installed capacity limit for conventional thermal power units, unit: MW; This represents the minimum output ratio for conventional thermal power units. / For conventional thermal power units, the ability to adjust ramp speed (up / down); / The minimum start-up / shutdown time for conventional thermal power units, in hours; The annualized investment cost of energy storage equipment is expressed in $ / kW-yr. The installed capacity limit for electric energy storage equipment, unit: MW; The energy storage capacity in hours for the energy storage device, in hours; The charging and discharging efficiency of energy storage devices; This represents the initial energy storage level of the energy storage device as a percentage. The annualized investment cost of a concentrated solar power (CSP) plant is expressed in $ / kW-yr. Variable operating costs of concentrated solar power (CSP) plants, in units of $ / kW; The installed capacity limit for concentrated solar power (CSP) plants, unit: MW; / For the up / down ramping capability of solar thermal power plants; This represents the initial percentage of heat storage level in the heat storage stage of a solar thermal power plant. The solar multiplier of a solar thermal power plant; The duration of thermal storage, in hours; The per-unit collector power curve of a solar thermal power plant; Annualized investment cost for wind farms or photovoltaic power plants, unit: $ / kW-yr; The installed capacity limit for wind farms or photovoltaic power plants, in MW; For wind farms / photovoltaic power plants, the per-unit predicted power curves are provided. This is the load demand curve, in MW. The annualized investment cost of transmission lines is expressed in $ / kW-yr. Line transmission capacity, unit: MW; For the reactance of the transmission line; / Renewable energy forecast error ratio / load forecast error ratio; The unit cost of load shedding for the system, in k$ / MWh; Let be the probability of the s-th running scenario; / Maximum budget for power supply investment / grid investment, in k$; The proportion of renewable energy generation in the system; The installed capacity of conventional thermal power units is expressed in MW. The installed capacity of energy storage equipment is expressed in MW. The installed capacity of wind farms or photovoltaic power plants is expressed in MW. The installed capacity of a solar thermal power unit is expressed in MW. This is a 0-1 state variable indicating whether or not to invest in the proposed railway line. Planned power output for conventional thermal power units, unit: MW; The online operating capacity of conventional thermal power units, in MW; / The rated capacity and shutdown capacity of conventional thermal power units are given in MW. / Power output for the planned charging / discharging of energy storage devices, unit: MW; The energy storage level of the electrical energy storage device, in MWh; The maximum possible output of the energy storage device, in MW; Planned power output of solar thermal power units, unit: MW; The thermal storage level of the solar thermal power unit is expressed in MWh. The maximum possible output of a solar thermal power unit, in MW; Power contribution to a wind farm or solar power plant, in MW; Load shedding capacity, unit: MW; Transmission power of a power transmission line, unit: MW; The phase angle of the node.

[0042] The electric vehicle load sub-model reconstructs the random behavior characteristics of EVs through random behavior analysis based on the Monte Carlo algorithm. By observing or sampling user behavior habits and analyzing the development trends of power battery technology, it calculates the probability distributions of daily grid connection time, grid-connected time, daily mileage, and battery capacity parameters of the EV. Random parameters are then repeatedly generated from these probability distributions as the basic input parameters for the simulation of each electric vehicle unit in the model. Specifically:

[0043] EV network usage time, battery capacity and daily driving range Where μ and σ are the expectation and standard deviation of the Gaussian distribution, respectively. The initial remaining charge of an EV upon first grid connection is linearly related to the battery capacity and daily mileage, and therefore also follows a Gaussian distribution. The daily grid connection time of an EV follows a weighted Gaussian mixture distribution, i.e. Where: K is the number of Gaussian distributed components (K=3 in this study), k i Let μ be the weight of the i-th Gaussian distribution. i and σ i Let be the expected value and standard deviation of the i-th Gaussian distribution, respectively. Suppose there are N EVs in the region, then for the n-th EV, the initial connection to the power grid... , where: Cap ev (n) represents the battery capacity, S day (n) represents the daily mileage, C pwr (n) represents the energy consumption per unit distance, f cha (n) represents the average daily charging frequency.

