A demand-side resource adjustable power domain aggregation method considering timing correlation

Abstract

The present invention relates to a method for aggregating adjustable power domains of demand-side resources taking into account time series correlation, comprising: obtaining grid parameters to be evaluated, and using time-divided linear boundary constraints to represent the adjustable power domains of adjacent time periods of demand-side resources; aggregating the adjustable power domains of each demand-side resource to obtain an aggregated result of the adjustable power domains of the cluster demand-side resources; based on the aggregated result of the adjustable power domains, using adaptive distributed robust optimization to solve the maximum area parameterized set within the approximate feasible domain to obtain an optimal feasible domain; based on the obtained optimal feasible domain of the adjustable power of a single time period, solving the union of multiple adjacent time period adjustable power domains to obtain a multi-time period adjustable power domain for the management and utilization of demand-side resources. Compared with the prior art, the present invention can accurately describe the dynamic characteristics of the power regulation capability of demand-side resources with higher accuracy, and is more conducive to the management and utilization of demand-side resources.

Description

Technical Field

[0001] The present invention relates to the technical field of demand-side resource prediction, and in particular to a method for demand-side resource adjustable power domain aggregation considering time sequence correlation. Background Art

[0002] Current research primarily uses general models, such as virtual generator models and virtual energy storage models, to characterize the adjustable power domain and quantify the power regulation capability of demand-side resources. However, virtual generator models lack consideration of demand-side resource state constraints, and virtual energy storage models fail to account for resource device ramping constraints. This inadequate consideration of these constraints can lead to inaccurately calculated adjustable power domains. Therefore, traditional general models are unsuitable for accurately describing the dynamic characteristics of demand-side resource power regulation capability.

[0003] During grid dispatch, demand-side resources report their power regulation range to the upper-level grid, and the grid dispatch center optimizes dispatch based on this reported regulation capability. Therefore, accurately quantifying the power regulation capability of demand-side resources is crucial for their participation in grid dispatch.

[0004] However, due to the characteristics of demand-side resources such as a wide variety, large number, small single capacity and geographical dispersion, separate modeling and quantitative analysis of each adjustable resource will bring a huge burden in computing and communication. At the same time, it also fails to take into account that the power boundary of the current period will be directly affected by the power of the previous period, ignores the temporal state correlation between the power variables of demand-side resources, and is difficult to accurately describe the dynamic characteristics of the power regulation capability of demand-side resources. The traditional external approximation method directly superimposes the parameters of each device as the parameters of the aggregated device, which easily makes the adjustable power domain larger than the original adjustable power domain and contains infeasible points.

[0005] Therefore, it is necessary to consider the time-series-correlated demand-side resource adjustable power domain aggregation method and build a general model that can reflect the adjustable capability of aggregated resources, which will help to manage and utilize demand-side resources and maximize the value of demand-side resources. Summary of the Invention

[0006] The purpose of the present invention is to overcome the defects of the above-mentioned existing technology that the classical external approximation method for adjustable power domain approximation is not suitable for the aggregation calculation of the power regulation capability of demand-side resources and cannot guarantee the safety and feasibility of the solution, and to provide a demand-side resource adjustable power domain aggregation method considering timing correlation.

[0007] The purpose of the present invention can be achieved by the following technical solutions:

[0008] A method for adjusting power domain aggregation of demand-side resources considering timing correlation includes the following steps:

[0009] Obtain the grid parameters to be evaluated, consider the temporal correlation of power regulation of demand-side resources, and use time-divided linear boundary constraints to represent the adjustable power domain of demand-side resources in adjacent time periods;

[0010] Aggregate the adjustable power domains of each demand-side resource to obtain an aggregated result of the adjustable power domains of the cluster's demand-side resources;

[0011] Based on the results of the adjustable power domain aggregation of the cluster demand-side resources, the adaptive distributed robust optimization is used to solve the maximum area parameterization set within the approximate feasible domain and obtain the optimal feasible domain.

