Microgrid emergency power supply capability assessment method based on aggregation operation domain

Abstract

The invention discloses a micro-grid emergency power supply capability assessment method based on an aggregation operation domain. The method comprises the following steps: firstly, establishing an aggregation operation domain general model based on Enry hub; according to the decomposable characteristic of the model, constructing an aggregation operation domain model based on partial decomposition of time domain coupling and time domain decoupling; for the time domain decoupling part, obtaining upper and lower power boundaries of multi-cycle decoupling by solving a linear optimization problem; for the time domain coupling part, based on a homomorphic polyhedron aggregation operation domain solving algorithm, the time domain coupling part model with the energy storage property is approximated by optimizing a basic homomorphic polyhedron, a scaling factor and a translation factor at the same time; and finally, constructing a microgrid emergency power supply capability evaluation system, solving the model and calculating evaluation indexes to obtain a final evaluation result. The method provided by the invention proves that the aggregation operation domain can be equivalently modeled as a time domain decoupling and time domain coupling partial decomposition model, and the method can be effectively applied to various scenes such as power distribution network fault recovery and the like.

Description

Technical Field

[0001] The present invention belongs to the technical field of integrated energy system operation, and mainly relates to a method for evaluating the emergency power supply capacity of a microgrid based on an aggregated operation domain. Background Art

[0002] With the rapid development of distributed resources and distribution network automation technologies, the evaluation of the emergency power consumption capacity of microgrids has gradually become a research hotspot for distribution network fault recovery and network reconstruction. Through the distributed resource power aggregation operation domain model, the emergency power supply capacity of a microgrid can be effectively evaluated. However, considering the multi-period coupling characteristics of distributed resource clusters, establishing an efficient and accurate aggregation operation domain model remains a very challenging research difficulty.

[0003] Distributed multi-energy flexible resources, usually including distributed generation, energy storage systems, and multi-energy flexible loads, have great potential to provide operation flexibility for power systems. Existing inner approximation aggregation models usually sacrifice a large amount of feasible space due to dependence on preset geometric models, while outer approximation methods cannot guarantee the decomposition feasibility of the model. Therefore, how to establish a distributed resource power aggregation operation domain model and construct a microgrid emergency power supply capacity evaluation system to effectively evaluate the dynamic emergency power supply capacity of a microgrid, and then quickly isolate the fault area, optimize the network structure, and reduce power losses, has become a key issue for improving the emergency power supply guarantee capacity after distribution network faults and is an important basis for the flexible coordinated planning of power systems. Summary of the Invention

[0004] In view of the problems of high model complexity and poor analytical properties existing in traditional aggregation modeling techniques, as well as the problems of low solution efficiency and high conservatism existing in existing parameter estimation methods, the present invention proposes a method for evaluating the emergency power supply capacity of a microgrid based on an aggregated operation domain. First, a general model of the aggregated operation domain based on Energy hub is established; according to the model decomposability characteristics, an aggregated operation domain model based on time-domain coupling and time-domain decoupling partial decomposition is constructed; for the time-domain decoupling part, the upper and lower power boundaries of multi-period decoupling are obtained by solving a linear optimization problem; for the time-domain coupling part, an algorithm for solving the aggregated operation domain based on a homomorphic polyhedron is proposed, and by simultaneously optimizing the basic homomorphic polyhedron, the scaling factor, and the translation factor, the time-domain coupling part model with energy storage properties is approximated; finally, a microgrid emergency power supply capacity evaluation system is constructed, the model is solved, and the evaluation index is calculated to obtain the final evaluation result. The method of the present invention proves that the aggregated operation domain can be equivalently modeled as a model of time-domain decoupling and time-domain coupling partial decomposition. Compared with the existing inner approximation model, this method has clear physical meaning and lower conservatism, and can be effectively applied to many scenarios such as distribution network fault recovery.

[0005] To achieve the above object, the technical solution adopted by the present invention is as follows: A method for evaluating the emergency power supply capacity of a microgrid based on an aggregated operation domain, comprising the following steps:

[0006] S1. Establish a general model of the aggregated operation domain based on the Energy hub: The model at least includes an energy balance constraint, a device capacity constraint, and a storage SOC constraint, and the mathematical form of the model is specifically as follows:

[0007]

[0008] Among them, represents the port variable, the time-domain decoupled variable represents the input / output power of non-storage devices, the charge / discharge power variable represents the power flow into / out of the energy storage, the SOC change variable characterizes the energy change of the energy storage device, AFR represents its operation domain under operation constraint conditions, A represents the equality constraint coefficient matrix, D represents the inequality constraint coefficient matrix, f represents the inequality constraint coefficient, G represents the energy storage constraint coefficient matrix, h represents the energy storage constraint coefficient, t is the index of the time interval, v is the index of the time interval t, T is the time index set, N bus represents the number of buses, N b ,N in and N str respectively represent each variable and the dimension of, N ∑ = N b + N in + 3N str ,

