Active power distribution network multi-mode electric vehicle charging and discharging bearing capacity evaluation method

Abstract

The invention discloses an active power distribution network multi-mode electric vehicle charging and discharging bearing capacity evaluation method, which belongs to the technical field of smart power grids, and comprises the following steps: firstly, aiming at the behavior randomness of an electric vehicle, constructing a time sequence scene containing arrival time, departure time and an initial charge state, and establishing three types of charging and discharging mode constraints; generating a time sequence demand curve through Latin hypercube sampling; establishing each-phase voltage and current steady-state constraint and distributed power supply constraint by taking the maximum bearing capacity as a target and combining the characteristics of the power distribution network, and constructing an evaluation model; and finally, solving by adopting mixed integer linear programming, and linearizing the nonlinear constraint to obtain a linear model capable of being efficiently solved, so as to realize evaluation of the bearing capacity of the electric vehicle. The method focuses on the core demand of large-scale electric vehicle bearing capacity evaluation, overcomes the limitation that a traditional model is difficult to adapt to complex characteristics of a three-phase unbalanced active power distribution network, and realizes collaborative optimization of the electric vehicle bearing capacity and safe and stable operation of the power distribution network.

Description

Technical Field

[0001] This invention belongs to the field of smart grid technology, specifically a method for evaluating the charging and discharging capacity of multimodal electric vehicles in active distribution networks. Background Technology

[0002] With the rapid growth of electric vehicle ownership, large-scale electric vehicle integration into distribution networks has become an inevitable trend in smart grid development. However, the charging behavior of electric vehicles is highly random, with random arrival and departure times and initial state of charge leading to problems such as exacerbated three-phase imbalance, voltage exceeding limits, and line overload in distribution networks. Traditional distribution network control is mostly based on the assumption of three-phase balance and often uses a single charging mode to assess the carrying capacity of electric vehicles, making it difficult to adapt to the complex characteristics of widespread distributed generation and load fluctuations in active distribution networks. Furthermore, the unschedulable charging power of electric vehicles under disordered charging modes further increases the operational pressure on distribution networks. Existing research on ordered charging and V2G modes largely fails to fully consider key factors such as voltage and current constraints and distributed generation operation limitations under three-phase imbalance scenarios, making it impossible to achieve coordinated optimization of electric vehicle carrying capacity and the safe and stable operation of the distribution network. In addition, traditional carrying capacity assessment models often face the problems of complex nonlinear constraint solutions and low computational efficiency, making it difficult to meet the real-time requirements of large-scale electric vehicle integration scenarios.

[0003] To effectively address the aforementioned challenges, a capacity assessment method is needed that can accurately characterize the randomness of electric vehicle behavior, adapt to the characteristics of three-phase unbalanced active distribution networks, and integrate multi-modal charging and discharging modes. This method would solve the uncertainty problems brought about by large-scale electric vehicle access, overcome the limitations of traditional models, and provide technical support for the safe and economical operation of distribution networks. Summary of the Invention

[0004] To address the aforementioned technical problems, this invention provides a method for evaluating the charging and discharging capacity of multimodal electric vehicles in active distribution networks. First, considering the stochastic nature of large-scale electric vehicle behavior, a time-series scenario encompassing arrival time, departure time, and initial state of charge is constructed. Charging and discharging power and state of charge constraints are established for disordered, ordered, and V2G modes, respectively. Based on the aforementioned time-series scenario and constraints for each mode, a Latin hypercube sampling method is used to generate electric vehicle time-series demand curves. Next, with maximizing the electric vehicle carrying capacity as the objective function, steady-state constraints are established for the voltage and current of each phase, taking into account the characteristics of a three-phase unbalanced active distribution network. Simultaneously, considering the rated capacity, active power upper and lower limits, and reactive power adjustment range constraints of various distributed power sources, a multimodal electric vehicle carrying capacity evaluation model is constructed under the three-phase unbalanced active distribution system scenario. A mixed-integer linear programming method is used to solve the model, linearizing nonlinear constraints such as voltage and current balance and electric vehicle operation to obtain an efficiently solvable linear model, achieving configuration optimization for multimodal electric vehicle charging and discharging.

[0005] To achieve the above objectives, the technical solution of the present invention is as follows: A method for evaluating the charging and discharging capacity of multimodal electric vehicles in an active distribution network, comprising the following steps:

[0006] S1. To address the stochastic nature of large-scale electric vehicle behavior, a time-series scenario is constructed, encompassing arrival time, departure time, and initial state of charge. Charging and discharging power and state of charge constraints are established under disordered, ordered, and V2G modes, respectively. Based on the aforementioned time-series scenario and constraints of each mode, the Latin hypercube sampling method is used to generate the time-series demand curve for electric vehicles.

[0007] S2. Taking the maximization of electric vehicle carrying capacity as the objective function, and considering the characteristics of a three-phase unbalanced active distribution network, steady-state constraints are established for the voltage and current of each phase. At the same time, considering the operating constraints of the three modes of electric vehicles, a multi-mode electric vehicle carrying capacity evaluation model is constructed in the scenario of a three-phase unbalanced active distribution system.

[0008] S3. Linearize the nonlinear constraints such as voltage and current balance and electric vehicle operating modes to obtain a linear model that can be solved efficiently. Then, use the branch and bound method or the cutting plane method to solve the model and realize the configuration optimization of charging and discharging of multi-mode electric vehicles.

[0009] Further in S1, considering the stochastic characteristics of large-scale electric vehicle behavior, a time-series scenario covering arrival time, departure time, and initial state of charge is constructed, specifically:

[0010] a) Electric vehicle arrival time model:

[0011] Using degrees of freedom Chi-square distribution modeling of electric vehicle arrival time The probability density function describes the probability trend of electric vehicles arriving in concentrated numbers starting in the evening:

[0012]

[0013] In the formula, Arrival time of electric vehicles , Let the degrees of freedom be the chi-square distribution. This is a gamma function.

[0014] In the interval The probability density function for inner truncation is:

[0015]

[0016] In the formula, for The probability of an electric vehicle arriving at a given time. It is a chi-square distribution with 8 degrees of freedom.

[0017] For the The probability that an electric vehicle will arrive within a given time interval. Equal to the truncated distribution Points on:

[0018]

[0019] In the formula, For the first The probability of an electric vehicle arriving within a time interval. , The first The start and end times of each time interval.

[0020] b) Electric vehicle departure time model:

[0021] Using degrees of freedom The chi-square distribution has the following probability density function:

[0022]

[0023] time The data is mapped to a scheduling time interval for truncation, and the continuous probability distribution is transformed into a discrete scheduling model.

[0024]

[0025]

[0026] In the formula, for Electric vehicle departure time The distribution, It is a chi-square distribution with 4 degrees of freedom. For the first The probability of an electric vehicle leaving within a given time interval.

[0027] c) Modeling the initial state of charge of electric vehicles:

[0028] The initial state of charge (SOC) of an electric vehicle is modeled using a truncated normal distribution:

[0029]

[0030]

[0031] In the formula, The mean Standard deviation The normal probability density function, The mean Standard deviation The cumulative distribution function of the normal distribution, For the first The electric vehicle's state of charge at each time interval.

[0032] By combining probability distributions, a time-series scenario containing the arrival time, departure time, and initial state of charge of electric vehicles is constructed, which accurately portrays the randomness of large-scale electric vehicle behavior and effectively avoids the prediction bias caused by traditional simplified distribution models.

