Newton-ADMM-based power distribution network distributed resource cooperative regulation and control method

Abstract

The invention relates to the field of power distribution network resource cooperative regulation and control, and discloses a Newton-ADMM-based power distribution network distributed resource cooperative regulation and control method, which comprises the following steps: S1, constructing a global optimization model comprising global coupling constraints and local constraints of each distributed resource, the global optimization model is decomposed into a plurality of local optimization sub-problems and a coordination problem processed by a power distribution system operator by introducing auxiliary variables corresponding to local regulation and control variables of all distributed resources; and S2, in a one-time iteration process, firstly solving the local optimization sub-problem in parallel by local nodes of each distributed resource so as to update the local regulation and control variable. A distributed resource collaborative regulation and control framework oriented to the three-phase unbalanced power distribution network is constructed through time sequence coupling characteristics of a temperature control load operation environment, and a global optimal solution can be obtained through limited information interaction between a DSO and a local user under the framework.

Description

Technical Field

[0001] This invention relates to the field of coordinated regulation of distribution network resources, specifically a method for coordinated regulation of distributed resources in distribution networks based on Newton-ADMM. Background Technology

[0002] With the increasing penetration of renewable energy and the accelerating trend of power system decentralization, modern distribution networks are facing increasingly complex operational challenges. The large-scale integration of distributed energy resources (DERs) enhances system flexibility but also brings significant challenges to regulation, especially for flexible loads with spatiotemporal coupling characteristics, such as temperature-controlled loads (TCLs). TCLs (air conditioners, electric water heaters, and refrigerators) account for a large proportion of residential and commercial electricity consumption, and their operating characteristics often result in peak electricity demand coinciding closely with system peak load. However, the inherent thermal energy storage capacity of TCLs also makes them ideal for demand response (DR) and load dispatching. Through reasonable coordinated control strategies, peak shaving and valley filling, and network congestion mitigation can be achieved without affecting user comfort.

[0003] Traditional distribution network resource scheduling primarily employs centralized optimization methods, where the Distribution System Operator (DSO) collects network-wide data and uniformly solves for the optimal decision. In recent years, distributed optimization strategies have gradually become a research hotspot. In a distributed framework, the decision-making process is decentralized among various participating entities; the DSO is only responsible for coordinating global constraints, while users autonomously adjust their loads based on local information. This approach not only reduces the demand for communication and computing resources but also effectively protects user privacy. Existing research has explored the application of various distributed optimization algorithms in distribution networks.

[0004] Defects and shortcomings of existing technology:

[0005] Current distributed resource regulation technologies have significant limitations. Centralized methods are difficult to apply to large-scale distribution networks due to privacy risks and communication / computation burdens. While existing distributed strategies protect user privacy, they generally ignore network constraints. For example, temperature-controlled load regulation fails to integrate power flow constraints of three-phase unbalanced distribution networks, easily leading to voltage exceedance safety issues. Furthermore, mainstream distributed algorithms such as ADMM rely on dual decomposition, resulting in slow iterative convergence and high computational and communication overhead. Although some studies have attempted to optimize ADMM by introducing adaptive parameters or nested column generation algorithms, they have not broken through the essence of dual methods, and the improvement in convergence performance is limited. On the other hand, existing temperature-controlled load models often simplify the coupling relationship between their dynamic thermal characteristics and grid operation constraints, resulting in insufficient adaptability of regulation strategies under complex actual operating conditions, making it difficult to balance user comfort with the economical and safe operation of the distribution network. Summary of the Invention

[0006] To address the shortcomings of existing technologies, this invention provides a distributed resource coordinated control method for distribution networks based on Newton-ADMM, which solves the problems of power flow limitations in unintegrated three-phase unbalanced distribution networks, which easily lead to voltage overruns, slow iterative convergence speed, and insufficient adaptability.

[0007] To achieve the above objectives, the present invention provides the following technical solution: a distributed resource coordinated control method for distribution networks based on Newton-ADMM, comprising the following steps:

[0008] S1: Construct a global optimization model that includes global coupling constraints and local constraints of each distributed resource, and decompose the global optimization model into multiple local optimization sub-problems and a coordination problem handled by the power distribution system operator by introducing auxiliary variables corresponding to the local control variables of each distributed resource.