[0009] The coefficient matrix A is specifically as follows:

[0010]

[0011] Among them,

[0012]

[0013] The coefficient matrices D and f are specifically as follows:

[0014]

[0015] Among them,

[0016] The coefficient matrices G and h are specifically as follows:

[0017]

[0018] in,

[0019] S2. Constructing an aggregate operation domain model: Constructing an aggregate operation domain model based on partial decomposition of time domain coupling and time domain decoupling. The mathematical form of the model is as follows:

[0020] DAFR: in, represents the augmented bound variable, represents the time domain decoupling part variable, Represents the time domain coupling part variable, Ω in and Ω es They represent the two parts of the feasible domain respectively, and DAFR represents its operating domain under the operating constraints;

[0021] S3. Aggregate Operation Domain Solution: For the time-domain decoupling part in step S2, the upper and lower power bounds of multi-cycle decoupling are obtained by solving a linear optimization problem. For the time-domain coupling part, a homomorphic polyhedron-based aggregate operation domain solution algorithm is proposed. By simultaneously optimizing the basic homomorphic polyhedron, scaling factor, and translation factor, the time-domain coupling part model with energy storage properties is approximated.

[0022] S4. Constructing a microgrid emergency power supply capability assessment system: Using the aggregated operation domain model constructed in step S2, confirming assessment indicators, and assessing the system emergency power supply capability based on the assessment indicators; the assessment indicators include but are not limited to ramp rate, ramp duration, and power adjustable range;

[0023] S5. Use the aggregated operation domain solving algorithm proposed in step S3 to solve the decomposition aggregation model obtained in step S2, calculate the microgrid emergency power supply capability evaluation index constructed in step S4, and summarize the final evaluation results.

[0024] As an improvement of the present invention, the energy balance constraint of the general model of the aggregation operation domain in step S1 is specifically:

[0025]

[0026] in, Represents the charging and discharging efficiency of energy storage,

[0027] The equipment capacity constraints are specifically:

[0028]

[0029] in, as well as They are and The upper and lower bounds of Represent the upper and lower bounds of internal variables (time-domain decoupled variables / non-energy storage variables);

[0030] Define The specific energy storage SOC constraint is as follows:

[0031]

[0032] Among them, E i , Represents the upper and lower bounds of the energy storage capacity and the initial energy storage capacity of the i-th energy storage, and σ i Represents the self-discharge coefficient, (β T×1 ) v = σ v , 1 ≤ v ≤ T.

[0033] As an improvement of the present invention, in the step S2, an augmented boundary variable including a virtual port is defined Define Simplify the inequality constraint and the equality constraint respectively. According to the decomposable characteristics of the model, define The feasible region of is

[0034]

[0035] Define The feasible region of is X es :

[0036]

[0037] Among them

[0038] Construct an aggregated operation domain model based on time-domain coupling and time-domain decoupled partial decomposition, where the feasible region of the time-domain decoupled part is defined as Ω in , and the feasible region of the time-domain coupling part is defined as Ω es :

[0039]

[0040] As another improvement of the present invention, in the time-domain decoupled part of the step S3, the upper and lower power boundaries of multi-period decoupling are obtained by solving a linear optimization problem:

[0041]

[0042] The feasible region of the time-domain decoupled part is equivalent to: Among them, Is the upper and lower bounds of the feasible region of the time-domain decoupled part.

[0043] As another improvement of the present invention, in the time-domain coupling part of step S3, an aggregated energy storage variable is defined

[0044]

[0045] Time-domain coupling part model It is expressed as:

[0046]

[0047] Wherein, is an approximation of the aggregated energy storage variable,

[0048] Combined with the time-domain decoupling part model Ω in And the time-domain coupling part model The total aggregated feasible region is expressed as:

[0049]

[0050] Wherein,

[0051] As yet another improvement of the present invention, the algorithm for solving the aggregated operation region based on the homomorphic polyhedron in step S3 is specifically as follows:

[0052] S31: Establish a feasible region model for the i-th energy storage individual:

[0053]

[0054] Wherein, respectively represent the charge and discharge power of the i-th energy storage, x chr,bt,i , x dis,bt,i , respectively represent the upper and lower bounds of the charge and discharge power, E bt,i , represents the initial capacity of the energy storage and the upper and lower bounds of the energy storage, σ bt,i ,η chr,bt,i ,η dis,bt,i represent the self-discharge coefficient and the charge and discharge efficiency;