[0033] Furthermore, in S1, charging and discharging power and state of charge constraints are established for disordered, ordered, and V2G modes, respectively, as follows:

[0034] Constructing binary variables , , The three charging modes for electric vehicles are identified by the following method:

[0035]

[0036]

[0037] In the formula, , , This is a binary variable; a value of 1 indicates participation in the current mode, and a value of 0 indicates non-participation in the current mode. To participate in electric vehicle charging nodes.

[0038] In three modes Constant-time electric vehicle state-of-charge constraints:

[0039]

[0040]

[0041] In the formula, , For electric vehicles Time Node Upper and lower limits of the state of charge at time, , These are the upper and lower limits of the state of charge of electric vehicles. , The state of charge of electric vehicles upon arrival and departure. For electric vehicles at nodes The upper limits of charging and discharging power. , To improve the charging and discharging efficiency of electric vehicles, This refers to the battery capacity of electric vehicles.

[0042] a) Disorderly charging mode of electric vehicles

[0043] In unordered charging mode, electric vehicle charging is not schedulable and is only limited by the charging station and the electric vehicle's own needs.

[0044]

[0045] In the formula, For nodes The Middle The disordered charging power of electric vehicles at the end of each time interval for Time, node Electric vehicle charging power, Binary variables for electric vehicle pattern recognition

[0046] Consider the constraints for updating the state of charge of an electric vehicle after charging:

[0047]

[0048]

[0049]

[0050]

[0051] In the formula, For nodes The Middle The state of charge of the electric vehicle at the end of a time interval. For time intervals, For nodes Electric vehicle charging efficiency , For nodes The Middle The upper and lower limits of the electric vehicle's state of charge at the end of each time interval.

[0052] b) Orderly charging mode for electric vehicles

[0053] Power can be dispatched in the orderly charging mode to adapt to changes in distribution network load.

[0054]

[0055]

[0056] In the formula, For nodes The Middle The orderly charging power of electric vehicles at the end of each time interval. node The Middle Electric vehicle charging power at the end of a time interval For nodes The upper limit of electric vehicle charging power, Binary variables for electric vehicle pattern recognition.

[0057]

[0058]

[0059]

[0060]

[0061] In the formula, For nodes The Middle The state of charge of the electric vehicle at the end of a time interval. For nodes Initial state of charge of electric vehicles For time intervals, For nodes Electric vehicle charging efficiency for Limits to the state of charge of electric vehicles at all times. for The upper limit of the state of charge of an electric vehicle at any given time.

[0062] c) Electric vehicle V2G mode

[0063] In V2G mode, the active power fed back from electric vehicles to the grid must meet the maximum power constraint:

[0064]

[0065]

[0066] In the formula, For nodes The Middle The electric vehicle charging and discharging power at the end of each time interval. For nodes The upper limit of charging and discharging power of electric vehicles The charging / discharging coefficient for electric vehicles is 0 or 1.

[0067]

[0068]

[0069]

[0070]

[0071] The above establishes charging and discharging power and state of charge constraints for disordered, ordered, and V2G modes respectively, distinguishes the load regulation characteristics under different modes, reduces evaluation errors caused by mode confusion, and significantly improves the accuracy and multi-mode adaptability of active distribution networks for electric vehicle load analysis.

[0072] Furthermore, in S1, based on the aforementioned time-series scenarios and constraints of each mode, the Latin hypercube sampling method is used to generate the time-series demand curve for electric vehicles, specifically as follows:

[0073] Latin hypercube sampling was performed on the arrival time, departure time, and initial state of charge (SOC) of the electric vehicle. The cumulative distribution functions of arrival time, departure time, and initial SOC were each divided into N equally probable intervals, with a probability width of [value missing] for each interval. For the first The probability range of each interval is... .

[0074] Arrival time, departure time, and initial state of charge samples:

[0075]

[0076]

[0077]

[0078]

[0079]

[0080]

[0081] In the formula, , where is the cumulative distribution function of arrival time. Let be the cumulative distribution function of departure time. Let be the cumulative distribution function of the charged state.

[0082] The N groups will be drawn The samples are randomly combined, and invalid samples are removed.

[0083] Based on binary variables, assign a charge / discharge mode to each sample:

[0084]

[0085] in Disordered mode Ordered mode V2G mode.

[0086] Combining the above-mentioned disordered, ordered, and V2G charging and discharging constraints, for each sample in the effective sample set... Calculate its value at each time interval. The charging and discharging power was used to obtain the demand curves for electric vehicles under different modes.

[0087]

[0088] In the formula For nodes No. Total power demand of electric vehicles in intervals.

[0089] Furthermore, in S2, with maximizing the carrying capacity of electric vehicles as the objective function, steady-state constraints are established for the voltage and current of each phase, taking into account the characteristics of a three-phase unbalanced active distribution network. Specifically:

[0090] The objective function is to maximize the load-bearing capacity of the electric vehicle.

[0091]

[0092] In the formula, The total load-bearing capacity of the electric vehicle system, for Time Node Upper limit of electric vehicle capacity in China for Time Node The number of electric vehicles that can be connected. This is a set of electric vehicle access nodes.

[0093] For a three-phase unbalanced active distribution network, steady-state operating constraints are formulated for each phase voltage and current using the right-angle components of the voltage and current phasors:

[0094] (1) Real part current balance constraint:

[0095]

[0096] In the formula, For nodes To the node The line in phase ,time The real part of the current below; For nodes To the node The line in phase ,time The real part of the current below; Electric vehicle injection nodes participating in V2G mode In phase ,time The real part of the current below; Injecting nodes into substations In phase ,time The real part of the current below; Distributed generator node In phase ,time The real part of the current below; Injecting nodes into switched capacitor banks In phase ,time The real part of the current below; For energy storage devices from nodes In phase ,time The real part of the current absorbed below; , Injecting nodes into orderly and disorderly charging modes for electric vehicles In phase ,time The real part of the current below; For nodes In phase ,time The real part of the load current; For demand response at nodes In phase ,time The real part of the current absorbed below; For the set of routes; For nodes time The number of electric vehicles.

[0097] (2) Imaginary part current balance constraint:

[0098]

[0099] In this formula, the meanings of the variables are similar to those in formula (17), but the imaginary part of the current is different.

[0100] (3) Voltage real and imaginary part drop constraints:

[0101]

[0102] In the formula, , For nodes In phase ,time The real and imaginary parts of the voltage. , For nodes In phase ,time The imaginary part of the voltage is below. For the line In phase and The resistance between them For the line In phase and The reactance between them For the line In phase ,time The real part of the current below, For the line In phase ,time The real part of the current below, It is a phase set.

[0103] (4) Nonlinear load current constraint:

[0104]

[0105] In the formula, , They are nodes In phase ,time The real and imaginary parts of the load current. , They are nodes In phase ,time The active and reactive power demand, , They are nodes In phase ,time The real and imaginary parts of the voltage.

[0106] (5) Voltage amplitude constraint:

[0107]

[0108] In the formula, , These are the lower and upper voltage limits, respectively.

[0109] (6) Line capacity constraints:

[0110]

[0111] In the formula, , The lines are respectively In phase ,time The real and imaginary parts of the current. For the line Maximum current limit.

[0112] The above approach establishes steady-state constraints on voltage and current for each phase based on the characteristics of a three-phase unbalanced active distribution network. This avoids the problems of voltage exceeding limits and line overload misjudgment caused by the traditional three-phase balance assumption, significantly reduces the operational risks caused by a single constraint, and improves the operational safety and carrying capacity assessment reliability of the distribution network.