[0012] According to the obtained optimal feasible region of adjustable power in a single period, the union of adjustable power domains of multiple adjacent periods is solved to obtain a multi-period adjustable power domain, which is used for the management and utilization of demand-side resources of the power grid to be evaluated.

[0013] Furthermore, the time-divided linear boundary constraints include resource power capacity constraints, resource ramp rate constraints, and resource status constraints.

[0014] Furthermore, the linear boundary constraints of the time periods are divided according to the upward power constraints and the downward power constraints, so as to obtain the adjustable power domains of the demand-side resources in adjacent time periods:

[0015]

[0016] Where, Ω i,k is the adjustable power range of the i-th flexibility resource in the current period k, is the upper boundary of the adjustable power range of the i-th flexibility resource in the current period k, is the lower boundary of the adjustable power range of the i-th flexibility resource in the current period k, P i,k,max 、P i,k,min are the upper and lower limits of the power of the i-th flexibility resource under the current time period k road density, r i,k,max 、r i,k,min are the upper and lower limits of the ramp rate of the i-th flexibility resource in the current period k, respectively. a, b, and c are dynamic characteristic parameters; x i,0 is the initial state of the i-th flexibility resource, x i,k,max 、x i,k,min are the upper and lower limits of the state variables of the i-th flexibility resource in the current period k; u i,k-1 are other variables in the i-th flexible resource period k-1, P i,k-1 is the adjustable power in the i-th flexible resource period k-1, P i,k is the adjustable power of the i-th flexibility resource in the current period k.

[0017] Furthermore, the process of aggregating the adjustable power domains of the various demand-side resources specifically includes the following steps:

[0018] Averaging the adjustable power domains of each demand-side resource to be aggregated to obtain a basic domain;

[0019] By translating and scaling the basic domain, the adjustable power domain of each demand-side resource is approximated to obtain the adjustable power approximate domain of each demand-side resource;

[0020] The translation and scaling parameters corresponding to the adjustable power approximate domains of each demand-side resource are linearly added to obtain the adjustable power domain of the cluster resources.

[0021] Furthermore, the expression of the adjustable power approximate domain is:

[0022]

[0023] Where, is the adjustable power approximate domain, are the approximate values ​​of the upper and lower bounds of power, i is the index of the flexibility resource, and k is the index of the time period; are the scaling coefficient and translation coefficient corresponding to the upper power boundary respectively; are the scaling coefficient and translation coefficient corresponding to the lower power boundary respectively; represents the actual power P of device i in time period k i,k Should be located at the upper and lower approximate boundaries and between;

[0024] The adjustable power approximate domain must satisfy: area maximum:

[0025]

[0026] For each flexibility resource i, the best approximation of its original power upper and lower bounds must be found:

[0027]

[0028] Furthermore, the expression of the adjustable power range of the cluster resources is:

[0029]

[0030] Where, is the lower boundary of the adjustable power range of cluster resources in time period k, is the upper boundary of the adjustable power range of cluster resources in time period k, P agg,kis the adjustable power domain of cluster resources in time period k, and N is the number of flexible resources in the cluster resources.

[0031] Furthermore, the method solves the maximum area parameterization set within the approximate feasible region by establishing an approximate feasible region boundary identification model based on two-stage ARO, and obtains the optimal feasible region;

[0032] The construction process of the approximate feasible region boundary identification model includes:

[0033] Introducing the uncertain variable ε k The actual power value P in time period k i,k Expressed as:

[0034]

[0035] In the formula, the actual power value P i,k Upper boundary and the lower boundary A linear combination of k ∈[0,1], T is the time set;

[0036] Will Equivalent conversion to about ε k Constraints where Ω i is the adjustable power domain of the i-th flexibility resource, and G is the uncertainty set:

[0037]

[0038] The expression of the approximate feasible region boundary identification model is obtained as follows:

[0039]

[0040] Where, is the approximate upper bound of the power of the i-th flexibility resource in the current period k, is the approximate value of the lower bound of the power of the i-th flexibility resource in the current period k, is the upper boundary of the adjustable power range of the i-th flexibility resource in the current period k, is the lower boundary of the adjustable power domain of the i-th flexibility resource in the current period k, Ω i,k is the adjustable power range of the i-th flexibility resource in the current period k.