[0055] Define the model parameters M bt,i and N bt,i and write them in a compact form:

[0056] Γ es,bt,i ={x es,bt,i |M bt,i x es,bt,i ≤Nbt,i}

[0057] Wherein:

[0058]

[0059]

[0060]

[0061]

[0062]

[0063] S32: Aggregate the exact model of the feasible region Γ es,bt,ag From Γ es,bt,i , 1 ≤ i ≤ N str,bt The Minkowski sum of gives:

[0064]

[0065] S33: Define the basic homomorphic polyhedron Γ es,bt,0 :

[0066] Γ es,bt,0 = {x es,bt,0 | M bt,0 x es,bt,0 ≤ N bt,0}

[0067] Wherein, is the coefficient matrix of the basic homomorphic polyhedron to be optimized;

[0068] S34: Define the scaling factor of the i-th energy storage individual as The translation factor is By scaling and translating the basic homomorphic polyhedron Γ es,bt,0 , the inner approximation polyhedron feasible region of the i-th energy storage unit is obtained

[0069]

[0070] S35: Represent the aggregate feasible region approximation model with the basic homomorphic polyhedron Γ es,bt,0

[0071]

[0072] Define the scaling factor β of the aggregate model bt,ag And the translation factor t bt,ag As:

[0073]

[0074] Aggregate feasible region approximation model Expressed as:

[0075]

[0076] in, Indicates the aggregated energy storage variable on the busbar. The subscript bt is omitted in the following steps.

[0077] S36: Establish an optimization model to solve the basic homomorphic polyhedron, scaling factor and translation factor:

[0078]

[0079] Among them, ρ is a penalty coefficient, s i =1 / β i ,r i =-t i / β i , G i is the variable to be optimized, is the coefficient matrix of the basic homomorphic polyhedron to be optimized, It is the empirical parameter used in the past for basic homomorphic polyhedrons. It is used as the benchmark value of M0 and N0 in this model for further optimization. str Represents the number of energy storage units, represents the square of the Frobenius norm;

[0080] According to Farkas' theorem, the optimal scaling factor β can be obtained from the optimal solution of the optimization problem. bt,i* =1 / s bt,i* and the translation factor t bt,i* =-r bt,i* / s bt,i* , and the basic homomorphic polyhedral system matrix

[0081] As a further improvement of the present invention, the bilinear optimization problem of the optimization model in step S36 is decomposed into two linear sub-problems for iterative solution:

[0082] Sub-question 1:

[0083]

[0084] Sub-question 2:

[0085]

[0086] when and The iteration is terminated when ∈ is 10 -3Magnitude.

[0087] As a further improvement of the present invention, the ramp rate in step S4 is specifically:

[0088]

[0089] Wherein, R u (T u ), R d (T d ) respectively represent the ramp rates of the upward ramp time T u and the downward ramp time T d . P0 and P1 respectively represent the aggregated power at the initial moment and the next moment, and FR(P1) represents the feasible region of the aggregated power at the next moment;

[0090] The specific ramp duration is:

[0091] T u (R u ) = maxk s.t. P0 + kR u ∈ FR(P1)

[0092] T d (R d ) = maxk s.t. P0 - kR d ∈ FR(P1)

[0093] Wherein, T u , T d respectively represent the upward ramp duration and the downward ramp duration, P0 represents the aggregated power at the initial moment, and FR(P1) represents the feasible region of the aggregated power at the next moment;

[0094] The specific power adjustable range is:

[0095]

[0096] Wherein, respectively represent the upper and lower limits of the aggregated power, and FR represents the feasible region of the aggregated power.

[0097] Compared with the prior art, the beneficial effects of the present invention are:

[0098] (1) The method of the present invention establishes a general model of the aggregated operation domain based on the Energy hub, ensuring the generality and rationality of the present invention;

[0099] (( (2) In step S2 of the method of the present invention, it is proved that the operation domain model can be equivalently decomposed into two parts: time-domain decoupling and time-domain coupling partial decomposition, further clarifying the physical meaning of the aggregated operation domain model;

[0100] (3) The aggregation operation domain solution algorithm based on homomorphic polyhedron proposed by the method of the present invention for the time-domain coupling part optimizes the basic homomorphic polyhedron, scaling factor, and translation factor at the same time, improves the drawback of strong conservatism of the basic homomorphic body using the experience of the previous homomorphic method, and the iterative solution model based on coordinate descent proposed by the algorithm transforms the bilinear problem into a linear optimization model that is easy to solve. By solving the time-domain coupling part model established above through this algorithm, the NP-hard problem brought by directly solving the Minkowski sum of individual operation domains is avoided, and the solution efficiency is greatly improved while ensuring the solution accuracy;

[0101] (4) The emergency power supply capacity evaluation system constructed by the method of the present invention can improve the decision-making efficiency during the fault recovery process of the distribution network and provides strong technical support for the flexible coordinated planning of the power system.