[0113] Furthermore, in S2, considering the operational constraints of the three modes of electric vehicles simultaneously, a multimodal electric vehicle load-bearing capacity assessment model is constructed under the scenario of a three-phase unbalanced active power distribution system. Specifically:

[0114] Electric vehicle operation constraint modeling:

[0115] (1) Active and reactive power constraints of electric vehicles:

[0116]

[0117]

[0118] (2) Constraints on disordered charging of electric vehicles:

[0119]

[0120] (3) Constraints on orderly charging of electric vehicles:

[0121]

[0122] (4) Constraints on electric vehicles participating in V2G mode:

[0123]

[0124] Distributed power source constraint modeling:

[0125] (1) Active power and reactive power constraints:

[0126]

[0127]

[0128] In the formula, , For distributed power at nodes ,time The active and reactive power of the following , For injection nodes In phase ,time The real and imaginary parts of the current.

[0129] (2) Active power and reactive power limits:

[0130]

[0131]

[0132] In the formula, For distributed power at nodes The upper limit of active power, For distributed power at nodes The power factor limit.

[0133] The constraint modeling of the switched capacitor bank is as follows:

[0134] (1) Constraints on the real and imaginary parts of the switching capacitor bank current:

[0135]

[0136]

[0137] In the formula, , The switched capacitor bank at the node phase ,time The real and imaginary parts of the injected current. node The number of switched capacitor banks, , The first Group of switched capacitor banks at nodes phase ,time The real and imaginary parts of the injected current.

[0138] (2) Current constraints provided by each phase and each switched capacitor bank module:

[0139]

[0140]

[0141] In the formula, node The switching capacitor module susceptance.

[0142] Modeling the operational constraints of energy storage devices:

[0143] (1) Net active power injection constraint:

[0144]

[0145] In the formula, , For energy storage devices at nodes ,time Under the charging and discharging power, , For energy storage devices at nodes phase ,time The real and imaginary parts of the absorption current.

[0146] (2) Reactive power constraint:

[0147]

[0148] (3) Charge state update constraints:

[0149]

[0150]

[0151] In the formula, Energy storage devices at nodes ,time The state of charge under these conditions, For energy storage devices at nodes The initial state of charge, For energy storage devices at nodes ,time The charging power at the following levels For energy storage devices at nodes Charging efficiency, For energy storage devices at nodes ,time The discharge power at that time, For energy storage devices at nodes The discharge efficiency, , These are energy storage devices at nodes The minimum and maximum states of charge,

[0152] (4) Charging and discharging power limits:

[0153]

[0154] In the formula, , For nodes Minimum and maximum charging power , For nodes Minimum and maximum discharge power.

[0155] Time-sharing demand response modeling:

[0156] (1) Constraints on changes in active and reactive power:

[0157]

[0158]

[0159] In the formula, For the demand response procedure at the node Phase ,time The increase in active power, For the demand response procedure at the node Phase ,time Reduced active power, , For the demand response procedure at the node Phase ,time The real and imaginary parts of the current.

[0160] (2) Demand response procedure power variation constraint:

[0161]

[0162]

[0163]

[0164] In the formula, The maximum percentage of load change allowed for the node. Binary variable: Whether to allow increased load on the demand response node. A binary variable indicating whether load reduction is allowed at the demand response node.

[0165] (3) Reactive power constraints in demand response:

[0166]

[0167] In the formula, The power constraint angle for demand response.

[0168] The above considerations take into account the operational constraints of distributed power sources, energy storage devices, switched capacitor banks, and time-of-use demand response. A multi-constraint collaborative evaluation model is constructed to reduce operational risks caused by a single constraint and improve the reliability of distribution network operation safety and carrying capacity assessment.

[0169] Furthermore, in S3, nonlinear constraints such as voltage and current balance and electric vehicle operating modes are linearized, specifically as follows:

[0170] The nonlinear terms in constraint (24) are handled by first-order Taylor series expansion, and the function and At the estimated point Expand nearby:

[0171]

[0172]

[0173] In the formula, , , , functions respectively and Partial derivatives with respect to the real and imaginary parts of the voltage, , These are the estimated values ​​for the real and imaginary parts of the voltage, respectively.

[0174] For the nonlinear constraint of voltage in constraint (25) The real part of the voltage and the virtual part The relationship is defined by the phase angle boundary, and the constraint is ultimately decomposed into side boundary constraint 1, side boundary constraint 2, side boundary constraint 3, upper boundary constraint and lower boundary constraint.

[0175] Side boundary constraint 1 connects the minimum voltage amplitude and the maximum positive phase angle deviation, and the maximum voltage amplitude and the reference phase angle, limiting the upper limit of the imaginary part of the voltage. This corresponds to the boundary combination of the minimum voltage amplitude and the maximum positive phase angle deviation.

[0176]

[0177] In the formula, Phase A is the reference phase angle for each phase. Phase B C phase , For the maximum positive phase angle deviation, For the maximum negative phase angle deviation, This is the per-unit value of the lower limit of voltage amplitude.

[0178] Side boundary constraint 2 connects the maximum voltage amplitude and the reference phase angle with the maximum voltage amplitude and the maximum positive phase angle deviation, further constraining the upper limit of the imaginary part. On the other hand, it adapts to the positive phase angle deviation scenario under the maximum voltage amplitude.

[0179]

[0180] In the formula, Phase A is the reference phase angle for each phase. Phase B C phase , For the maximum positive phase angle deviation, This is the per-unit value of the lower limit of voltage amplitude.

[0181] Side boundary constraint 3 connects the two points of maximum voltage amplitude / reference phase angle and minimum voltage amplitude / maximum negative phase angle deviation, limiting the lower limit of the imaginary part of the voltage, corresponding to the boundary combination of maximum voltage amplitude and maximum negative phase angle deviation:

[0182]

[0183] In the formula, Phase A is the reference phase angle for each phase. Phase B C phase , For the maximum negative phase angle deviation, These are the per-unit values ​​of the upper limit of voltage amplitude.

[0184] The upper boundary constraint is defined by the tangent of the maximum positive phase angle deviation, which limits the upper slope of the imaginary part relative to the real part. To avoid the phase angle exceeding the maximum positive deviation:

[0185]

[0186] In the formula, Phase A is the reference phase angle for each phase. Phase B C phase , This represents the maximum positive phase angle deviation.

[0187] The lower boundary constraint is defined by the tangent of the maximum negative phase angle deviation, which limits the lower slope of the imaginary part relative to the real part. To avoid the phase angle exceeding the maximum negative deviation:

[0188]

[0189] In the formula, For each phase reference phase angle Phase B C phase , This represents the maximum negative phase angle deviation.

[0190] Line capacity nonlinearity constraint in constraint (26) The constraint is transformed into a linear constraint through variable decomposition and piecewise summation.

[0191] The real part of the current and the virtual part Decomposed into positive and negative components:

[0192]

[0193] In the formula, , These are the positive and negative components of the real part, respectively. , These are the positive and negative components of the imaginary part, respectively.

[0194]

[0195] Define real segment components and imaginary segment components The range of values ​​for each segment variable is: ,in :

[0196]

[0197] In the formula, For real segment components, For imaginary segment components, This represents the number of equally spaced intervals.

[0198] The piecewise linear fitting expression for the quadratic term is:

[0199]

[0200] In the formula, This is a linear approximation of the second term of the current. This is limited by the maximum heat capacity.

[0201] The above linearization of nonlinear constraints such as voltage and current balance and distributed power source operation addresses the problems of low solution efficiency and difficulty in adapting traditional nonlinear models to large-scale electric vehicle access scenarios.