[0009] S2: In one iteration, the local optimization subproblem is first solved in parallel by the local nodes of each distributed resource in order to update the local control variables;

[0010] S3: Then, the power distribution system operator updates the auxiliary variables based on the local control variables updated by all local nodes;

[0011] S4: Next, using the Hessian matrix of the objective function of the global optimization model, update the dual variable used to coordinate the local control variable and the auxiliary variable;

[0012] S5: Repeat steps S2 to S4 until the convergence condition is met, and obtain the optimal control strategy for each distributed resource.

[0013] Preferably, the objective function of the global optimization model includes user electricity cost and user thermal comfort.

[0014] Preferably, the global coupling constraints of the global optimization model include linearized distribution network power flow equation constraints and distribution network power flow security constraints.

[0015] Preferably, the distributed resource is a temperature-controlled load.

[0016] Preferably, the local constraints of the temperature-controlled load are modeled based on a discretized linear thermal model that describes the relationship between indoor temperature, outdoor temperature and temperature-controlled load power.

[0017] Preferably, the local control variable is the active power sequence of the temperature-controlled load for each user.

[0018] Preferably, the decomposition is achieved by using the consistency between the local control variable and the auxiliary variable as a coupling constraint.

[0019] Preferably, the update of the dual variable is based on the value of the dual variable in the previous round and a correction term consisting of the difference between the Hessian matrix and the local control variable and auxiliary variable in the current round.

[0020] Preferably, the global optimization model is constructed as a convex quadratic programming problem with a separable objective function, and the Hessian matrix is ​​a constant matrix.

[0021] Preferably, the linearized power flow equations of the distribution network are DistFlow power flow equations.

[0022] This invention provides a method for coordinated control of distributed resources in a distribution network based on Newton-ADMM.

[0023] It has the following beneficial effects:

[0024] 1. This invention takes into account the temporal coupling characteristics of temperature-controlled load operating environment and constructs a distributed resource collaborative control framework for three-phase unbalanced distribution networks. Under this framework, DSO and local users can obtain the global optimal solution through limited information interaction.

[0025] 2. This invention transforms the distributed resource collaborative regulation problem into a standard convex quadratic programming problem in a compact form through derivation and transformation, and further introduces auxiliary variables to decompose the constraints of the optimization problem, ensuring the distributed implementation of the proposed strategy.

[0026] 3. This invention proposes a Newton-Alternating Directional Multiplier Method (Newton-ADMM) based on ADMM. This algorithm updates the dual variables using the Hessian matrix of the objective function, significantly improving convergence performance through Newton-based acceleration. The proposed algorithm is used in the distributed energy management framework presented in this paper, effectively reducing computational workload and communication costs while ensuring optimality. Attached Figure Description

[0027] Figure 1 This invention provides a distributed resource collaborative regulation framework.

[0028] Figure 2 This is a schematic diagram of the three-phase unbalanced radial power distribution network of the present invention;

[0029] Figure 3 This is a schematic diagram of the improved IEEE 123 node unbalanced distribution network of the present invention;

[0030] Figure 4 This is a schematic diagram of a user's 24-hour non-temperature-controlled load in an embodiment of the present invention;

[0031] Figure 5This is a schematic diagram of the 24-hour electricity price curve for users in an embodiment of the present invention;

[0032] Figure 6 This is a schematic diagram of the user's ambient temperature over 24 hours in an embodiment of the present invention;

[0033] Figure 7 This is a schematic diagram of the lowest 24-hour voltage of the system before and after using the strategy described herein in an embodiment of the present invention;

[0034] Figure 8 This is a schematic diagram of the 24-hour load curves of the system before and after using the strategy described herein in an embodiment of the present invention;

[0035] Figure 9 This is a schematic diagram of the voltage distribution of phase a in a 15h distribution network system according to an embodiment of the present invention;

[0036] Figure 10 This is a schematic diagram of the temperature-controlled load situation of phase a of the power distribution network during a 15-hour period, as described in an embodiment of the present invention.

[0037] Figure 11 This is a schematic diagram of voltage transformation during the convergence process of phase a at node 96 of the distribution network in a 15-hour system according to an embodiment of the present invention.