[0041] Furthermore, the solution process of the approximate feasible region boundary identification model includes:

[0042] Phase 1: As the decision variable, the goal is to maximize the flexibility area of ​​the approximate feasible region boundary identification model, and obtain the uncertain variable εk The value of

[0043] The second stage: get the uncertain variable ε k After finding the value of G, we search for the optimal adjustable resource scheduling strategy under the worst scenario of G.

[0044] Furthermore, the expression of the multi-period adjustable power domain is:

[0045] Ω agg =Ω agg,1 ∪Ω agg,2 ∪…∪Ω agg,m

[0046] Where, Ω agg is the multi-period adjustable power domain after the aggregation of n flexibility resources on the demand side, Ω agg,1 ,…,Ω agg,k k adjustable power domains in a single period after the demand-side resource cluster aggregation.

[0047] Furthermore, the method further includes quantifying the power adjustable capability of the aggregated resources on the demand side by using the power adjustable capacity, and the calculation expression of the power adjustable capacity is:

[0048]

[0049]

[0050] Where R is the power adjustable capacity, P k is the actual power of the aggregated resource in time period k, and are the upper and lower boundaries of the adjustable power domain of the aggregated resources in time period k, and are the power capacities of the aggregated resources that can be adjusted up and down in time period k.

[0051] Compared with the prior art, the present invention has the following advantages:

[0052] (1) The present invention considers the temporal correlation of power regulation of demand-side resources and adopts time-divided linear boundary constraints to represent the adjustable power domain of adjacent time periods of demand-side resources; then, the adjustable power domain is aggregated based on the Minkowski sum and upper and lower boundary approximation methods, and the adaptive distributed robust optimization is used to solve the maximum area parameterization set in the approximate feasible domain for the aggregation result to obtain the optimal feasible domain, ensuring its aggregation optimality and the feasibility of the solution; finally, the multi-time period adjustable power domain suitable for day-ahead scheduling is obtained through the time period clustering method, which reduces the computational complexity; this scheme considers the temporal state correlation between the power variables of demand-side resources, can accurately describe the dynamic characteristics of the power regulation capability of demand-side resources, and is more conducive to the management and utilization of demand-side resources, as well as maximizing the value of demand-side resources.

[0053] (2) Experimental verification shows that the proposed scheme can reduce the proportion of infeasible points to 0, and the total scheduling cost only increases by 9.8%. Compared with the traditional aggregation method, the approximate method proposed in the present invention has higher accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] Figure 1 A flowchart of a method for demand-side resource adjustable power domain aggregation considering timing correlation provided in an embodiment of the present invention;

[0055] Figure 2 A schematic diagram of reference domains of three devices in an embodiment of a method for demand-side resource adjustable power domain aggregation considering timing correlation provided by one embodiment of the present invention;

[0056] Figure 3a A schematic diagram of the original domain and approximate domain of device 1 in an embodiment of a method for demand-side resource adjustable power domain aggregation considering timing correlation provided by one embodiment of the present invention;

[0057] Figure 3b A schematic diagram of the original domain and approximate domain of device 2 in an embodiment of a method for demand-side resource adjustable power domain aggregation considering timing correlation provided by one embodiment of the present invention;

[0058] Figure 3c A schematic diagram of the original domain and approximate domain of device 3 in an embodiment of a method for demand-side resource adjustable power domain aggregation considering timing correlation provided by one embodiment of the present invention;