[0202] Furthermore, in S3, a linear model that can be solved efficiently is obtained, and the branch-and-bound method or the cutting plane method is used to solve the model, thereby achieving configuration optimization for charging and discharging of multimodal electric vehicles, specifically:

[0203] Through the above linearization process, the complete model is transformed into a mixed-integer linear programming model.

[0204] Consider the model of electric vehicles participating in three charging modes:

[0205]

[0206] The model that maximizes the load-bearing capacity of electric vehicles when they only participate in disordered charging:

[0207]

[0208] The model for maximizing the load-bearing capacity of electric vehicles when participating in disordered and ordered charging:

[0209]

[0210] The transformed mixed-integer linear programming model is solved using the branch-and-bound method or the cutting plane method, and the total carrying capacity of electric vehicles in the target distribution network is finally obtained.

[0211] Construct a quantitative index model for the load-bearing capacity of electric vehicles, and the load-bearing capacity improvement rate of the ordered mode relative to the disordered mode:

[0212]

[0213] The capacity improvement rate of V2G mode compared to ordered mode:

[0214] .

[0215] Compared to existing technologies, this solution has the following advantages:

[0216] 1) By constructing a time-series scenario including the arrival time, departure time, and initial state of charge of electric vehicles, and combining probability distribution and Latin hypercube sampling to generate time-series demand curves, the randomness of large-scale electric vehicle behavior is characterized, effectively avoiding demand forecasting bias caused by traditional simplified distribution models. Charging and discharging power and state of charge constraints are established for disordered, ordered, and V2G modes respectively, distinguishing the load regulation characteristics under different modes, reducing evaluation errors caused by mode confusion, and significantly improving the accuracy and multi-mode adaptability of the active distribution network for electric vehicle load analysis.

[0217] 2) This invention aims to maximize the carrying capacity of electric vehicles. It establishes steady-state constraints on voltage and current for each phase based on the characteristics of a three-phase unbalanced active distribution network, while also incorporating operational constraints from distributed power sources, energy storage devices, switched capacitor banks, and time-sharing demand response. This constructs a multi-constraint collaborative evaluation model. It avoids the voltage over-limit and line overload misjudgment problems caused by the traditional three-phase balance assumption, significantly reduces operational risks caused by a single constraint, and improves the operational safety and carrying capacity assessment reliability of the distribution network.

[0218] 3) By employing mixed-integer linear programming, nonlinear constraints such as voltage and current balance and distributed power source operation are linearized, addressing the issues of low solution efficiency and difficulty in adapting to large-scale electric vehicle access scenarios in traditional nonlinear models. Simultaneously, a quantitative index of carrying capacity is constructed to quantify the effect of multi-mode integration on carrying capacity improvement, meeting real-time assessment needs and providing quantitative basis for the planning and scheduling strategies of power distribution network charging facilities. Attached Figure Description

[0220] Figure 1 This is a flowchart of the method described in this invention;

[0221] Figure 2 This is a diagram showing the load nodes (1, 2, and 3) and charging station nodes.

[0222] Figure 3 This is a schematic diagram of the overall strategy framework of the method described in this invention. Detailed Implementation

[0224] To enhance understanding of the present invention, the application is described in detail below with reference to embodiments.

[0225] Example: Figure 1 A schematic flowchart of a method for evaluating the charging and discharging capacity of multimodal electric vehicles in an active distribution network, provided as an example of the present invention; Figure 1 As shown, the method of the present invention includes the following steps:

[0226] S1. To address the stochastic nature of large-scale electric vehicle (EV) integration into the active distribution network, a time-series scenario is constructed, encompassing EV arrival time, departure time, and initial state of charge (SOC). Binary variables are introduced to identify three charging / discharging modes: disordered, ordered, and V2G (Vehicle-to-Grid). Charging / discharging power limits, SOC updates, and boundary constraints are established for each mode. Finally, based on the aforementioned time-series scenario, a Latin hypercube sampling method is used to generate a time-series demand curve with load characteristics. Specifically:

[0227] First, considering the stochastic nature of large-scale electric vehicle (EV) access to the active distribution network, a time-series scenario covering EV arrival time, departure time, and initial state of charge is constructed. The steps are as follows:

[0228] a) Electric vehicle arrival time model:

[0229]

[0230] In the formula, for The probability of an electric vehicle arriving at a given time. It is a chi-square distribution with 8 degrees of freedom.

[0231] For the The probability that an electric vehicle will arrive within a given time interval. Equal to the truncated distribution Points on:

[0232]

[0233] In the formula, For the first The probability of an electric vehicle arriving within a time interval. , The first The start and end times of each time interval.

[0234] b) Electric vehicle departure time model:

[0235] Using degrees of freedom The chi-square distribution has the following probability density function:

[0236]

[0237] time The data is mapped to a scheduling time interval for truncation, and the continuous probability distribution is transformed into a discrete scheduling model.

[0238]

[0239]

[0240] In the formula, for Electric vehicle departure time The distribution, It is a chi-square distribution with 4 degrees of freedom. For the first The probability of an electric vehicle leaving within a time interval;

[0241] c) Modeling the initial state of charge of electric vehicles;

[0242] The initial state of charge (SOC) of an electric vehicle is modeled using a truncated normal distribution:

[0243]

[0244]

[0245] In the formula, The mean Standard deviation The normal probability density function, The mean Standard deviation The cumulative distribution function of the normal distribution, For the first The electric vehicle's state of charge at each time interval.

[0246] Subsequently, binary variables are introduced to identify three charging and discharging modes: disordered, ordered, and V2G. Charging and discharging power limits, state of charge updates, and boundary constraints are established for each mode, clarifying the operating rules and load regulation characteristics of different modes. The steps are as follows:

[0247] Constructing binary variables , , The three charging modes for electric vehicles are identified by the following method:

[0248]

[0249]

[0250] In the formula, , , This is a binary variable; a value of 1 indicates participation in the current mode, and a value of 0 indicates non-participation in the current mode. To participate in electric vehicle charging nodes.

[0251] In three modes Constant-time electric vehicle state-of-charge constraints:

[0252]

[0253]

[0254] In the formula, , For electric vehicles Time Node Upper and lower limits of the state of charge at time, , These are the upper and lower limits of the state of charge of electric vehicles. , The state of charge of electric vehicles upon arrival and departure. For electric vehicles at nodes The upper limits of charging and discharging power. , To improve the charging and discharging efficiency of electric vehicles, For electric vehicle battery capacity,

[0255] a) Disorderly charging mode of electric vehicles

[0256] In unordered charging mode, electric vehicle charging is not schedulable and is only limited by the charging station and the electric vehicle's own needs.

[0257]

[0258] In the formula, For nodes The Middle The disordered charging power of electric vehicles at the end of each time interval for Time, node Electric vehicle charging power, Binary variables for electric vehicle pattern recognition

[0259] Consider the constraints for updating the state of charge of an electric vehicle after charging:

[0260]

[0261]

[0262]

[0263]

[0264] In the formula, For nodes The Middle The state of charge of the electric vehicle at the end of a time interval. For time intervals, For nodes Electric vehicle charging efficiency , For nodes The Middle At the end of each time interval, the upper and lower limits of the electric vehicle's state of charge are defined.

[0265] b) Orderly charging mode for electric vehicles

[0266] Power can be dispatched in the orderly charging mode to adapt to changes in distribution network load.