[0102] (5) The method of the present invention proves that the aggregation operation domain can be equivalently modeled as a model of time-domain decoupling and time-domain coupling part decomposition. Compared with the existing inner approximation model, this method has clear physical meaning and lower model complexity. The proposed aggregation operation domain solution algorithm based on homomorphic polyhedron optimizes the basic homomorphic polyhedron for the first time, further reduces the conservatism of the approximation method, and can be effectively applied to many scenarios such as distribution network fault recovery. Description of the Drawings

[0103] Figure 1 is the schematic structural diagram of the general model of flexible resources of the integrated energy system based on Energy hub of the present invention;

[0104] Figure 2 is the step flow chart of the microgrid emergency power supply capacity evaluation method based on the aggregation operation domain of the present invention;

[0105] Figure 3 is the relationship diagram of the ramp rate and the continuous ramp time in Embodiment 1 of the present invention;

[0106] Figure 4 is the schematic diagram of the power adjustable range in Embodiment 1 of the present invention;

[0107] Figure 5 is the schematic structural diagram of the distribution system in the test example of the present invention;

[0108] Figure 6 is the schematic diagram of the park node scheduling power curve under different methods in the test example of the present invention;

[0109] Figure 7 is the schematic diagram of the cost growth curve under different energy storage ratios under different methods in the test example of the present invention. Detailed Embodiments

[0110] The present invention will be further illustrated below in conjunction with the accompanying drawings and specific embodiments. It should be understood that the following specific embodiments are only used to illustrate the present invention and not to limit the scope of the present invention.

[0111] Embodiment 1

[0112] A method for evaluating the emergency power supply capacity of a microgrid based on an aggregated operation domain is applied to an integrated energy system, and its structure is as Figure 1 shown. The integrated energy system includes three energy forms: electricity, heat, and natural gas. Different energies are coupled and energy-converted through P2G devices, P2H devices, etc. The flow of this method is as Figure 2 shown, and it includes the following steps:

[0113] Step S1: Establish a general model of the aggregated operation domain based on Energy hub;

[0114] Energy hub consists of an input port and an output port. The input port includes electricity and natural gas, while the output port includes different energy loads. This model allows various resources to be modeled within a unified framework. Energyhub includes electrical, gas, heat, and cold buses, and the conversion units in Energy hub include typical flexible resources such as distributed generation, energy storage, and other energy converters. Therefore, first define a general model of flexible resources based on Energy hub.

[0115] The variables of the Energy hub model include input power, non-storage device power, load power, charging power, discharging power, change in storage SOC, etc. In this embodiment, the input variable represents the input power of Energy hub (such as electricity, natural gas), where t represents the index of the time interval and T represents the set of time indices. The non-storage variable represents the input and output power of non-storage devices except for renewable energy generation. The renewable energy variable represents the power of renewable energy generation. The load power variable represents the electrical, heat, and cold loads of the output port. The charging / discharging power variable represents the power flow in and out of the energy storage system. The SOC change variable represents the change in the energy in storage. N b ,N f ,N res ,N load and N str respectively represent and the dimensions of.

[0116] Establish the mathematical form of the general model, including at least energy balance constraints, equipment capacity constraints, and energy storage SOC constraints:

[0117] Establish energy balance constraints and define all non-energy storage variables as internal variables

[0118]

[0119] Among them, represents the charge and discharge efficiency of the energy storage.

[0120] Establish equipment capacity constraints:

[0121]

[0122] Among them, and are respectively and the upper and lower bounds of, represents the upper and lower bounds of the internal variables.

[0123] Establish energy storage SOC constraints:

[0124]

[0125] Among them, E i , represents the upper and lower bounds of the energy storage capacity and the initial energy storage capacity, σ i represents the self-discharge coefficient, (β T×1 ) v =σ v , 1 ≤ v ≤ T.

[0126] Give the compact form of the mathematical form of the general model. First, define the variables of the entire system as:

[0127]

[0128] Among them, N Σ =N b +N in +3N str .