[0038] Figure 12 This is a schematic diagram of power transformation during the convergence process of phase a at node 96 of the distribution network in a 15-hour system according to an embodiment of the present invention. Detailed Implementation

[0039] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0040] Example:

[0041] Please see the appendix Figure 1 - Appendix Figure 2 This invention provides a method for coordinated control of distributed resources in a distribution network based on Newton-ADMM, comprising:

[0042] This scheme first establishes a distributed resource collaborative control framework for three-phase unbalanced distribution networks, such as... Figure 1Within this framework, the DSO (Distributed System Optimization) is responsible for coordinating global security constraints (power flow equations, voltage limits), while local users autonomously adjust the operating status of temperature-controlled loads based on their own thermal comfort preferences and electricity costs. By introducing auxiliary variables, the original problem is decomposed into local optimization sub-problems and global coordination problems, ensuring the feasibility of distributed implementation. The DSO and users only need to exchange limited information such as aggregated power and boundary voltage, without sharing detailed privacy data, thus protecting user privacy and reducing communication burden.

[0043] 1) Network Model

[0044] An unbalanced distribution network model is introduced, and subsequent distributed resource regulation strategy modeling is based on this model. Given... Figure 1 The diagram shows a structure with N+1 nodes and an unbalanced phase {ph}. a ,ph b ,ph c The radial distribution network has a set of nodes as follows: Where 0 represents the first term

[0045] point, It is the index set of all non-first segment nodes. For a node... set up Let represent the node directly connected to node j and preceding node j, so the set of all paths can be represented as set up Let T be the set of nodes that follow node j but do not contain j. Let the scheduling time window consist of NT equal intervals, i.e., T = {1, 2, ..., NT}. For any node... and time period Use V i (t) represents the corresponding voltage value, v i (t)=|V i (t)| 2 s is the square of the voltage amplitude. i (t)=p i (t)+jq i (t) represents the active and reactive power consumed by the node. For each line and time period Let z ij =r ij +jx ij S represents the line impedance. ij (t)=P ij (t)+jQ ij (t) represents the active and reactive power flowing from node i to j. All of the above parameters are expressed in per-unit values.

[0046] Subsequently, the linear DistFlow power flow equations are used to model the power flow of the distribution network in a compact form:

[0047]

[0048] in,

[0049] P(t) = [P BP(1)1 (t),…,P BP(N)N (t)] T ;

[0050] Q(t) = [Q BP(1)1 (t),…,Q BP(N)N (t)] T ;

[0051] p(t) = [p1(t), ..., p N (t)] T ;

[0052] q(t)=[q1(t),…,q N (t)] T ;

[0053] v(t) = [v1(t), ..., v N (t)] T ;

[0054] D r It is an N×N dimensional diagonal matrix, where the j-th diagonal block corresponds to r. ij D x Similarly, corresponding to x ij , Let m0 be the correlation matrix of a radial distribution network. The first line. When A certain column of corresponding lines Then, all elements in this column except for the elements corresponding to i and j are 0. Substituting equation 1c and...

[0055] Substituting equation 1d into equation 1e, we can derive:

[0056]

[0057] in

[0058] 2) User Model

[0059] Each node is configured with an aggregation device to adjust the electricity consumption behavior of its users. This paper categorizes user loads into temperature-controlled loads (TCL) and non-temperature-controlled loads (non-TCL), and sets the temperature-controlled loads to have significant control flexibility. Taking user u connected to node i as an example, its non-TCL within the scheduling time window NT is represented by load forecasting as follows: and For temperature-controlled load equipment, assuming it operates under constant power factor, the user set under each node is represented as follows: Users accessing node i The temperature-controlled load operation satisfies the following relationship:

[0060] q i,u =η i,u p i,u (Equation 3a);

[0061]

[0062] Where p i,u =[p i,u (1),…,p i,u (NT)] T and q i,u =[q i,u (1),…,q i,u (NT)] T PF represents the predicted temperature control load for user u at node i. i,u The power factor is used. A discretized linear thermal model is used to simulate indoor temperature variations:

[0063]