[0267]

[0268]

[0269] In the formula, For nodes The Middle The orderly charging power of electric vehicles at the end of each time interval. For nodes The upper limit of electric vehicle charging power, Binary variables for electric vehicle pattern recognition

[0270]

[0271]

[0272]

[0273]

[0274] In the formula, For nodes The Middle The state of charge of the electric vehicle at the end of a time interval. For nodes Initial state of charge of electric vehicles For time intervals, For nodes Electric vehicle charging efficiency This is the minimum state of charge for electric vehicles. This represents the upper limit of the state of charge for electric vehicles.

[0275] c) Electric vehicle V2G mode

[0276] In V2G mode, the active power fed back from electric vehicles to the grid must meet the maximum power constraint:

[0277]

[0278]

[0279] In the formula, For nodes The Middle The electric vehicle charging and discharging power at the end of each time interval. For nodes The upper limit of charging and discharging power of electric vehicles The charging / discharging coefficient for electric vehicles is 0 or 1;

[0280]

[0281]

[0282]

[0283] .

[0284] Finally, based on the aforementioned time-series scenarios and multi-mode constraints, the Latin hypercube sampling method is used to generate time-series demand curves with load characteristics. The steps are as follows:

[0285] Latin hypercube sampling was performed on the arrival time, departure time, and initial state of charge (SOC) of the electric vehicle. The cumulative distribution functions of arrival time, departure time, and initial SOC were each divided into N equally probable intervals, with a probability width of [value missing] for each interval. For the first The probability range of each interval is... ,

[0286] Arrival time, departure time, and initial state of charge samples:

[0287]

[0288]

[0289]

[0290]

[0291]

[0292]

[0293] In the formula, , where is the cumulative distribution function of arrival time. Let be the cumulative distribution function of departure time. Let be the cumulative distribution function of the state of charge.

[0294] The N groups will be drawn The samples are randomly combined, and invalid samples are removed.

[0295] Based on binary variables, assign a charge / discharge mode to each sample:

[0296]

[0297] in Disordered mode Ordered mode V2G mode,

[0298] Combining the above-mentioned disordered, ordered, and V2G charging and discharging constraints, for each sample in the effective sample set... Calculate its value at each time interval. The charging and discharging power was used to obtain the electric vehicle demand curves under different modes.

[0299]

[0300] In the formula For nodes No. Total power demand of electric vehicles in intervals.

[0301] S2. Taking maximizing the carrying capacity of the active distribution network for electric vehicles as the objective function, and considering the operating characteristics of a three-phase unbalanced active distribution network, independent steady-state constraints are established for each phase from three dimensions: current, voltage, and line. Operating constraints of distributed power sources, energy storage devices, switched capacitor banks, and time-of-use demand response are incorporated to construct a carrying capacity assessment model for a three-phase unbalanced active distribution system that considers the charging and discharging demands of electric vehicles and the safe operation of the distribution network. Specifically:

[0302] First, taking maximizing the carrying capacity of the active distribution network for electric vehicles as the objective function, and considering the operating characteristics of a three-phase unbalanced active distribution network, independent steady-state constraints are established for each phase from three dimensions: current, voltage, and line. The steps are as follows:

[0303] The objective function is to maximize the load-bearing capacity of the electric vehicle.

[0304]

[0305] In the formula, The total load-bearing capacity of the electric vehicle system, for Time Node Upper limit of electric vehicle capacity in China for Time Node The number of electric vehicles that can be connected. For electric vehicle access nodes,

[0306] For a three-phase unbalanced active distribution network, steady-state operating constraints are formulated for each phase voltage and current using the right-angle components of the voltage and current phasors:

[0307] (1) Real part current balance constraint:

[0308]

[0309] In the formula, For nodes To the node The line in phase ,time The real part of the current below; For nodes To the node The line in phase ,time The real part of the current below; Electric vehicle injection nodes participating in V2G mode In phase ,time The real part of the current below; Injecting nodes into substations In phase ,time The real part of the current below; Distributed generator node In phase ,time The real part of the current below; Injecting nodes into switched capacitor banks In phase ,time The real part of the current below; For energy storage devices from nodes In phase ,time The real part of the current absorbed below; , Injecting nodes into orderly and disorderly charging modes for electric vehicles In phase ,time The real part of the current below; For nodes In phase ,time The real part of the load current; For demand response at nodes In phase ,time The real part of the current absorbed below; For the set of routes; For nodes time The number of electric vehicles.

[0310] (2) Imaginary part current balance constraint:

[0311]

[0312] In the formula, in the formula, For nodes To the node The line in phase ,time The imaginary part of the current below; For nodes To the node The line in phase ,time The imaginary part of the current below; Electric vehicle injection nodes participating in V2G mode In phase ,time The imaginary part of the current below; Injecting nodes into substations In phase ,time The imaginary part of the current below; Distributed generator node In phase ,time The imaginary part of the current below; Injecting nodes into switched capacitor banks In phase ,time The imaginary part of the current below; For energy storage devices from nodes In phase ,time The imaginary part of the current absorbed below; , Injecting nodes into orderly and disorderly charging modes for electric vehicles In phase ,time The imaginary part of the current below; For nodes In phase ,time The imaginary part of the load current; For demand response at nodes In phase ,time The imaginary part of the current absorbed below; This is a set of routes.

[0313] (3) Voltage real and imaginary part drop constraints:

[0314]

[0315] In the formula, , For nodes In phase ,time The real and imaginary parts of the voltage. , For nodes In phase ,time The imaginary part of the voltage is below. For the line In phase and The resistance between them For the line In phase and The reactance between them For the line In phase ,time The real part of the current below, For the line In phase ,time The real part of the current below, It is a phase set.

[0316] (4) Nonlinear load current constraint:

[0317]

[0318] In the formula, , They are nodes In phase ,time The real and imaginary parts of the load current. , They are nodes In phase ,time The active and reactive power demand, , They are nodes In phase ,time The real and imaginary parts of the voltage.

[0319] (5) Voltage amplitude constraint:

[0320]

[0321] In the formula, , These are the lower and upper voltage limits, respectively.

[0322] (6) Line capacity constraints:

[0323]

[0324] In the formula, , The lines are respectively In phase ,time The real and imaginary parts of the current. For the line Maximum current limit.

[0325] Electric vehicle operation constraint modeling:

[0326] (1) Active power and reactive power constraints of electric vehicles

[0327]

[0328]

[0329] (2) Constraints on disordered charging of electric vehicles:

[0330]

[0331] (3) Constraints on orderly charging of electric vehicles:

[0332]

[0333] (4) Constraints on electric vehicles participating in V2G mode:

[0334]

[0335] Secondly, considering the operational limitations of distributed power sources, energy storage devices, switched capacitor banks, and time-of-use demand response, a capacity assessment model for the three-phase unbalanced active distribution system is constructed to evaluate the charging and discharging demands of multimodal electric vehicles and the safe operation of the distribution network. The steps are as follows:

[0336] Distributed power source constraint modeling:

[0337] (1) Active power and reactive power constraints:

[0338]

[0339]

[0340] In the formula, , For distributed power at nodes ,time The active and reactive power of the following , For injection nodes In phase ,time The real and imaginary parts of the current.

[0341] (2) Active power and reactive power limits:

[0342]

[0343]

[0344] In the formula, For distributed power at nodes The upper limit of active power, For distributed power at nodes The power factor limit,

[0345] The constraint modeling of the switched capacitor bank is as follows:

[0346] (1) Constraints on the real and imaginary parts of the switching capacitor bank current:

[0347]

[0348]

[0349] In the formula, , The switched capacitor bank at the node phase ,time The real and imaginary parts of the injected current. node The number of switched capacitor banks, , The first Group of switched capacitor banks at nodes phase ,time The real and imaginary parts of the injected current.