[0129] Then establish the compact mathematical form of the general model:

[0130]

[0131] Among them,

[0132] Step S2: constructing an aggregated operation domain model based on partial decomposition of time domain coupling and time domain decoupling according to the decomposable characteristics of the model;

[0133] S21. Define the augmented boundary variable containing the virtual port as definition

[0134] S22, simplify the expression;

[0135] S221. Simplify inequality constraints:

[0136]

[0137]

[0138] S222, the equality constraint is simplified to:

[0139]

[0140] in,

[0141] S23. Based on the decomposable characteristics of the model, an aggregated operation domain model is constructed based on partial decomposition of time domain coupling and time domain decoupling;

[0142] S231. Definition The feasible domain is

[0143]

[0144] S232, Definition The feasible region is X es :

[0145]

[0146] in,

[0147] By eliminating variables It can be further simplified to:

[0148]

[0149] S233. Define augmented boundary variables The feasible domain of:

[0150]

[0151] S234, Definition Defining variables The T-dimensional feasible region is Ω in, further simplify:

[0152]

[0153] S235. Define DAFR as the augmented boundary variable of the feasible region:

[0154] DAFR:

[0155] Step S3. For the time-domain decoupled part, the algorithm obtains the upper and lower power boundaries of multi-period decoupling by solving a linear optimization problem to describe power flexibility; for the time-domain coupled part, an aggregation operation domain solution algorithm based on the homomorphic polyhedron is proposed. By simultaneously optimizing the basic homomorphic polyhedron, the scaling factor, and the translation factor, the time-domain coupled part model with energy storage properties is approximated to reduce the conservatism of the approximation algorithm.

[0156] S31. Aggregation operation domain solution algorithm for the time-domain decoupled part:

[0157] Obtain the upper and lower power boundaries of multi-period decoupling by solving a linear optimization problem

[0158]

[0159] In this way, the feasible region of the time-domain decoupled part is equivalent to:

[0160]

[0161] S32. For the time-domain coupled part, an aggregation operation domain solution algorithm based on the homomorphic polyhedron is proposed. By simultaneously optimizing the basic homomorphic polyhedron, the scaling factor, and the translation factor, the time-domain coupled part model with energy storage properties is approximated;

[0162] S321. Establish the feasible region model of the i-th energy storage unit:

[0163]

[0164] Define the model parameters M bt,i and N bt,i , and write it in a compact form:

[0165] Γ es,bt,i ={x es,bt,i |M bt,i x es,bt,i ≤N bt,i}

[0166] where:

[0167]

[0168]

[0169] S322. Aggregation Feasible Region Exact Model Γ es,bt,ag From Γ es,bt,i , 1 ≤ i ≤ N str,bt The Minkowski sum is obtained as follows:

[0170]

[0171] S323. Define the basic homomorphic polyhedron Γ es,bt,0 :

[0172] Γ es,bt,0 = {x es,bt,0 | M bt,0 x es,bt,0 ≤ N bt,0}

[0173] where is the coefficient matrix of the basic homomorphic polyhedron to be optimized.

[0174] S324. Define the scaling factor of the i-th energy storage individual as and the translation factor as By scaling and translating the basic homomorphic polyhedron Γ es,bt,0 , the inner approximation polyhedron feasible region of the i-th energy storage unit is obtained

[0175]

[0176] S325. Use the basic homomorphic polyhedron Γ es,bt,0 to represent the aggregation feasible region approximation model

[0177]

[0178] Furthermore, define the scaling factor β bt,ag and the translation factor t bt,ag of the aggregation model as:

[0179]

[0180] Therefore, the aggregation feasible region approximation model can be further expressed as:

[0181]

[0182] where represents the aggregated energy storage variable on this bus.

[0183] S326. Establish an optimization model for solving the basic homomorphic polyhedron, scaling factor, and translation factor:

[0184]

[0185] Among them, ρ is a penalty coefficient.

[0186] From the optimal solution of this optimization problem, the optimal scaling factor β bt,i* = 1 / s bt,i* and the translation factor t bt,i* = -r bt,i* / s bt,i* , and the basic homomorphic polyhedron system coefficient matrix

[0187] Since the optimization model is a bilinear optimization problem, it cannot be directly solved by an off-the-shelf solver. The present invention uses the coordinate descent idea to decompose the above bilinear optimization problem into two linear sub-problems for iterative solution:

[0188] Sub-problem 1:

[0189]

[0190] Sub-problem 2:

[0191]

[0192] When and , terminate the iteration, where ∈ is a small value.

[0193] S33. Obtain an approximate aggregated feasible region model;

[0194] Define the aggregated energy storage variable

[0195]

[0196] Through the above solution algorithm, the time-domain coupling part model can be approximately expressed as:

[0197]

[0198] Among them, is the approximation of the aggregated energy storage variable,

[0199] Combined with the time-domain decoupled part model Ω in and the time-domain coupling part model The total aggregated feasible region can be expressed as:

[0200]

[0201] Among them,

[0202] Step S4: Construct a microgrid emergency power supply capacity evaluation system. By constructing an aggregated operation domain model, evaluation indicators such as the ramping rate, ramping duration, and power adjustment range are obtained, so as to evaluate the emergency power supply capacity of the system.