[0059] Figure 4 A schematic diagram of adjustable power domains aggregated by three devices in an embodiment of a method for aggregating adjustable power domains of demand-side resources considering timing correlation provided by one embodiment of the present invention;

[0060] Figure 5A schematic diagram of power boundary comparison based on different approximation methods in an embodiment of a method for demand-side resource adjustable power domain aggregation considering timing correlation provided by one embodiment of the present invention;

[0061] Figure 6 A schematic diagram of an adjustable power range of a demand-side resource cluster calculated in an embodiment of a method for aggregating adjustable power domains of demand-side resources considering timing correlation provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0062] To make the objectives, technical solutions, and advantages of the embodiments of the present invention more clear, the technical solutions of the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Generally, the components of the embodiments of the present invention described and shown in the drawings herein can be arranged and designed in various different configurations.

[0063] Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the invention as claimed, but rather merely represents selected embodiments of the present invention. All other embodiments derived by persons of ordinary skill in the art based on the embodiments of the present invention without creative effort shall fall within the scope of protection of the present invention.

[0064] It should be noted that similar reference numerals and letters denote similar items in the following drawings, and therefore, once an item is defined in one drawing, it does not need to be further defined or explained in subsequent drawings.

[0065] Example 1

[0066] like Figure 1 As shown, this embodiment provides a method for demand-side resource adjustable power domain aggregation considering timing correlation, including the following steps:

[0067] S1: Obtain the grid parameters to be evaluated, consider the temporal correlation of the power regulation of demand-side resources, use time-divided linear boundary constraints to represent the adjustable power domain of demand-side resources in adjacent time periods, and quantify their power regulation capabilities;

[0068] S2: Based on the Minkowski sum and upper and lower bounds approximation method, an adjustable power domain aggregation method is proposed to aggregate the adjustable power domains of each demand-side resource and obtain the aggregated adjustable power domain results of the cluster's demand-side resources.

[0069] S3: Based on the results of the adjustable power domain aggregation of the cluster's demand-side resources, adaptive distributed robust optimization is used to solve the maximum area parameterization set within the approximate feasible domain, obtaining the optimal feasible domain to ensure its aggregation optimality and the feasibility of the solution;

[0070] S4: Based on the obtained optimal feasible region of adjustable power for a single time period, the union of adjustable power domains for multiple adjacent time periods is solved to obtain a multi-period adjustable power domain suitable for day-ahead scheduling. This reduces the computational complexity and is used for the management and utilization of demand-side resources of the power grid to be evaluated, such as day-ahead scheduling and other power grid scheduling schemes.

[0071] This embodiment uses scientific demonstration to verify and illustrate the effectiveness of the technology used in the present invention.

[0072] The verification environment is set as follows: Taking a specific demand-side resource device as an example, the time period is set to 1 hour, and the adjustable power domain of its adjacent time periods is depicted. Considering three distributed energy storage devices with different parameters as the demand-side resources to be aggregated, the parameters of the three devices and the benchmark domain are shown in Table 1:

[0073] Table 1 Equipment parameters

[0074]

[0075] In order to analyze the effectiveness of this method when processing a large amount of data, the Monte Carlo sampling method is used to randomly generate 1000 data. It is assumed that the device parameters obey the probability distribution function in Table 2:

[0076] Table 2 Probability distribution of equipment parameters

[0077]

[0078] Specifically, in step S1, the adjustable power domain of adjacent time periods of the demand-side resources is represented by linear boundary constraints in time periods in consideration of the temporal correlation of the power regulation of the demand-side resources.

[0079] It should be noted that the linear boundary constraints of the time periods are used to represent the adjustable power domain of the demand-side resources in adjacent time periods.

[0080] Assume that the scheduling period T of a flexible resource is divided into m periods, and the power of the current period is P k , the power of the previous period is P k-1 , where k = 1, 2, ..., m.