[0350] (2) Current constraints provided by each phase and each switched capacitor bank module:

[0351]

[0352]

[0353] In the formula, node The switching capacitor module susceptance,

[0354] Modeling the operational constraints of energy storage devices:

[0355] (1) Net active power injection constraint:

[0356]

[0357] In the formula, , For energy storage devices at nodes ,time Under the charging and discharging power, , For energy storage devices at nodes phase ,time The real and imaginary parts of the absorption current.

[0358] (2) Reactive power constraint:

[0359]

[0360] (3) Charge state update constraints:

[0361]

[0362]

[0363] In the formula, Energy storage devices at nodes ,time The state of charge under these conditions, For energy storage devices at nodes The initial state of charge, For energy storage devices at nodes ,time The charging power at the following levels For energy storage devices at nodes Charging efficiency, For energy storage devices at nodes ,time The discharge power at that time, For energy storage devices at nodes The discharge efficiency, , These are energy storage devices at nodes The minimum and maximum states of charge,

[0364] (4) Charging and discharging power limits:

[0365]

[0366] In the formula, , For nodes Minimum and maximum charging power , For nodes Minimum and maximum discharge power

[0367] Time-sharing demand response modeling:

[0368] (1) Constraints on changes in active and reactive power:

[0369]

[0370]

[0371] In the formula, For the demand response procedure at the node Phase ,time The increase in active power, For the demand response procedure at the node Phase ,time Reduced active power, , For the demand response procedure at the node Phase ,time The real and imaginary parts of the current.

[0372] (2) Demand response procedure power variation constraint:

[0373]

[0374]

[0375]

[0376] In the formula, The maximum percentage of load change allowed for the node. Binary variable: Whether to allow increased load on the demand response node. Binary variables, whether load reduction is allowed at the demand response node,

[0377] (3) Reactive power constraints in demand response:

[0378]

[0379] In the formula, The power constraint angle for demand response.

[0380] S3. A mixed-integer linear programming method is used to solve the evaluation model. Through first-order Taylor series expansion, variable decomposition, and piecewise linear fitting, nonlinear constraints such as voltage and current balance and line capacity are linearized, transforming the model into a linear model. Finally, a multimodal electric vehicle load-bearing capacity model is constructed to complete the load-bearing capacity assessment. Specifically:

[0381] First, nonlinear constraints such as voltage and current balance and line capacity are linearized. The steps are as follows:

[0382] After linearizing constraint (24), we get:

[0383]

[0384]

[0385] In the formula, , , , functions respectively and Partial derivatives with respect to the real and imaginary parts of the voltage, , These are the estimated values ​​for the real and imaginary parts of the voltage, respectively.

[0386] After linearizing the voltage in constraint (25), we get:

[0387]

[0388] In the formula, Phase A is the reference phase angle for each phase. Phase B C phase , For the maximum positive phase angle deviation, For the maximum negative phase angle deviation, This is the per-unit value of the lower limit of voltage amplitude.

[0389]

[0390] In the formula, Phase A is the reference phase angle for each phase. Phase B C phase , For the maximum positive phase angle deviation, This is the per-unit value of the lower limit of voltage amplitude.

[0391]

[0392] In the formula, Phase A is the reference phase angle for each phase. Phase B C phase , For the maximum negative phase angle deviation, These are the per-unit values ​​of the upper limit of voltage amplitude.

[0393]

[0394] In the formula, Phase A is the reference phase angle for each phase. Phase B C phase , For the maximum positive phase angle deviation,

[0395]

[0396] In the formula, For each phase reference phase angle Phase B C phase , This represents the maximum negative phase angle deviation.

[0397] Line capacity nonlinearity constraint in constraint (44) By decomposing variables and summing pieces, the problem is transformed into a linear constraint. The piecewise linear fitting expression for the quadratic term is:

[0398]

[0399] In the formula, This is a linear approximation of the second term of the current. Due to maximum heat capacity limitations, For the first The segmentation coefficient of the segment. , The lines are respectively No. At time t, the first The real and imaginary parts of a segment are segment variables. This represents the total number of segments in the piecewise linear fit.

[0400] Finally, the model is transformed into a linear model, and a multimodal electric vehicle load-bearing capacity model is constructed. Load-bearing capacity evaluation indices are also developed. The steps are as follows:

[0401] Consider the model of electric vehicles participating in three charging modes:

[0402]

[0403] Maximize the load-bearing capacity of electric vehicles when they only participate in disordered charging. Model:

[0404]

[0405] When electric vehicles participate in disorderly or orderly charging, the carrying capacity of electric vehicles should be maximized. Model:

[0406]

[0407] Construct a quantitative index model for the load-bearing capacity of electric vehicles, and the load-bearing capacity improvement rate of the ordered mode relative to the disordered mode:

[0408]

[0409] The capacity improvement rate of V2G mode compared to ordered mode:

[0410] .

[0411] Experimental verification

[0412] The above model was tested on an IEEE 4.16kV, 5MVA, 123-node unbalanced system. The distributed energy sources installed in the system are as follows: two distributed power sources with the following parameters. , Two sets of switching capacitor banks, with the following parameters: 5 energy storage devices, with the following parameters: , , , , Consider two types of electric vehicles: parameters are... Charging and discharging power ; parameters are Charging and discharging power The node voltage limit is set to... , .

[0413] The simulation was performed on a computer equipped with an Intel(R) Xeon(R) CPU E5-2650v4 (2.20GHz) processor and 64GB of memory. The simulation environment was a 64-bit Windows operating system with an Intel(R) Core(TM) i7-7700 CPU @3.6GHz and 64GB of memory.

[0414] Two distributed power sources are installed at nodes 13 and 97; three sets of switching capacitor banks are installed at nodes 85 (phase c), 96 (phase b), and 114 (phase a). The access node and phase of each electric vehicle are randomly assigned, treating it as an independent load. The simulation considers two electric vehicle fleets: 2500 vehicles and 5000 vehicles.

[0415]

[0416] As shown in Table 1, the first two columns represent the number of electric vehicles and the objective function value, respectively. The first set of simulation results shows that when the number of electric vehicles is 2500, the objective function value is 4.46 MVA. The second set of simulation results shows that as the number of electric vehicles increases to 5000, the objective function value decreases to 4.07 MVA.

[0417] Furthermore, the demand for electric vehicles has been largely met: despite However, the energy supply from electric vehicles remains within this limit. Finally, due to the rising demand curve, electric vehicle charging leads to increased daily energy consumption.

[0418]

[0419] As shown in Table 2, the system is simulated considering disordered charging, ordered charging, and V2G modes for electric vehicles to evaluate the maximization effect of carrying capacity. Two distributed power sources are installed at nodes 13 and 97; three sets of switching capacitor banks are installed at nodes 85 (phase c), 96 (phase b), and 114 (phase a). Electric vehicle type is considered: parameters are... The quantity is 2500.

[0420] The above description is merely a preferred embodiment of the present invention and is not intended to further limit the present invention. All equivalent changes made based on the description and drawings of the present invention are within the protection scope of the present invention.