[0203] Establish the ramping rate index:

[0204]

[0205] Among them, R u (T u ), R d (T d ) represent the upward ramping rate and the downward ramping rate respectively, and are functions of the upward ramping time T u and the downward ramping time T d respectively. P0 and P1 represent the aggregated power at the initial moment and the next moment respectively, and FR(P1) represents the feasible region of the aggregated power at the next moment.

[0206] Establish the ramping duration index:

[0207] T u (R u ) = maxk s.t. P0 + kR u ∈ FR(P1)

[0208] T d (R d ) = maxk s.t. P0 - kR d ∈ FR(P1)

[0209] Among them, T u and T d represent the upward ramping duration and the downward ramping duration respectively, P0 represents the aggregated power at the initial moment, and FR(P1) represents the feasible region of the aggregated power at the next moment.

[0210] Establish the power adjustment range index:

[0211]

[0212] Among them, represent the upper and lower limits of the aggregated power respectively, and FR represents the feasible region of the aggregated power.

[0213] Step S5: Solve the model and calculate the evaluation indicators.

[0214] S51: Based on MATLAB, use Gurobi to solve the aggregated operation domain model established in Step S3;

[0215] S52. Calculate various evaluation indicators based on the emergency power supply capacity evaluation system constructed in step S4. Conduct experiments on the relationship between the ramp rate and the continuous ramp time and the power adjustable range as follows:

[0216] Give the system an initial state. Take the feasible region of the port electric power of the system at the next moment as the research object, and perform optimization and solution through the above model. Evaluate the results of the aggregation methods introduced above respectively, calculate the three evaluation indicators of the ramp rate, ramp duration, and power adjustable range proposed above respectively, and make comparisons. Since the cross-section indicators calculated by the robust method are consistent with the accurate results, only the differences between the evaluation indicators under the three models of the geometric model (PA), the analytical model (AN) proposed by the present invention, and the accurate model (EX) are compared.

[0217] Figure 3 The upward ramp rate and ramp time curves and the downward ramp rate and ramp time curves of the three methods are plotted respectively. The experimental results show that the index performance under the analytical model is very close to the accurate model, with a larger upward and downward ramp rate compared to the geometric model under the same ramp time, and a shorter ramp time required under the same ramp rate, reflecting that the analytical model has better emergency power supply capacity.

[0218] To more intuitively compare the power adjustable range indicators on the time cross-section under each model, successively take the intermediate state of the system at the previous moment to calculate the feasible region of the electric power at the next moment, and plot as Figure 4 the box plot of the cross-section feasible region shown. Figure 4 It can be found from that the power adjustable range under the analytical model is basically the same as that of the accurate model, while the geometric model loses more power adjustable range compared to the accurate model.

[0219] The above results show that the analytical method has obvious advantages over the geometric method in terms of ramp rate and power adjustable range, and its index data is similar to that of the accurate model.

[0220] Test example

[0221] The system applicable to this test example is a 33-node distribution system, as Figure 5 shown. The distribution network has 4 photovoltaic power generations and 4 Energy Hubs. The Energy Hub includes an electric boiler, a combined heat and power unit, etc. The optimization period is 24h, and the scheduling time interval is taken as 1h.

[0222] Calculate the feasible region of the electric power at the ports of the four parks according to the steps of the present invention, and provide it to the distribution network for economic scheduling. To evaluate the economy and accuracy of the proposed method, in this example, the scheduling results of all prediction methods are compared with the accurate scheduling result M0, where M3 represents the method of the present invention. The total power scheduling results of the 4 Energy Hubs are asFigure 6 as shown Figure 6 It can be seen from the figure that the total power dispatch result obtained by M3 is the closest to that obtained by M0, verifying that the proposed method has the least conservatism.

[0223] To analyze the performance of each method under different energy storage ratios, this example tests the scenario of increasing the energy storage ratio. Compared with M0, the increased costs at different energy storage ratios are as Figure 7 shown. It can be concluded that as the energy storage capacity increases, the increased cost of M1 gradually increases, while the increased costs of M2 and M3 do not show an obvious growth trend and are always less than 15%, confirming the low conservatism of these two methods. In most cases, the increased cost of M3 is the smallest, which verifies the superiority of the method proposed in the present invention. It can be seen from the economic dispatch problem in this example that the model proposed in the present invention has good performance in many scenarios. This method can make good use of the flexibility of distributed resources and provides an aggregation model with good performance and less computational effort for the optimal operation of the distribution network.