[0064] In the formula and (℃ / kWh) is a positive parameter. T(p) i,u (0))(°C) is the initial indoor temperature of user u at access node i, T(p i,u (t) (℃) represents the indoor temperature during time period t. o (t)(°C) represents the outdoor temperature within time period t, Δt(h) represents the duration of each time period t, and P base (kWh) is the reference power. Let T(p) i,u )=[T(p i,u (1)),…,T(p i,u (NT))] t This allows (4a) and (4b) to be further transformed into a compact form:

[0065] T(p i,u ) = J i,u -E i,u p i,u (Equation 5);

[0066] The expressions for each term in the formula are as follows:

[0067] J i，u =H i，u T i，u (0)+D i,u To

[0068] T i，u (0) = [T i，u (0),T i，u (0)…,T i，u (0)] T

[0069] T o =[T0(1),T o (2)…T o (NT)] T

[0070]

[0071] 3) Distributed resource collaborative regulation strategy

[0072] 1. User comfort calculation:

[0073] During time period t, the thermal comfort level U(p) of user u accessing node i is... i,u (t)) using the current room temperature T(p) i,u Comfort level under (t) and user's maximum thermal comfort U max The differences between them are expressed as follows:

[0074] U(p i,u (t))=U max -f*(T(p i,u (t))-TB) 2 (Equation 6);

[0075] In the formula, f>0 is a constant, and the comfort level of user u accessing node i within the scheduling time T can be characterized as:

[0076]

[0077] Where TB = [TB, TB, ..., TB] T The optimal temperature for the user.

[0078] 2. Distributed resource regulation strategy model:

[0079] A. Objective function

[0080] The goal of power distribution system operators is to minimize electricity costs for users while ensuring their thermal comfort. The objective function is shown below:

[0081]

[0082] In the formula, μ represents the monetary marginal utility of electricity cost, and π T =[π(1),…π(NT)] T(Yuan / kWh) represents the predicted user-side electricity price for the dispatch period. sum =[p sum (1),…p sum (NT)] T This represents the total electricity consumption of the power distribution system during each time period within the dispatch cycle.

[0083] B. Constraints

[0084] The constraints of the scheduling strategy proposed in this paper mainly include the linear DistFlow equation corresponding to Equation 1, the upper and lower limits of active power of the temperature-controlled load corresponding to Equation 4c, and the discretized linear thermal model corresponding to Equation 5. Power flow constraints are further considered under the above constraints.

[0085]

[0086] v min <v(t)<v max (Equation 9d);

[0087]

[0088] This represents the upper limit of the power supply capacity of the distribution network. Observing (Equation 8), it can be found that under the current objective function, the coordinated control strategy of distributed resources needs to be solved independently by the distribution system operator, i.e., centralized calculation.

[0089] 3. Distributed solution framework based on Newton-ADMM

[0090] I. Objective Function Reconstruction

[0091] To enable the objective function to be solved using a distributed algorithm, substitute (Equation 5), (Equation 9a), and (Equation 9c) into the objective function (Equation 8):

[0092]

[0093] in:

[0094]

[0095] The above constraints include the linear DistFlow equation corresponding to Equation 1, the upper and lower limits of active power for the temperature-controlled load corresponding to Equation 4c, and the power flow constraint (Equation 9). It can be observed that 'a' does not affect the optimal solution of the objective function, which is equivalent to:

[0096]

[0097] At this point, all terms in the objective function are related to the local agent, and a distributed algorithm can be used to solve it. Furthermore, (Equation 11) is rearranged into a compact quadratic programming form. Let... This represents the active power sequence of n user temperature-controlled loads under node i. Given the active power sequence of the temperature-controlled loads at all nodes of the system, the objective function can be expressed in a compact form as a general rigorous convex quadratic programming problem with separable objective functions, coupled constraints, and local constraints:

[0098]

[0099] x in the objective function i The local constraints are (Equation 4c), and the coupled constraints with respect to x include (Equation 2) and (Equations 9a)-(Equations 9e). All of the above constraints are convex sets, and for any node i, we have:

[0100]

[0101] B i =-2f[(J i,u1 -TB) T E i,u1 ,…,(J i,u1 -TB) T E i,un(i) )] T +μP base Δt[π T ,…π T ] T (Equation 13b);

[0102] The objective function is expressed as:

[0103]

[0104] Introducing auxiliary variables Let z = x, the objective function is reconstructed as follows:

[0105]

[0106] stx-z=0 (Equation 15b);

[0107] Where g(z) is a closed positive convex function, and the closed convex set Through exponential function Included in the objective function:

[0108]

[0109] II. Distributed Solving Based on Newton-ADMM

[0110] Based on the ADMM algorithm, a Newton-ADMM algorithm is proposed that accelerates convergence by utilizing the Hessian matrix of the objective function, and is used to solve the strategy proposed in this paper.