[0081] The constraint space of flexible resource operation can be expressed by a set of inequality constraints, namely:

[0082] 1) Resource power capacity constraints:

[0083] P k,min ≤P k ≤P k,max (1)

[0084] Among them, P k,max、P k,min are the upper and lower limits of power under the road density k in the current period.

[0085] 2) Resource ramp rate constraints, i.e., constraints on the power change between adjacent time periods. The power that a demand-side resource can generate in the current period is affected not only by its own equipment parameters and scheduling timescale, but also by the power constraints of the previous period.

[0086] r k,min ≤P k -P k-1 ≤r k,max (2)

[0087] Where: r k,max 、r k,min are the upper and lower limits of the climbing rate in the current period k.

[0088] 3) Resource status constraint: The operating status of demand-side resources is a continuous cumulative quantity that changes with the power changes in each period, such as the remaining capacity of energy storage, the temperature of the temperature control system, etc.

[0089] The capacity constraint status of energy storage load is:

[0090]

[0091] SOC k,min ≤SOC k ≤SOC k,max (4)

[0092] Where: E is the rated capacity of the energy storage load; SOC k is the state of charge of the energy storage load in the current period k, SOC k,max , SOC k,min They are the upper and lower limits of the state of charge of the energy storage load in the current time period k; the parameter η represents the charging and discharging efficiency.

[0093] For temperature control loads such as air conditioners, it is necessary to control the indoor temperature within the human comfort temperature range during operation. The relationship between indoor temperature and air conditioner power is as follows:

[0094]

[0095] T r,k,min ≤T r,k ≤T r,k,max (6)

[0096] Where: T r,k is the indoor temperature in the current period k, T r,k,max 、T r,k,min are the highest and lowest temperatures that meet the comfort requirements in the current period k; To,k-1 is the outdoor temperature; R and C are the heat capacity and equivalent resistance of the temperature control system respectively; parameter η is the energy efficiency ratio.

[0097] The operational characteristics of adjustable resources are different, and the state constraints that need to be satisfied are also different. The state constraints of various demand-side resources can be uniformly expressed as:

[0098] x k,min ≤ax k-1 +bu k-1 +cP k ≤x k,max (7)

[0099] Where: a, b, c are dynamic characteristic parameters; x k-1 is the state variable of the previous period, x k,max 、x k,min are the upper and lower limits of the state variables in the current period k; u k-1 For other variables.

[0100] In order to highlight the influence of state constraints on the power relationship between two adjacent time periods, given the initial state x0, equation (7) can be rewritten as follows:

[0101] x k,min ≤a(x0+cP k-1 )+bu k-1 +cP k ≤x k,max (8)

[0102] Rearranging equations (1), (2), and (8), the constraints are divided into upward power constraints ① to ③ and downward power constraints ④ to ⑥, and the adjustable power domain Ω of the demand-side flexibility resource is obtained. i,k as follows:

[0103]

[0104] Where: is the upper boundary of the adjustable power domain; is the lower boundary of the adjustable power domain.

[0105] In step S2, a method for approximating and clustering the adjustable power domain of demand-side flexibility resources is proposed, which includes three steps: setting the reference domain, approximating the adjustable power domain, and aggregating the adjustable power domain. It should be noted that:

[0106] Since demand-side flexibility resources are diverse and numerous, and their individual capacities are relatively small, they need to be aggregated to form a cluster of adjustable power domains:

[0107]

[0108] In the formula: Symbol represents Minkowski summation; Ω agg,k Ω is the adjustable power domain after the aggregation of n flexibility resources on the demand side in time period k; 1,k ,…,Ω n,k is the adjustable power domain of each of the n flexibility resources.

[0109] The specific process is:

[0110] 1) The benchmark domain of resource aggregation, the results are as follows Figure 2 As shown in this paper, the adjustable power domain Ω of all resource devices to be aggregated is i,k (i=1,..,n) and average to get the reference domain Ω 0,k ,Ω 0,k The form of is the same as that of formula (9), and its parameters are the average of n resource parameters.

