A distributed decision-making method, device, equipment and medium for energy management of distribution network

Abstract

The present invention belongs to the technical field of distribution network energy management, and specifically relates to a distributed decision-making method, device, equipment and medium for distribution network energy management. This scheme modifies and embeds the personalized second-order information of the user utility function into the ADMM, and uses the Hessian matrix of the user utility function to develop a personalized diagonal matrix for each user through a preprocessing method. This personalized diagonal matrix provides independent distribution system operators with valuable information that respects user privacy without infringing on user privacy. The personalized diagonal matrix is ​​used to adjust the update direction of the dual variables in the fast quasi-Newton ADMM, rather than the steepest descent / ascent direction in the classic ADMM. This method can improve the overall convergence performance.

Description

Technical Field

[0001] The present invention belongs to the technical field of distribution network energy management, and particularly relates to a distributed decision-making method, device, equipment and medium for distribution network energy management. Background Art

[0002] In recent years, with the continuous increase in the penetration rate of distributed energy in the distribution network, the operation and planning of the distribution network have faced new challenges and opportunities. In order to improve the efficiency and reliability of the distribution network, it has become particularly important to study how to utilize the flexibility of distributed energy for energy management. Distribution network energy management can generally be divided into methods such as centralized optimization and local control. In centralized optimization, power companies or distribution system operators collect all necessary network parameters and user information, and determine the optimal coordinated control strategy for distributed energy through centralized calculation. Although centralized optimization has been widely studied at the distribution network level, it faces problems such as high computational complexity, poor scalability, and user privacy leakage. Local control relies on the independent decision-making of each local agent and only regulates the operating state of distributed energy based on local information. Although local control has high operability and implementability, it is difficult to ensure the optimality of the overall system performance due to the lack of communication and coordination between distributed energy. Summary of the Invention

[0003] The purpose of the present invention is to provide a distributed decision-making method, device, equipment and medium for distribution network energy management to solve the problems existing in distribution network energy management.

[0004] To achieve the above purpose, the present invention adopts the following technical solutions:

[0005] In the first aspect of the present invention, a distributed decision-making method for distribution network energy management is provided, including:

[0006] Obtain a pre-constructed unbalanced distribution network model;

[0007] Obtain a pre-constructed user electrical equipment model;

[0008] Based on the unbalanced distribution network model and the user electrical equipment model, construct an energy management optimization model managed by an independent distribution system operator; wherein, the energy management optimization model includes an objective function for maximizing the net benefit of all users and constraint conditions corresponding to the objective function;

[0009] Calculate the second-order information of the energy management optimization model, and equivalently reconstruct the energy management optimization model according to the second-order information to obtain an equivalent representation model;

[0010] Use the fast quasi-Newton ADMM algorithm to solve the equivalent representation model to obtain a distributed decision-making scheme.

[0011] Further, in the step of obtaining the pre - constructed user electrical equipment model, the user electrical equipment model includes:

[0012] The utility function of each electrical appliance of the user, and the constraint set corresponding to the actual power consumption of each electrical appliance; wherein, the types of electrical appliances include thermostatic control load electrical appliances and electric vehicles.

[0013] Further, in the step of constructing an energy management optimization model managed by an independent distribution system operator based on the unbalanced distribution network model and the user electrical equipment model, the energy management optimization model is expressed as:

[0014]

[0015]

[0016]

[0017]

[0018]

[0019]

[0020] In the formula, represents the active power of the corresponding phase at node i; represents the active power of the constant - power device; represents the reactive power of the corresponding phase at node i; represents the reactive power of the constant - power device; represents the set of users at node i; represents the power factor; represents the voltage upper limit; represents the voltage lower limit; represents the maximum active power; represents the user, represents the set of users; represents the electrical appliance, represents the set of electrical appliances of the user, is the user for each electrical appliance of the utility function; is the set of time periods the actual power consumption; is the retail electricity price of the user, and the superscript T represents transpose; represents the marginal utility of money of the user's electrical appliance; represents the time period; represents the distribution network power flow voltage; represents the transpose of the matrix of node connection relationships, A vector representing the connection relationship between the head node and the remaining nodes, represents the reference voltage value; , respectively represent the active power flow and the reactive power flow; and are both standard incidence matrices of the distribution network; is the constraint set of the electrical appliances; is for each time period k of the common duration; represents the rated capacity of the electric vehicle; represents the node , represents the index set of all non-head nodes in the distribution network; represents the corresponding specific phase, represents the three-phase phase.