Claims

1. A method for evaluating the charging and discharging capacity of multi-mode electric vehicles in an active distribution network, characterized in that, Includes the following steps: S1. To address the stochastic nature of large-scale electric vehicle behavior, a time-series scenario encompassing arrival time, departure time, and initial state of charge is constructed. Charging and discharging power and state of charge constraints are established for disordered, ordered, and V2G modes, respectively. Based on the aforementioned time-series scenario and constraints for each mode, the Latin hypercube sampling method is used to generate the electric vehicle time-series demand curve. S2. Taking maximizing the load-bearing capacity of electric vehicles as the objective function, and considering the characteristics of a three-phase unbalanced active distribution network, steady-state constraints are established for the voltage and current of each phase. Simultaneously, considering the operating constraints of three modes of electric vehicles, a multi-modal electric vehicle load-bearing capacity evaluation model is constructed for a three-phase unbalanced active distribution system scenario. S3. Linearize the nonlinear constraints such as voltage and current balance and electric vehicle operating modes to obtain a linear model that can be solved efficiently. Then, use the branch and bound method or the cutting plane method to solve the model and realize the configuration optimization of charging and discharging of multi-mode electric vehicles.

2. The method for evaluating the charging and discharging capacity of multi-mode electric vehicles in an active distribution network according to claim 1, characterized in that, In S1, considering the stochastic characteristics of large-scale electric vehicle behavior, a time-series scenario covering arrival time, departure time, and initial state of charge is constructed, specifically: a) Electric vehicle arrival time model: Using degrees of freedom Chi-square distribution modeling of electric vehicle arrival time The probability density function describes the probability trend of electric vehicles arriving in concentrated numbers starting in the evening: In the formula, Arrival time of electric vehicles , Let the degrees of freedom be the chi-square distribution. For gamma function, In the interval The probability density function for inner truncation is: In the formula, for The probability of an electric vehicle arriving at a given time. It is a chi-square distribution with 8 degrees of freedom. For the The probability that an electric vehicle will arrive within a given time interval. Equal to the truncated distribution Points on: In the formula, For the first The probability of an electric vehicle arriving within a time interval. , The first The start and end times of each time interval. b) Electric vehicle departure time model: Using degrees of freedom The chi-square distribution has the following probability density function: time The data is mapped to a scheduling time interval for truncation, and the continuous probability distribution is transformed into a discrete scheduling model. In the formula, for Electric vehicle departure time The distribution, It is a chi-square distribution with 4 degrees of freedom. For the first The probability of an electric vehicle leaving within a time interval; c) Modeling the initial state of charge of electric vehicles; The initial state of charge (SOC) of an electric vehicle is modeled using a truncated normal distribution: In the formula, The mean Standard deviation The normal probability density function, The mean Standard deviation The cumulative distribution function of the normal distribution, For the first The electric vehicle's state of charge at each time interval.

3. The method for evaluating the charging and discharging capacity of multi-mode electric vehicles in an active distribution network according to claim 1, characterized in that, In S1, the charging and discharging power and state of charge constraints are established for disordered, ordered, and V2G modes, respectively, as follows: Constructing binary variables , , The three charging modes for electric vehicles are identified by the following method: In the formula, , , This is a binary variable; a value of 1 indicates participation in the current mode, and a value of 0 indicates non-participation in the current mode. To participate in electric vehicle charging nodes. In three modes Constant-time electric vehicle state-of-charge constraints: In the formula, , For electric vehicles Time Node Upper and lower limits of the state of charge at time, , The state of charge of electric vehicles upon arrival and departure. For electric vehicles at nodes The upper limits of charging and discharging power. , To improve the charging and discharging efficiency of electric vehicles, For electric vehicle battery capacity, a) Disorderly charging mode of electric vehicles In unordered charging mode, electric vehicle charging is not schedulable and is only limited by the charging station and the electric vehicle's own needs. In the formula, For nodes The Middle The disordered charging power of electric vehicles at the end of each time interval for Time, node Electric vehicle charging power, Binary variables for electric vehicle pattern recognition Consider the state-of-charge update constraints after disordered charging of electric vehicles: In the formula, For nodes The Middle The state of charge of the electric vehicle at the end of a time interval. For time intervals, For nodes Electric vehicle charging efficiency , For nodes The Middle At the end of each time interval, the upper and lower limits of the electric vehicle's state of charge are defined. b) Orderly charging mode for electric vehicles Power can be dispatched in the orderly charging mode to adapt to changes in distribution network load. In the formula, For nodes The Middle The orderly charging power of electric vehicles at the end of each time interval. Binary variables for electric vehicle pattern recognition For nodes The upper limit of electric vehicle charging power, Consider the state-of-charge update constraints after orderly charging of electric vehicles: In the formula, For nodes The Middle The state of charge of the electric vehicle at the end of a time interval. For nodes middle The electric vehicle's state of charge at all times. For time intervals, c) Electric vehicle V2G mode In V2G mode, the active power fed back from electric vehicles to the grid must meet the maximum power constraint: In the formula, For nodes The Middle The electric vehicle charging and discharging power at the end of each time interval. The charging / discharging coefficient for electric vehicles is 0 or 1. For nodes The Middle Electric vehicle charging and discharging power at time intervals Consider the state-of-charge update constraints of electric vehicles after charging in V2G mode: 。 4. The method for evaluating the charging and discharging capacity of multi-mode electric vehicles in an active distribution network according to claim 1, characterized in that, In S1, based on the aforementioned time-series scenarios and constraints of each mode, the Latin hypercube sampling method is used to generate the time-series demand curve for electric vehicles, specifically: Latin hypercube sampling was performed on the arrival time, departure time, and initial state of charge (SOC) of the electric vehicle. The cumulative distribution functions of arrival time, departure time, and initial SOC were each divided into N equally probable intervals, with a probability width of [value missing] for each interval. For the first The probability range of each interval is... , Arrival time, departure time, and initial state of charge samples: In the formula, , where is the cumulative distribution function of arrival time. Let be the cumulative distribution function of departure time. Let be the cumulative distribution function of the state of charge. The N groups will be drawn The samples are randomly combined, and invalid samples are removed. Based on binary variables, assign a charge / discharge mode to each sample: in Disordered mode Ordered mode V2G mode, Combining the above-mentioned disordered, ordered, and V2G charging and discharging constraints, for each sample in the effective sample set... Calculate its value at each time interval. The charging and discharging power was used to obtain the electric vehicle demand curves under different modes. In the formula For nodes No. Total power demand of electric vehicles in intervals.