[0111]

[0112] 2) Approximation of the adjustable power domain of a single resource, the results are as follows Figure 3a-3c As shown. By translating and scaling the reference domain to approximate the boundaries of the original adjustable power domain, the adjustable power approximate domain of each resource is obtained. First, the upper and lower boundaries of the reference domain are translated and scaled to obtain the approximate values ​​of the upper and lower power boundaries:

[0113]

[0114] Where: are the approximate values ​​of the upper and lower bounds of power, respectively; are the scaling coefficient and translation coefficient corresponding to the upper power boundary respectively; are the scaling coefficient and translation coefficient corresponding to the lower power boundary respectively; Formula (14) represents the actual power P of device i in time period k i,k Should be located at the upper and lower approximate boundaries between.

[0115] By the approximate values ​​of the upper and lower power limits The adjustable power approximation domain

[0116]

[0117] The approximate domain needs to meet the following conditions:

[0118] ① In order to obtain more flexibility and adjustability, the approximate domain is the original feasible region Ω i,k The best approximation domain must satisfy the area maximum.

[0119]

[0120] ② The feasibility of the solution must be guaranteed. Any power within the range All of these can be achieved by scheduling demand-side resource devices without violating constraints. Specifically, for each device i, the optimal approximation of its original power upper and lower bounds must be found.

[0121]

[0122] 3) Adjustable power domain aggregation of cluster demand-side resources, the results are as follows Figure 4 After the adjustable power domain of a single resource is approximated, the adjustable power domain of the cluster resources can be aggregated by directly linearly adding the translation and rotation parameters of each approximate adjustable power domain.

[0123]

[0124] In step S3, in order to ensure the optimality of the approximate feasible region aggregation and the feasibility of the solution, an approximate feasible region boundary identification model based on two-stage ARO is established. It should be noted that:

[0125] The feasible region is composed of decision variables Described, and the upper and lower boundaries is a time variable, for which we introduce the uncertain variable ε k The actual power value P in time period k i,k Expressed as:

[0126]

[0127] In the formula, the actual power value P i,k Can be seen as the upper boundary and the lower boundary A linear combination of k ∈[0,1]. When ε k =1, When ε k =0,

[0128] Through the above variable transformation, we can Equivalent conversion to about ε k Constraints Where G is an uncertain set:

[0129]

[0130] The problem of approximate aggregation of adjustable power domains is formulated as a model using ARO:

[0131]

[0132] Satisfy ①-⑥ in formula (8).

[0133] A parameterized set is used to characterize the variation of aggregate power over time, and a two-stage adaptive robust optimization model is established to solve the maximum area parameterized set within the approximate feasible region. The first stage objective is to maximize the flexibility area, and the decision variable is In the second stage, when the uncertain variable ε k After the value of is determined, the second-stage objective function is to find the optimal adjustable resource scheduling strategy under the worst scenario of G.

[0134] In step S4, the single-period adjustable power domain of the demand-side resource cluster needs to be aggregated into multiple periods. Therefore, the present invention proposes a method for calculating the multi-period adjustable power domain of the demand-side resource cluster. It should be noted that:

[0135] According to step S3, the adjustable power domain of the demand-side resource cluster in a single period is obtained, and the union of the adjustable power domains of multiple adjacent periods is solved to obtain the multi-period adjustable power domain, thereby achieving dimensionality reduction calculation of the high-dimensional problem, namely:

[0136] Ω agg =Ω agg,1 ∪Ω agg,2 ∪…∪Ω agg,m (25)

[0137] Where: the symbol “∪” means to find the union; Ω agg Ω is the multi-period adjustable power domain after the aggregation of n flexibility resources on the demand side; agg,1 ,…,Ω agg,k It is a single-period k-adjustable power domain aggregated by the demand-side resource cluster.