[0044] A certain number of EV clusters are considered as a batch. Input parameters are generated through random behavior analysis based on the Monte Carlo algorithm to simulate EV charging and discharging behavior. When not participating in system regulation, EVs maintain rated power charging upon grid connection until the battery is fully charged or disconnected from the grid. After participating in system regulation, EVs can participate through electricity price signals or dispatch signals. However, regardless of the method used to guide user participation, the fundamental determining factor is the net load of the power system. A larger net load indicates a stronger demand for reducing EV load, while a smaller net load indicates a stronger demand for increasing EV load. For ease of analysis, this model simplifies the relationship between net load and EV regulation demand to a linear relationship. Regulation demand guides EV regulation actions batch by batch. When a batch of EVs participates in system regulation, the regulation demand data is synchronously corrected to guide the charging and discharging behavior of the next batch of EVs, causing EVs to gradually approach the regulation demand. Specifically, the nth EV reduces its load at time t. Where: Pev(n-1,t) is the charging power of the (n-1)th batch of vehicles at time t. When EVs participate in the adjustment, the maximum depth of discharge and off-grid SOC constraints are further considered to ensure user travel needs, specifically: , where: SOC min It is the minimum depth of discharge of the battery, SOC. end T represents the minimum expected battery level when the EV is disconnected from the grid. end This is a collection of EV disconnection times.

[0045] Based on user travel needs, the model's V2G adjustment strategy further incorporates elements such as... Figure 4 The V2G regulation inertia constraint shown fully considers the subsequent impact on the power system after V2G behavior regulation ends. In actual scenarios where EVs participate in V2G regulation, their on-grid time often cannot fully cover the peak and off-peak periods of the power grid to achieve sufficient peak shaving and valley filling. After EVs discharge during peak periods, travel demand constraints may lead to a concentrated charging of vehicle clusters during subsequent peak periods, causing new peak loads and causing the regulation result to deviate from the system's needs. Therefore, based on the current power system demand as the regulation target and user travel demand as the main constraint, this model adds a V2G regulation inertia constraint to account for the subsequent impact of V2G on the power system. The constraint conditions include: Where: n is the car serial number, t is the time, and P evb (n,t) represents the EV charging power when not involved in regulation, P ev (n,t) represents the EV charging power after participating in V2G (positive for charging, negative for discharging), L lim+L(t) is the maximum net load limit within the region, and L(t) is the net load value within the region. EVs that do not meet the inertial constraints cease V2G discharge regulation and only optimize their own charging behavior. The optimization process specifically includes: during the period of a single EV connection to the grid, the model combines the GTEP sub-model to adjust demand data, and the target adjustment power at each time point is updated in descending order, specifically as follows: , where: P sf (n,t) represents the required load reduction of the system, P sf '(n,k) is P sf (n,t) represents the new sequence after sorting, where k is the index, and I(n,k) represents the time corresponding to index k. Based on the above formula, the EV charging strategy and SOC changes are as follows: , For k0, we have , where: P chg The rated power of the charging pile is SOC0(n), which is the initial SOC when connected to the power grid. ev (n,t) represents the EVSOC when no adjustment is involved, and Cap ev (n) represents the battery capacity.

[0046] Through specific experimental simulations of the above detection methods, we obtained 8760 load data (sampling interval of 1 hour) before and after electric vehicle clusters participated in V2G, as well as power system power generation and energy storage capacity and 8760 hours of power generation data (or charge and discharge data, sampling interval of 1 hour) before and after electric vehicle V2G. Table 1 shows a portion of the original data, that is, the electric vehicle load (in 10,000 kilowatts) in China and various regions before and after V2G in the first 100 hours of the 8760 hours.

[0047] Table 1

[0048]

[0049] Table 2. Power system energy storage capacity requirements and annual discharge when charging pile power is 5kW

[0050] Table 3. Power system energy storage capacity requirements and annual discharge when charging pile power is 7kW

[0051] Table 4. Power system energy storage capacity requirements and annual discharge when charging pile power is 10kW

[0052] Table 5. Power system energy storage capacity requirements and annual discharge when charging pile power is 15kW.

[0053] Table 6. Power system energy storage capacity requirements and annual discharge when charging pile power is 20kW

[0054] Table 7. Power system energy storage capacity requirements and annual discharge when charging pile power is 25kW.

[0055] Table 8. Power system energy storage capacity requirements and annual discharge when charging pile power is 30kW

[0056] Table 9. Power system energy storage capacity requirements and annual discharge when charging pile power is 35kW

[0057] Taking my country in 2025 as an example, and assuming that the vehicle-to-grid interaction parameters in my country in 2050 are as shown in Table 10 and the number of electric vehicles is as shown in Table 11, this model is used to analyze the V2G regulation potential and impact on the power system in my country in 2050. Net load refers to the difference between electricity load and wind and solar power output.