[0224] In summary, the method of the present invention establishes a general model of the aggregated operation domain based on the Energy hub, constructs an aggregated operation domain model with time-domain coupling and time-domain decoupling decomposition based on the general model; then respectively proposes solution algorithms for the time-domain decoupling and time-domain coupling parts, and finally constructs a microgrid emergency power supply capacity evaluation system based on evaluation indexes such as ramp rate, ramp duration, and power adjustable range to complete the evaluation of the emergency power supply capacity. The method of the present invention greatly improves the solution efficiency while ensuring the solution accuracy. The constructed emergency power supply capacity evaluation system can improve the decision-making efficiency of the distribution network during the fault recovery process and provides strong technical support for the flexible and coordinated planning of the power system.

[0225] It should be noted that the above content only illustrates the technical idea of the present invention and cannot be used to limit the protection scope of the present invention. For those of ordinary skill in the art in this technical field, without departing from the principle of the present invention, several improvements and refinements can be made, and these improvements and refinements all fall within the protection scope of the claims of the present invention.

Claims

1. A method for evaluating the emergency power supply capacity of a microgrid based on an aggregated operation domain, characterized in that, It includes the following steps: S1. Establish a general model for the aggregated operation domain based on the Energy hub: The model at least includes energy balance constraints, device capacity constraints, and energy storage SOC constraints. The specific mathematical form of the model is as follows: AFR: Among them, x b represents the port variable, and its variable at the t-th moment is time-domain decoupled variable represents the input / output power of non-energy storage devices, charge / discharge power variable represents the power flow into / out of the energy storage, SOC change variable characterizes the energy change of the energy storage device. AFR represents its operating domain under operating constraints. A represents the equality constraint coefficient matrix, D represents the inequality constraint coefficient matrix, f represents the inequality constraint coefficient, G represents the energy storage constraint coefficient matrix, h represents the energy storage constraint coefficient, t is the index of the time interval, v is the index of the time interval t, T is the time index set, N b ,N in and N str respectively represent each variable and the dimension of, N ∑ = N b + N in + 3N str ; S2. Construct an aggregated operation domain model: Construct an aggregated operation domain model based on time-domain coupling and time-domain decoupling partial decomposition. The mathematical form of the model is as follows: DAFR: Among them, represents the augmented boundary variable, represents the time-domain decoupled part variable, represents the time-domain coupled part variable, Ω in and Ω es respectively represent the feasible regions of these two parts, and DAFR represents its operating region under operating constraint conditions; S3. Solve the aggregated operation domain: For the time-domain decoupling part in step S2, obtain the upper and lower power boundaries of multi-period decoupling by solving a linear optimization problem; for the time-domain coupling part, based on the aggregated operation domain solution algorithm of the homomorphic polyhedron, approximate the time-domain coupling part model with energy storage properties by simultaneously optimizing the basic homomorphic polyhedron, scaling factor, and translation factor. S4. Construct a microgrid emergency power supply capacity evaluation system: Through the aggregated operation domain model constructed in step S2, confirm the evaluation indexes, and evaluate the system emergency power supply capacity according to the evaluation indexes; the evaluation indexes include but are not limited to ramp rate, ramp duration, and power adjustable range. S5. Use the aggregated operation domain solution algorithm proposed in step S3 to solve the decomposed aggregation model obtained in step S2, calculate the microgrid emergency power supply capacity evaluation indexes constructed in step S4, and summarize the final evaluation results.

2. The method for evaluating the emergency power supply capacity of a microgrid based on an aggregated operation domain according to claim 1, wherein: The energy balance constraint of the general model of the aggregated operation domain in step S1 is specifically: Among them, represents the charge and discharge efficiency of energy storage; The device capacity constraint is specifically: Among them, and are respectively and the upper and lower bounds of, representing the lower bound of the time-domain decoupled variable, representing the upper bound of the time-domain decoupled variable; Definition The energy storage SOC constraint is specifically as follows: Among them, E i , represent the upper and lower bounds of the energy storage capacity of the i-th energy storage and the initial energy storage capacity, and σ i represents the self-discharge coefficient, (β T×1 ) v = σ v , 1 ≤ v ≤ T.