[0138] In order to more intuitively quantify the power adjustment capability of aggregated resources, the power adjustment capacity R is defined as follows:

[0139]

[0140]

[0141] Where: P k is the actual power of the aggregated resource in time period k, are the upper and lower boundaries of the adjustable power domain of the aggregated resources in time period k, respectively; are the power capacities of the aggregated resources that can be adjusted up and down in time period k.

[0142] Since the adjustable power domain of a single demand-side flexibility resource is approximated and then aggregated, the demand-side resource adjustment power cost function needs to undergo corresponding equivalent changes:

[0143]

[0144] Where: r k is the power regulation amount; is the cluster scaling parameter, is the cluster translation parameter; a i 、b i and c i is the operating cost coefficient of equipment i; is the operating cost coefficient of the aggregated resource.

[0145] Figure 5 This is a comparison chart of the approximate method proposed in this paper and other current approximate methods, using the average scale factor φ of the approximate adjustable power domain relative to the original adjustable power domain. ave is 0.8124 compared to the φ of the "zonotope" method. ave The value is closer to 1, Figure 6 This is the result of aggregating the adjustable power domains of clustered ES, EV, and HVAC. Compared to the RO algorithm, the proposed ARO algorithm can reduce the proportion of infeasible points to 0, while increasing the total scheduling cost by only 9.8%. Under the same conditions, the proportion of infeasible points using the external approximation method reaches 13.62. Therefore, compared to traditional aggregation methods, the proposed approximation method has higher accuracy.

[0146] The above describes in detail the preferred embodiments of the present invention. It should be understood that those skilled in the art can make numerous modifications and variations based on the concepts of the present invention without inventive effort. Therefore, any technical solutions that can be derived by those skilled in the art through logical analysis, reasoning, or limited experimentation based on the concepts of the present invention and the prior art should be within the scope of protection defined by the claims.

Claims

1. A method for demand-side resource adjustable power domain aggregation considering timing correlation, characterized in that: The following steps are involved: Obtain the grid parameters to be evaluated, consider the temporal correlation of power regulation of demand-side resources, and use time-divided linear boundary constraints to represent the adjustable power domain of demand-side resources in adjacent time periods; Aggregate the adjustable power domains of each demand-side resource to obtain an aggregated result of the adjustable power domains of the cluster's demand-side resources; Based on the results of the adjustable power domain aggregation of the cluster demand-side resources, the adaptive distributed robust optimization is used to solve the maximum area parameterization set within the approximate feasible domain and obtain the optimal feasible domain. Based on the obtained optimal feasible region of adjustable power in a single period, the union of adjustable power domains in multiple adjacent periods is solved to obtain a multi-period adjustable power domain, which is used for the management and utilization of demand-side resources of the power grid to be evaluated; The linear boundary constraints of the time periods are divided according to the upward power constraints and downward power constraints to obtain the adjustable power domain of the demand-side resources in adjacent time periods: Where, For the i Flexible resources in the current period Adjustable power domain under For the i Flexible resources in the current period The upper boundary of the lower adjustable power domain, For the i Flexible resources in the current period The lower boundary of the lower adjustable power domain, 、 Respectively i Flexible resources in the current period The upper and lower limits of power under road density, 、 Respectively i Flexible resources in the current period The upper and lower limits of the climbing rate under is the dynamic characteristic parameter; For the i The initial state of a flexibility resource, 、 Respectively i Flexible resources in the current period The upper and lower limits of the state variables under For the i Flexible resource time slots Other variables under For the i Flexible resource time slots Adjustable power under For the i Flexible resources in the current period Adjustable power under 2. A method for demand-side resource adjustable power domain aggregation considering timing correlation according to claim 1, characterized in that: The time-divided linear boundary constraints include resource power capacity constraints, resource ramp rate constraints, and resource status constraints.