[0058] Table 10 Vehicle-to-Network Interaction Parameter Settings in 2050

[0059] Table 11. Number of Private Cars in Various Regions in 2050

[0060] The model can simulate and analyze the V2G regulation potential. By 2050, simulation results of V2G mode regulation, extracted from the 8760-hour load curve, for typical days and weeks, are as follows: Figure 5 and Figure 6As shown in Table 10, when the charging pile power distribution is within the range of 10~30kW, the peak shaving and valley filling are calculated to reach 0.2~2.8 billion kW and 1.1~2.5 billion kW respectively, based on the vehicle-to-grid interaction parameters. The system net load peak value decreased from 1.95 billion kW to 1.67~1.93 billion kW, and the valley value increased from the initial -1.94 billion kW to -1.83~-1.76 billion kW, reducing the peak-valley difference by 1.2~5.2 billion kW, nearly 12% smaller than before the adjustment. Considering willingness and simultaneity factors, the peak capacity provided by the V2G mode reaches 1.1 billion kW. The effective power generation of electric vehicles during peak load periods reaches approximately 378.7 billion kWh, and the maximum charging load increases from the initial 5.5 billion kW to 6.5 billion kW. During the net load peak period, the discharge mode is activated, with the maximum discharge power occurring at 6:00 AM on November 15th, reaching a maximum of 0.93 billion kW.

[0061] The model can simulate and analyze the sensitivity of vehicle-to-grid (V2G) interaction regulation to the parameter of charging power. Taking 2050 as an example, when the charging power increases from 7kW to 15kW, the effect of V2G on peak shaving and valley filling in the system significantly improves. However, as the charging power further increases, the regulation effect exhibits diminishing marginal returns, and the difference in impact on the system becomes negligible. Figure 7 As shown in Table 12, by 2050, when the charging power increases from 5kW to 10kW, the reduction in peak-valley load will increase by 130 million kW; however, when the charging power increases from 10kW to 15kW, the reduction will only increase by 55 million kW. When the charging power increases to 20kW, the reduction will reach saturation; and when the charging power increases from 35kW to 40kW, the reduction will only increase by 4.77 million kW. Detailed data for each region is shown in Table 12, based on a charging pile power of 20kW.

[0062] Table 12. Effects of Vehicle-to-Network Interaction on Peak Shaving and Valley Filling in Different Regions

[0063] The model can simulate and analyze the impact of V2G on power system planning. Using the predicted demand for new energy storage as recorded in the study "Research on the Development Potential and Path of New Energy Storage to Support Dual Carbon Goals" as the initial boundary conditions, the initial total installed energy storage capacity demand in China in 2050 is 950 million kW, of which 650 million kW is for new energy storage. By 2050, based on a charging pile power of 20 kW, vehicle-to-grid interaction will replace 280 million kW of new energy storage capacity, representing a replacement rate of approximately 26%. The demand for 6-hour short-term energy storage and 720-hour long-term energy storage will decrease by approximately 180 million kW and 15 million kW respectively, and the demand for new energy storage will decrease to approximately 460 million kW. The replacement rate of V2G for installed energy storage in Northeast, North, Northwest, and East China will reach 19%, 11%, 16%, and 44% respectively.

[0064] Compared to existing technologies, this model takes a macroscopic view of the power system and its demands, handling the V2G regulation process and quantitatively assessing its regulation capacity. It can provide the electric vehicle load characteristic curve and system power output curve for 8760 hours throughout the year. First, in quantitatively analyzing the impact of V2G on the power system, it organically combines the V2G model with the power system planning model. The model views the true regulation capacity of V2G from the power system perspective, enabling the calculation of V2G regulation potential that better reflects the actual needs of the power system. It also achieves a quantitative analysis of the impact of V2G on the power system, such as calculating the impact of V2G on the demand for energy storage capacity in the power system, which can guide power system planning under V2G. Second, when calculating the V2G potential, this method incorporates inertial regulation constraints. The V2G model will not show a situation where electric vehicle V2G causes a new round of load peaks within a continuous 8760-hour timescale throughout the year. The V2G regulation process fully considers that after EVs discharge during peak periods, travel demand constraints may lead to a certain probability that vehicle clusters will concentrate on charging during subsequent peak periods, causing new peak loads and causing the regulation results to deviate from the system's requirements. Considering that the power system will not allow this problem to occur, the model incorporates new constraints and regulation strategies, making the V2G regulation model simulation closer to the actual application scenario and meeting the dual actual needs of electric vehicle users and the power system.