[0111] The augmented Lagrangian function of (Equation 15) is as follows:

[0112]

[0113] Let F be a positive definite matrix, and define... as follows:

[0114]

[0115] The augmented Lagrangian function of (Equation 17) is:

[0116]

[0117] Let M = F T F, then (Equation 18) is equivalent to:

[0118]

[0119] Following the above derivation, we obtain a "Newton-ADMM" algorithm. Unlike the traditional ADMM algorithm, the "Newton-ADMM" algorithm, like Newton's method, uses matrix M to update the dual variable, achieving fast convergence. This algorithm can be applied to the reconstructed objective function described above.

[0120]

[0121] As can be seen from the above, x i Only related to local constraints, z = x is coupled with global constraints, H i Let M be the objective function and the Hessian matrix. i For H i When the algorithm achieves its fastest convergence performance, let... Algorithm 1 shows the solution process of the problem using the "Newton-ADMM" algorithm.

[0122] Algorithm 1, Newton-ADMM solution process:

[0123] Initialization: Set k = 0, x 0 ,z 0 ,λ 0 And each node i will F i Send to the independent power distribution system operator.

[0124] When k≥0, update the variables alternately according to steps (S1)-(S4) in the Newton-ADMM until convergence:

[0125] S1: Update x

[0126]

[0127] This step can be solved locally by node i: f i

[0128]

[0129] S2: Update z

[0130]

[0131] This step was updated by the independent power distribution system operator;

[0132] S3: Update λ

[0133] λ k+1 =λ k +βH(x k+1 -z k+1 );

[0134] Right now:

[0135]

[0136] It can be updated by an independent power distribution system operator or by each node i;

[0137] S4: Number of update iterations

[0138] In the process of coordinated regulation of temperature-controlled loads, the variable x in step 1 of Algorithm 1... i The local node i is responsible for updating the constraints, while steps 2 and 3 are performed by the distribution system operator. During the solution process, the local constraints are known only for the local node i, and the coupling constraints are known only for the DSO. Therefore, under the "Newton-ADMM" solution framework, the local node i and the distribution system operator only need to exchange limited information x, z, λ to converge to the optimal result.

[0139] See attached document Figure 3 - Appendix Figure 6 As shown, a numerical simulation was performed on the improved IEEE 123 node distribution network. The time interval Δt = 1 hour, the number of time intervals NT = 24, TB = 22 (°C), and T... i,u (0) = 23 (℃), f = 6.12, PF i,u=0.9, μ=1. Furthermore, let the upper and lower limits of the voltage be 1.05pu and 0.95pu respectively, and v0=1.05 2 Pbase = 100kW, step size β = 1. The user's 24-hour non-temperature-controlled load curve, electricity price curve, and ambient temperature are as follows: Figures 4-6 As shown.

[0140] Strategy effectiveness verification

[0141] First, the effectiveness of the proposed strategy is tested. The strategy is applied to the designed scenario, and the system operation before and after using the strategy is compared. Without the proposed strategy, the local node decides the power consumption behavior of the temperature-controlled load based on its own operating conditions. Figure 7 and Figure 8 The system's performance before and after using the strategy proposed in this paper is shown.

[0142] Figure 7 The lowest voltage in the system within 24 hours before and after using the strategy proposed in this paper was found when the local node adjusted the load based solely on its own information. In this case, the voltage exceeded the limit at 4 PM and 5 PM. However, under the collaborative control strategy proposed in this paper, the voltage exceeded the limit no longer occurred. Figure 8 The system's 24-hour load curve is similar to the voltage curve. The system also experienced load overruns between 4 PM and 5 PM, which were avoided under the strategy proposed in this paper. These results demonstrate that the proposed strategy can effectively improve system operation and ensure the safe operation of the distribution network.

[0143] This section compares the results of the proposed Newton-ADMM algorithm with those of the centralized algorithm and the ADMM algorithm to verify the accuracy and convergence ability of the Newton-ADMM algorithm.