3. The method for demand-side resource adjustable power domain aggregation considering timing correlation according to claim 1, characterized in that: The process of aggregating the adjustable power domains of various demand-side resources specifically includes the following steps: Averaging the adjustable power domains of each demand-side resource to be aggregated to obtain a basic domain; By translating and scaling the basic domain, the adjustable power domain of each demand-side resource is approximated to obtain the adjustable power approximate domain of each demand-side resource; The translation and scaling parameters corresponding to the adjustable power approximate domains of each demand-side resource are linearly added to obtain the adjustable power domain of the cluster resources.

4. The method for demand-side resource adjustable power domain aggregation considering timing correlation according to claim 3, characterized in that: The expression of the adjustable power approximate domain is: Where, is the adjustable power approximate domain, 、 are the approximate values ​​of the upper and lower bounds of power, i is the label of the flexibility resource, is the time period number; 、 are the scaling coefficient and translation coefficient corresponding to the upper power boundary respectively; 、 are the scaling coefficient and translation coefficient corresponding to the lower power boundary respectively; Representation device In the period The actual power Should be located at the upper and lower approximate boundaries and between; The adjustable power approximate domain must satisfy: area maximum: ； For each flexibility resource The best approximation to the upper and lower bounds of its original power must be found: 。 5. The method for demand-side resource adjustable power domain aggregation considering timing correlation according to claim 4, characterized in that: The expression of the adjustable power range of the cluster resources is: Where, For the period The lower boundary of the cluster resource adjustable power domain, For the period The lower cluster resources can adjust the upper boundary of the power domain. For the period The cluster resources can be adjusted in the power domain. N The number of flexible resources in the cluster.

6. The method for demand-side resource adjustable power domain aggregation considering timing correlation according to claim 1, characterized in that: The method solves the maximum area parameterization set in the approximate feasible region by establishing an approximate feasible region boundary identification model based on a two-stage ARO, and obtains the optimal feasible region; The construction process of the approximate feasible region boundary identification model includes: Introducing uncertain variables The period The actual power value under Expressed as: In the formula, the actual power value Upper boundary and the lower boundary A linear combination of , is the time collection; Will Equivalent conversion to Constraints ,in For the i Adjustable power domain with flexibility resources, For an uncertain set: The expression of the approximate feasible region boundary identification model is obtained as follows: Where, For the i Flexible resources in the current period An approximate upper bound on the lower power, For the i Flexible resources in the current period An approximate lower bound on the lower power, For the i Flexible resources in the current period The upper boundary of the lower adjustable power domain, For the i Flexible resources in the current period The lower boundary of the lower adjustable power domain, For the i Flexible resources in the current period The adjustable power domain below.

7. The method for demand-side resource adjustable power domain aggregation considering timing correlation according to claim 6, characterized in that: The solution process of the approximate feasible region boundary identification model includes: Phase 1: As the decision variable, the goal is to maximize the flexibility area of ​​the approximate feasible region boundary identification model, and obtain the uncertain variable The value of Phase 2: Obtaining uncertain variables After finding the value of G, we search for the optimal adjustable resource scheduling strategy under the worst scenario of G.

8. The method for demand-side resource adjustable power domain aggregation considering timing correlation according to claim 1, characterized in that: The expression of the multi-period adjustable power domain is: Where, For the demand side The multi-period adjustable power domain after the aggregation of flexible resources, A single period for the aggregation of demand-side resource clusters Adjustable power domain.

9. The method for demand-side resource adjustable power domain aggregation considering timing correlation according to claim 1, characterized in that: The method further includes quantifying the power adjustable capability of the aggregated resource on the demand side by using a power adjustable capacity, wherein the calculation expression of the power adjustable capacity is: Where, For power adjustable capacity, To aggregate resources in time period The actual power, and Aggregate resources in time periods Adjustable upper and lower boundaries of the power domain, and Aggregate resources in time periods Power handling that can be adjusted up and down.

Images