[0065] The above-described specific implementations can be partially adjusted by those skilled in the art in different ways without departing from the principles and purpose of the present invention. The scope of protection of the present invention is defined by the claims and is not limited to the above-described specific implementations. All implementation schemes within the scope of the claims are bound by the present invention.

Claims

1. A method for detecting the V2G regulation capability of an electric vehicle, characterized in that, include: Step 1: Construct a quantitative analysis model that includes an electric vehicle load sub-model and a generation and transmission extension planning (GTEP) sub-model for high-proportion renewable energy grid connection; Step 2: By adding new constraints to the quantitative analysis model, the electric vehicle load sub-model and the GTEP sub-model can interact and iterate to achieve integrated source-grid-load-storage collaborative simulation, thereby obtaining the load-side flexibility adjustment potential and the changes in installed capacity, power generation, and cost parameters before and after adjustment.

2. The method for detecting the V2G regulation capability of electric vehicles according to claim 1, characterized in that, The data interaction and iteration mentioned above include: A) Using the electric vehicle load sub-model, the probability distribution of the daily grid connection time, grid connection duration, daily mileage, and battery capacity parameters of each EV in the EV cluster is obtained based on the Monte Carlo algorithm. Considering the automatic charging of EVs after grid connection, the total load curve of electric vehicles for 8760 hours when EVs do not participate in V2G is simulated. After adding it with the other social loads, the total social load curve for 8760 hours is obtained, which is used as the input of the GTEP model. The power grid planning results are generated using the GTEP sub-model to obtain the power supply, energy storage capacity and 8760-hour output, as well as the power flow. The simulation results of the GTEP sub-model are output to the electric vehicle load sub-model. B) Based on the total 8760-hour social load curve obtained in the previous step, and combined with the wind and solar power output curves output by the GTEP submodel, the 8760-hour social net load curve is obtained by subtracting them. After setting a threshold upper limit for the net load (EVs discharge when the net load is greater than the threshold upper limit), the net load can be used as a V2G regulation signal and returned to the electric vehicle load submodel. Considering the charging and travel demand of electric vehicles, the battery capacity limit, the system regulation target, and the constraint that concentrated charging after electric vehicle discharge does not cause new load peaks, the Monte Carlo algorithm is used to discharge electric vehicles during periods of excessive net load and guide electric vehicles to charge during periods of low net load. The electric vehicle load submodel further simulates the 8760-hour total load curve of electric vehicles after participating in V2G regulation. C) The electric vehicle load sub-model outputs the simulation results to the GTEP sub-model again. The total 8760-hour load curve of electric vehicles participating in V2G is added to the remaining total social load to obtain the total 8760-hour total social load curve, which is used as the input of the GTEP model. The GTEP model is used to generate the power grid planning results after V2G participation in regulation, and obtain the power supply, energy storage capacity and 8760-hour output, and power flow situation. Finally, the 8760-hour load curve of electric vehicles, the 8760-hour load curve of the whole society, the power supply, energy storage capacity and 8760-hour output, and the power flow situation before and after V2G regulation are obtained, realizing the calculation of V2G regulation capacity and the analysis of its impact on the power system.

3. The method for detecting the V2G regulation capability of electric vehicles according to claim 1 or 2, characterized in that, The GTEP sub-model takes power generation capacity, grid capacity, power generation and transmission cost efficiency data, and renewable energy resource endowment and policy objectives as inputs. It comprehensively considers power supply and demand balance, unit operation, power transmission, and unit construction constraints. With the lowest overall cost per kilowatt-hour as the optimization objective, it uses a large-scale mixed integer programming algorithm (CPLEX) to optimize and output the power generation structure for the target year and the generation structure, power flow changes, fuel consumption, and carbon emission parameters for 8760 hours. Specifically: Among them: power investment cost Power grid investment costs System operating costs .

4. The method for detecting the V2G regulation capability of electric vehicles according to claim 3, characterized in that, The GTEP sub-model uses load time-series curves, taking the time-series variation characteristics of load levels as an important indicator. This lays the foundation for introducing constraints on the balance between power peak shaving and ramping flexibility. The load time-series curves are divided daily, and several typical scenarios are selected using a scenario reduction algorithm, which are then combined into an annual load time-series curve. Based on the load time-series curves, the model adds a set of conventional thermal power ramping constraints and a set of unit start-up and shutdown behavior constraints to the linear operation simulation constraint set, specifically: .