[0144] The results obtained under the centralized algorithm are set as the standard results to verify the accuracy of the proposed algorithm. Figure 9 , Figure 10 The figures show the distribution of voltage and temperature-controlled load at each node of the 15-hour distribution network a-phase system. The results of the Newton-ADMM algorithm are represented by blue rings, while the results of the centralized algorithm are represented by red flower-shaped dots. It can be seen that the red dots are almost all located at the center of the blue rings, indicating that the results of the Newton-ADMM algorithm are very close to those of the centralized solution, and the accuracy of the algorithm meets the calculation requirements.

[0145] Further analysis of the convergence performance of the Newton-ADMM algorithm is conducted. The parameter settings remain consistent with the previous analysis, assuming the ADMM algorithm step size β = 1. Figure 11 and Figure 12The convergence process of phase a at node 96 over 15 hours is shown when different algorithms are used to solve the problem. Observations reveal that the ADMM algorithm requires nearly 40 convergence iterations to obtain a value close to the lumped-mode solution, while the Newton-ADMM algorithm achieves the same result in only 10 iterations. This demonstrates the significant improvement in convergence performance of the proposed Newton-ADMM algorithm compared to the ADMM algorithm. In practical applications, this means that collaborative control strategies can be formulated with fewer information exchanges, which is beneficial for reducing communication costs.

[0146] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A method for coordinated control of distributed resources in a distribution network based on Newton-ADMM, characterized in that, Includes the following steps: S1: Construct a global optimization model that includes global coupling constraints and local constraints of each distributed resource, and decompose the global optimization model into multiple local optimization sub-problems and a coordination problem handled by the power distribution system operator by introducing auxiliary variables corresponding to the local control variables of each distributed resource. S2: In one iteration, the local optimization subproblem is first solved in parallel by the local nodes of each distributed resource in order to update the local control variables; S3: Then, the power distribution system operator updates the auxiliary variables based on the local control variables updated by all local nodes; S4: Next, using the Hessian matrix of the objective function of the global optimization model, update the dual variable used to coordinate the local control variable and the auxiliary variable; S5: Repeat steps S2 to S4 until the convergence condition is met, and obtain the optimal control strategy for each distributed resource.

2. The method for coordinated control of distributed resources in a distribution network based on Newton-ADMM according to claim 1, characterized in that, The objective function of the global optimization model includes user electricity costs and user thermal comfort.

3. The method for coordinated control of distributed resources in a distribution network based on Newton-ADMM according to claim 1, characterized in that, The global coupling constraints of the global optimization model include linearized distribution network power flow equation constraints and distribution network power flow security constraints.

4. The method for coordinated control of distributed resources in a distribution network based on Newton-ADMM according to claim 1, characterized in that, The distributed resource is a temperature-controlled load.

5. A method for coordinated control of distributed resources in a distribution network based on Newton-ADMM according to claim 4, characterized in that, The local constraints of the temperature-controlled load are modeled based on a discretized linear thermal model that describes the relationship between indoor temperature, outdoor temperature and temperature-controlled load power.

6. The method for coordinated control of distributed resources in a distribution network based on Newton-ADMM according to claim 1, characterized in that, The local control variable is the active power sequence of each user's temperature-controlled load.

7. A method for coordinated control of distributed resources in a distribution network based on Newton-ADMM according to claim 1, characterized in that, The decomposition is achieved by using the consistency between the local control variable and the auxiliary variable as a coupling constraint.

8. A method for coordinated control of distributed resources in a distribution network based on Newton-ADMM according to claim 4, characterized in that, The update of the dual variable is based on the value of the dual variable in the previous round and a correction term consisting of the difference between the Hessian matrix and the local control variable and auxiliary variable in the current round.

9. A method for coordinated control of distributed resources in a distribution network based on Newton-ADMM according to claim 1, characterized in that, The global optimization model is constructed as a convex quadratic programming problem with a separable objective function, and the Hessian matrix is ​​a constant matrix.

10. A method for coordinated control of distributed resources in a distribution network based on Newton-ADMM according to claim 3, characterized in that, The linearized power flow equations of the distribution network are called DistFlow power flow equations.

Images