5. The method for evaluating the charging and discharging capacity of multi-mode electric vehicles in an active distribution network according to claim 1, characterized in that, In S2, with maximizing the load-bearing capacity of electric vehicles as the objective function, steady-state constraints are established for the voltage and current of each phase, taking into account the characteristics of a three-phase unbalanced active distribution network. Specifically: The objective function is to maximize the load-bearing capacity of the electric vehicle. In the formula, For the total load capacity of the electric vehicle system, for Time Node Upper limit of electric vehicle capacity in China for Time Node The number of electric vehicles that can be connected. For electric vehicle access nodes, For a three-phase unbalanced active distribution network, steady-state operating constraints are formulated for each phase voltage and current using the right-angle components of the voltage and current phasors: (1) Real part current balance constraint: In the formula, For nodes To the node The line in phase ,time The real part of the current below; For nodes To the node The line in phase ,time The real part of the current below; Electric vehicle injection nodes participating in V2G mode In phase ,time The real part of the current below; Injecting nodes into substations In phase ,time The real part of the current below; Distributed generator node In phase ,time The real part of the current below; Injecting nodes into switched capacitor banks In phase ,time The real part of the current below; For energy storage devices from nodes In phase ,time The real part of the current absorbed below; , Injecting nodes into orderly and disorderly charging modes for electric vehicles In phase ,time The real part of the current below; For nodes In phase ,time The real part of the load current; For demand response at nodes In phase ,time The real part of the current absorbed below; For the set of routes; For nodes time The number of electric vehicles (2) Imaginary part current balance constraint: In the formula, in the formula, For nodes To the node The line in phase ,time The imaginary part of the current below; For nodes To the node The line in phase ,time The imaginary part of the current below; Electric vehicle injection nodes participating in V2G mode In phase ,time The imaginary part of the current below; Injecting nodes into substations In phase ,time The imaginary part of the current below; Distributed generator node In phase ,time The imaginary part of the current below; Injecting nodes into switched capacitor banks In phase ,time The imaginary part of the current below; For energy storage devices from nodes In phase ,time The imaginary part of the current absorbed below; , Injecting nodes into orderly and disorderly charging modes for electric vehicles In phase ,time The imaginary part of the current below; For nodes In phase ,time The imaginary part of the load current; For demand response at nodes In phase ,time The imaginary part of the current absorbed below; For the set of routes, (3) Voltage real and imaginary part drop constraints: In the formula, , For nodes In phase ,time The real and imaginary parts of the voltage. , For nodes In phase ,time The imaginary part of the voltage is below. For the line In phase and The resistance between them For the line In phase and The reactance between them For the line In phase ,time The real part of the current below, For the line In phase ,time The real part of the current below, For phase set, (4) Nonlinear load current constraint: In the formula, , They are nodes In phase ,time The real and imaginary parts of the load current. , They are nodes In phase ,time The active and reactive power demand is as follows. , They are nodes In phase ,time The real and imaginary parts of the voltage. (5) Voltage amplitude constraint: In the formula, , These are the lower and upper voltage limits, respectively. (6) Line capacity constraints: In the formula, , The lines are respectively In phase ,time The real and imaginary parts of the current. For the line Maximum current limit.

6. The method for evaluating the charging and discharging capacity of multi-mode electric vehicles in an active distribution network according to claim 1, characterized in that, In S2, considering the operational constraints of three modes of electric vehicles simultaneously, a multimodal electric vehicle load-bearing capacity assessment model is constructed under the scenario of a three-phase unbalanced active power distribution system. Specifically: Electric vehicle operation constraint modeling: (1) Active and reactive power constraints of electric vehicles: (2) Constraints on disordered charging of electric vehicles: (3) Constraints on orderly charging of electric vehicles: (4) Constraints on electric vehicles participating in V2G mode: Distributed power source constraint modeling: (1) Active power and reactive power constraints: In the formula, , For distributed power at nodes ,time The active and reactive power of the following , For injection nodes In phase ,time The real and imaginary parts of the current. (2) Active power and reactive power limits: In the formula, For distributed power at nodes The upper limit of active power, For distributed power at nodes The power factor limit, The constraint modeling of the switched capacitor bank is as follows: (1) Constraints on the real and imaginary parts of the switching capacitor bank current: In the formula, , The switched capacitor bank at the node phase ,time The real and imaginary parts of the injected current. node The number of switched capacitor banks, , The first Group of switched capacitor banks at nodes phase ,time The real and imaginary parts of the injected current. (2) Current constraints provided by each phase and each switched capacitor bank module: In the formula, node The switching capacitor module susceptance, Modeling the operational constraints of energy storage devices: (1) Net active power injection constraint: In the formula, , For energy storage devices at nodes ,time The charging and discharging power at the following levels , For energy storage devices at nodes phase ,time The real and imaginary parts of the absorption current. (2) Reactive power constraint: (3) Charge state update constraints: In the formula, Energy storage devices at nodes ,time The state of charge under these conditions, For energy storage devices at nodes The initial state of charge, For energy storage devices at nodes ,time The charging power at the following levels For energy storage devices at nodes Charging efficiency, For energy storage devices at nodes ,time The discharge power at that time, For energy storage devices at nodes The discharge efficiency, , These are energy storage devices at nodes The minimum and maximum states of charge, (4) Charging and discharging power limits: In the formula, , For nodes Minimum and maximum charging power , For nodes Minimum and maximum discharge power Time-sharing demand response modeling: (1) Constraints on changes in active and reactive power: In the formula, For the demand response procedure at the node Phase ,time The increase in active power, For the demand response procedure at the node Phase ,time Reduced active power, For the demand response procedure at the node Phase ,time The reactive power below, , For the demand response procedure at the node Phase ,time The real and imaginary parts of the current. (2) Demand response procedure power variation constraint: In the formula, The maximum percentage of load change allowed for the node. Binary variable: Whether to allow increased load on the demand response node. Binary variables: Are load reductions allowed at the demand response node? (3) Reactive power constraints in demand response: In the formula, The power constraint angle for demand response.

7. The method for evaluating the charging and discharging capacity of multi-mode electric vehicles in an active distribution network according to claim 1, characterized in that, In S3, nonlinear constraints such as voltage and current balance and electric vehicle operating modes are linearized, specifically as follows: After linearizing constraint (24), we get: In the formula, , , , functions respectively and Partial derivatives with respect to the real and imaginary parts of the voltage, , These are the estimated values ​​for the real and imaginary parts of the voltage, respectively. After linearizing the voltage in constraint (25), we get: In the formula, Phase A is the reference phase angle for each phase. Phase B Phase C , For the maximum positive phase angle deviation, For the maximum negative phase angle deviation, This is the per-unit value of the lower limit of voltage amplitude. In the formula, Phase A is the reference phase angle for each phase. Phase B Phase C , For the maximum positive phase angle deviation, This is the per-unit value of the lower limit of voltage amplitude. In the formula, Phase A is the reference phase angle for each phase. Phase B Phase C , For the maximum negative phase angle deviation, These are the per-unit values ​​of the upper limit of voltage amplitude. In the formula, Phase A is the reference phase angle for each phase. Phase B Phase C , For the maximum positive phase angle deviation, In the formula, For each phase reference phase angle Phase B Phase C , This represents the maximum negative phase angle deviation.

8. The method for evaluating the charging and discharging capacity of multi-mode electric vehicles in an active distribution network according to claim 1, characterized in that, In S3, a linear model that can be solved efficiently is obtained, and the branch-and-bound method or the cutting plane method is used to solve the model, thereby realizing the configuration optimization of charging and discharging of multimodal electric vehicles. Specifically: Line capacity nonlinearity constraint in constraint (44) By decomposing variables and summing pieces, the problem is transformed into a linear constraint. The piecewise linear fitting expression for the quadratic term is: In the formula, This is a linear approximation of the second term of the current. Due to maximum heat capacity limitations, For the first The segmentation coefficient of the segment. , The lines are respectively No. At time t, the first The real and imaginary parts of a segment are segment variables. This represents the total number of segments in the piecewise linear fit. Through the linearization process described above, the complete model is transformed into a mixed-integer linear programming model. Consider the model of electric vehicles participating in three charging modes: Maximize the load-bearing capacity of electric vehicles when they only participate in disordered charging. Model: When electric vehicles participate in disorderly or orderly charging, the carrying capacity of electric vehicles should be maximized. Model: The transformed mixed-integer linear programming model is solved using either the branch and bound method or the cutting plane method, ultimately yielding the total carrying capacity of electric vehicles in the target distribution network. Construct a quantitative index model for the load-bearing capacity of electric vehicles, and the load-bearing capacity improvement rate of the ordered mode relative to the disordered mode: The capacity improvement rate of V2G mode compared to ordered mode: 。 9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the active distribution network multimodal electric vehicle charging and discharging capacity assessment method as described in any one of claims 1 to 8.

10. A computer-readable storage medium storing computer instructions thereon, characterized in that, When executed by the processor, the computer instructions implement the active distribution network multi-mode electric vehicle charging and discharging capacity assessment method as described in any one of claims 1-8.

Images