5. The method for detecting the V2G regulation capability of electric vehicles according to claim 3 or 4, characterized in that, The constraints of the GTEP sub-model include: Maximum capacity constraints for conventional thermal power units: ; Maximum capacity constraints for intermittent renewable energy units: ; Maximum feasible capacity constraints for concentrated solar power (CSP) generator units: ; Maximum capacity constraints for energy storage units: ; Power investment budget constraints: ; Power grid investment budget constraints: ; Constraints on the proportion of renewable energy generation: ; The constraint on the proportion of renewable energy generation also corresponds to the renewable energy quota system requirements in energy policy: ; Generation-load balance constraints: Among them: the load power of the node bus cannot exceed the power consumption of the node. ; Power flow constraints of existing lines: , , ; Current constraints to be addressed: , ; Energy output constraints of intermittent renewable energy sources: ; Simplified operational constraints for solar thermal power plants: , , , , , , , , , ; Operating constraints of energy storage devices: , , , , , ; Operating constraints of conventional thermal power units: , , , , , , ; System backup constraints: Where: t is the t-th time period, and its value ranges from 1 to N. T ; g represents the g-th conventional thermal power unit, with a value ranging from 1 to N. G ; i represents the i-th type of conventional thermal power unit, with a value from 1 to N. I w represents the w-th wind farm or photovoltaic power station, with a value ranging from 1 to N. W c represents the c-th solar thermal power plant, with values ​​ranging from 1 to N. C b represents the b-th energy storage device, with a value ranging from 1 to N. B n is the nth bus node, with values ​​ranging from 1 to N. N ; l represents the l-th transmission line, with a value ranging from 1 to N. L The start and end nodes are l (+) and l (−) ; s represents the s-th running scenario, with values ​​ranging from 1 to N. S The nth bus node is the set of conventional thermal power units. For the nth bus node, it is a set of wind farms or photovoltaic power stations; For the nth bus node, the set of concentrated solar power plants; For the nth bus node, the set of energy storage devices; For the i-th type of conventional thermal power, there is a set of conventional thermal power units. / This is a collection of existing lines / a collection of lines to be built; The annualized investment cost for conventional thermal power units is expressed in $ / kW-yr. Start-up and shutdown costs for conventional thermal power units, in k$; Variable operating costs for conventional thermal power units, in $ / kW; The installed capacity limit for conventional thermal power units, unit: MW; This represents the minimum output ratio for conventional thermal power units. / For conventional thermal power units, the ability to adjust ramp speed (up / down); / The minimum start-up / shutdown time for conventional thermal power units, in hours; The annualized investment cost of energy storage equipment is expressed in $ / kW-yr. The installed capacity limit for electric energy storage equipment, unit: MW; The energy storage capacity in hours for the energy storage device, in hours; The charging and discharging efficiency of energy storage devices; This represents the initial energy storage level of the energy storage device as a percentage. The annualized investment cost of a concentrated solar power (CSP) plant is expressed in $ / kW-yr. Variable operating costs of concentrated solar power (CSP) plants, in units of $ / kW; The installed capacity limit for concentrated solar power (CSP) plants, unit: MW; / For the up / down ramping capability of solar thermal power plants; This represents the initial percentage of heat storage level in the heat storage stage of a solar thermal power plant. The solar multiplier of a solar thermal power plant; The duration of thermal storage, in hours; The per-unit collector power curve of a solar thermal power plant; Annualized investment cost for wind farms or photovoltaic power plants, unit: $ / kW-yr; The installed capacity limit for wind farms or photovoltaic power plants, in MW; For wind farms / photovoltaic power plants, the per-unit predicted power curves are provided. This is the load demand curve, in MW. The annualized investment cost of transmission lines is expressed in $ / kW-yr. Line transmission capacity, unit: MW; For the reactance of the transmission line; / Renewable energy forecast error ratio / load forecast error ratio; The unit cost of load shedding for the system, in k$ / MWh; Let be the probability of the s-th running scenario; / Maximum budget for power supply investment / grid investment, in k$; The proportion of renewable energy generation in the system; The installed capacity of conventional thermal power units is expressed in MW. The installed capacity of energy storage equipment is expressed in MW. The installed capacity of wind farms or photovoltaic power plants is expressed in MW. The installed capacity of a solar thermal power unit is expressed in MW. This is a 0-1 state variable indicating whether or not to invest in the proposed railway line. Planned power output for conventional thermal power units, unit: MW; The online operating capacity of conventional thermal power units, in MW; / The rated capacity and shutdown capacity of conventional thermal power units are given in MW. / Power output for the planned charging / discharging of energy storage devices, unit: MW; The energy storage level of the electrical energy storage device, in MWh; The maximum possible output of the energy storage device, in MW; Planned power output of solar thermal power units, unit: MW; The thermal storage level of the solar thermal power unit is expressed in MWh. The maximum possible output of a solar thermal power unit, in MW; Power contribution to a wind farm or solar power plant, in MW; Load shedding capacity, unit: MW; Transmission power of a power transmission line, unit: MW; The phase angle of the node.