3. The method for evaluating the emergency power supply capacity of a microgrid based on an aggregated operation domain according to claim 1, wherein: In the step S2, an augmented boundary variable including a virtual port is defined Define N bus represents the number of busbars, simplifies the inequality constraints and equality constraints respectively. According to the decomposable characteristics of the model, the feasible region of the simplified model expression is defined as Definition The feasible region of es is X Among them, Construct an aggregated operating region model based on time-domain coupling and time-domain decoupling partial decomposition, where the feasible region of the time-domain decoupling part is defined as Ω in , and the feasible region of the time-domain coupling part is defined as Ω es :

4. The method for evaluating the emergency power supply capacity of a microgrid based on an aggregated operation domain according to claim 2, wherein: For the time-domain decoupling part of step S3, the upper and lower power boundaries of multi-period decoupling are obtained by solving a linear optimization problem The feasible region of the time-domain decoupling part is equivalent to: Among them, are the upper and lower bounds of the time-domain decoupled partial feasible region.

5. The method for evaluating the emergency power supply capacity of a microgrid based on an aggregated operation domain according to claim 4, wherein: For the time-domain coupling part of step S3, an aggregated energy storage variable is defined Time-domain coupling partial model It is expressed as: Among them, is an approximation of the aggregated energy storage variable, Combined with the time-domain decoupled partial model Ω in and the time-domain coupled partial model The overall aggregated feasible region is expressed as: Among them, 6. The method for evaluating the emergency power supply capacity of a microgrid based on an aggregated operation domain according to claim 5, wherein: The aggregated operation domain solution algorithm based on the homomorphic polyhedron in step S3 is specifically: S31: Establish a feasible domain model for the i-th energy storage individual: Among them, respectively represent the charging and discharging power of the i-th energy storage, x chr,bt,i , x dis,bt,i , respectively represent the upper and lower bounds of the charging and discharging power, E bt,i , represents the initial capacity of the energy storage and the upper and lower bounds of the energy storage, σ bt,i , η chr,bt,i , η dis,bt,i represent the self-discharge coefficient and the charge-discharge efficiency; Define the model parameters M bt,i and N bt,i , written in a compact form: Γ es,bt,i ={x es,bt,i |M bt,i x es,bt,i ≤N bt,i} Where: (γ bt,i ) v = σ v , 1 ≤ v ≤ T, (Φ bt,i ) mn = σ m-n , S32: Aggregate Feasible Region Exact Model Γ es,bt,ag From the energy storage individual model Γ es,bt,i , 1 ≤ i ≤ N str,bt obtained by the Minkowski sum of: S33: Define the basic homomorphic polyhedron Γ es,bt,0 : Γ es,bt,0 = {x es,bt,0 | M bt,0 x es,bt,0 ≤ N bt,0} Among them, is the coefficient matrix of the basic homomorphic polyhedron to be optimized; S34: Define the scaling factor of the $i$-th energy storage individual as β bt,i ≥0, and the translation factor as By scaling and translating the basic homomorphic polyhedron Γ es,bt,0 , the inner approximation polyhedron feasible region of the $i$-th energy storage unit is obtained S35: Represent the approximate model of the aggregation feasible region using the basic homomorphic polyhedron Γ es,bt,0 and Define the scaling factor β of the aggregation model bt,ag and the translation factor t bt,ag as follows: Aggregate Feasible Region Approximation Model It is expressed as: Among them, represents the aggregated energy storage variable on this busbar; S36: Establish an optimization model for solving the basic homomorphic polyhedron, scaling factor, and translation factor: where ρ is a penalty coefficient, s i = 1 / β i , r i = -t i / β i , G i is the variable to be optimized, is the coefficient matrix of the basic homomorphic polyhedron to be optimized, is the empirical value, N str represents the number of energy storages, represents the square of the Frobenius norm.

7. The method for evaluating the emergency power supply capacity of a microgrid based on an aggregated operation domain according to claim 6, wherein: The bilinear optimization problem of the optimization model in step S36 is decomposed into two linear sub-problems for iterative solution: Sub-problem 1: Sub-problem 2: When and terminate the iteration, where ∈ takes the order of magnitude of 10 -3 order of magnitude.

8. The method for evaluating the emergency power supply capacity of a microgrid based on an aggregated operation domain according to claim 1, wherein: The ramp rate in step S4 is specifically: Among them, R u (T u ), R d (T d ) respectively represent the climbing rates of the upward climbing time T u and the downward climbing time T d . P0 and P1 respectively represent the aggregated power at the initial moment and the next moment, and FR(P1) represents the feasible region of the aggregated power at the next moment; The ramp duration is specifically: T u (R u ) = max k s.t. P0 + kR u ∈ FR(P1) T d (R d ) = maxk s.t. P0 - kR d ∈ FR(P1) Among them, T u , T d respectively represent the duration of upward climbing and the duration of downward climbing. P0 represents the aggregated power at the initial moment, and FR(P1) represents the feasible region of the aggregated power at the next moment; The power adjustable range is specifically: Among them, respectively represent the upper and lower limits of the aggregated power, and FR represents the feasible region of the aggregated power.

Images