6. The method for detecting the V2G regulation capability of electric vehicles according to claim 1 or 2, characterized in that, The electric vehicle load sub-model recreates the random behavior characteristics of EVs through random behavior analysis based on the Monte Carlo algorithm. By observing or sampling user behavior habits and analyzing the development trends of power battery technology, it calculates the probability distributions of the EV's daily grid connection time, grid connection duration, daily mileage, and battery capacity parameters. Random parameters are then repeatedly generated from these probability distributions as the basic input parameters for the simulation of each electric vehicle unit in the model. Specifically, these parameters are: EV grid connection duration, battery capacity, and daily mileage. Where μ and σ are the expectation and standard deviation of a Gaussian distribution, respectively. The initial remaining charge of the EV when it first connects to the grid is linearly related to the battery capacity and daily mileage, and therefore also follows a Gaussian distribution. The daily grid connection time of the EV follows a weighted Gaussian mixture distribution, i.e. Where: K is the number of Gaussian distribution components, k i Let μ be the weight of the i-th Gaussian distribution. i and σ i Let be the expectation and standard deviation of the i-th Gaussian distribution, respectively. Suppose there are N EVs in the region. Then, for the n-th EV, the initial connection to the power grid... , where: Cap ev (n) represents the battery capacity, S day (n) represents the daily mileage, C pwr (n) represents the energy consumption per unit distance, f cha (n) represents the average daily charging frequency.

7. The method for detecting the V2G regulation capability of electric vehicles according to claim 6, characterized in that, The relationship between net load and EV adjustment demand is simplified to a linear one. The adjustment demand guides the EV adjustment actions batch by batch. When a batch of EVs participates in system adjustment, the adjustment demand data is simultaneously corrected to guide the charging and discharging behavior of the next batch of EVs, so that the EVs gradually approach the adjustment demand. Specifically, the nth EV reduces its load at time t. Where: Pev(n-1,t) is the charging power of the (n-1)th batch of vehicles at time t. When EVs participate in the adjustment, the maximum depth of discharge and off-grid SOC constraints are further considered to ensure user travel needs, specifically: , where: SOC min It is the minimum depth of discharge of the battery, SOC. end T represents the minimum expected battery level when the EV is disconnected from the grid. end This is a collection of EV disconnection times.

8. The method for detecting the V2G regulation capability of electric vehicles according to claim 5, characterized in that, Add V2G regulation inertia constraints to account for the subsequent impact of V2G on the power system. The constraints include: Where: n is the car serial number, t is the time, and P evb (n,t) represents the EV charging power when not involved in regulation, P ev (n,t) represents the EV charging power after participating in V2G (positive for charging, negative for discharging), L lim+ L(t) is the maximum net load limit within the region, and L(t) is the net load value within the region. EVs that do not meet the inertial constraint conditions stop V2G discharge regulation and only optimize their own charging behavior.

9. The method for detecting the V2G regulation capability of electric vehicles according to claim 8, characterized in that, The optimization specifically includes: during the period of a single EV grid connection, the model combines the GTEP sub-model to adjust demand data, and the target adjustment power at each time point is updated in descending order, specifically as follows: , where: P sf (n,t) represents the required load reduction of the system, P sf '(n,k) is P sf The new sequence after sorting (n,t) is given, where k is the index and I(n,k) is the time corresponding to index k. This leads to the specific EV charging strategy and SOC changes: , For k0, we have , where: P chg The rated power of the charging pile is SOC0(n), which is the initial SOC when connected to the power grid. ev (n,t) represents the EVSOC when no adjustment is involved, and Cap ev (n) represents the battery capacity.

Images