[0021] Furthermore, calculate the second-order information of the energy management optimization model, and perform an equivalent reconstruction on the energy management optimization model according to the second-order information to obtain an equivalent representation model, including:

[0022] For each user , use to represent the actual electricity consumption of all electrical appliances of the user in the time period set of the column vector, and use to represent the constraint of the column vector :

[0023]

[0024] Use and to represent the total utility function and the cost function of the user respectively;

[0025]

[0026]

[0027] Define the parameters and as follows respectively:

[0028]

[0029]

[0030] Represent the energy management optimization model as the following first intermediate model:

[0031]

[0032]

[0033]

[0034] Among them, represents the voltage column vector of all users at time period k, represents the power column vector of all users at time period k;

[0035] Let and be respectively:

[0036]

[0037]

[0038] the Hessian matrix of is a block diagonal matrix, expressed as:

[0039]

[0040] the Hessian matrix of is a block diagonal matrix, expressed as:

[0041]

[0042] Introduce an auxiliary variable , make Let the parameter be:

[0043]

[0044] Among them, represents the voltage column vector of all users at time period after introducing the auxiliary variable, represents the power column vector of all users at time period after introducing the auxiliary variable;

[0045] Let the parameter represent the indicator function of the closed convex set :

[0046]

[0047] Reconstruct the first intermediate model equivalently into an equivalent representation model:

[0048]

[0049]

[0050] Among them, x is the control variable and z is the auxiliary variable.

[0051] Furthermore, the equivalent representation model is solved by using the fast quasi-Newton ADMM algorithm, including:

[0052] For the customer introduce a positive definite matrix , and define a diagonal positive definite matrix ;

[0053] Define the parameter as follows:

[0054]

[0055]

[0056] Express the equivalent representation model as the following second intermediate model:

[0057]

[0058]

[0059] Among them, 、 represent the transformed x and z respectively;

[0060] Determine the augmented Lagrangian function of the second intermediate model as:

[0061]

[0062] Define the parameter and :

[0063]

[0064]

[0065]

[0066] Among them, ;

[0067] The steps to solve the augmented Lagrangian function are as follows:

[0068] Iteration number = 0, each user sends its own positive definite matrix to the corresponding independent distribution system operator; ≥ 0, iteratively update x 、 z and Until convergence.

[0069] Furthermore, update iteratively in sequence x , z and until convergence, including:

[0070] S1. By the user update x :

[0071]

[0072] S2. By the independent distribution system operator update z :

[0073]

[0074] S3. Update :

[0075]

[0076] S4. Update the iteration count .

[0077] Furthermore, for the customer introduce a positive definite matrix , in the step of defining let the Hessian matrix H be a symmetric positive definite real matrix, and solve the following semi-definite convex optimization to calculate the diagonal positive definite matrix F that minimizes the condition number t:

[0078]

[0079]

[0080]

[0081] is a diagonal matrix;

[0082] wherein, .

[0083] In the second aspect of the present invention, a distributed decision-making device for power distribution network energy management is provided, including:

[0084] A first acquisition module for acquiring a pre-constructed unbalanced power distribution network model;

[0085] A second acquisition module for acquiring a pre-constructed user electrical equipment model;

[0086] A model construction module, configured to construct an energy management optimization model managed by an independent distribution system operator based on the unbalanced distribution network model and the user electrical equipment model; wherein, the energy management optimization model includes an objective function for maximizing the net benefit of all users and constraint conditions corresponding to the objective function;

[0087] A model conversion module, configured to calculate the second-order information of the energy management optimization model, and perform equivalent reconstruction on the energy management optimization model according to the second-order information to obtain an equivalent representation model;

[0088] A model solving module, configured to solve the equivalent representation model by using a fast quasi-Newton ADMM algorithm to obtain a distributed decision-making scheme.

[0089] In a third aspect of the present invention, an electronic device is provided, including a processor and a memory, and the processor is configured to execute a computer program stored in the memory to implement the distributed decision-making method for distribution network energy management as described above.

[0090] In a fourth aspect of the present invention, a computer-readable storage medium is provided, and the computer-readable storage medium stores at least one instruction, and when the at least one instruction is executed by a processor, the distributed decision-making method for distribution network energy management as described above is implemented.

[0091] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0092] The distributed decision-making method for distribution network energy management of the present invention is a distributed optimization method of fast quasi-Newton ADMM. In this solution, the personalized second-order information of the utility function is modified and embedded in ADMM, and the Hessian matrix of the customer utility function is used to establish a personalized diagonal matrix for each customer, so as to improve the convergence performance of fast quasi-Newton ADMM in a personalized distributed decision-making manner without disclosing private customer information, and provide insensitive but valuable information for independent distribution system operators.

[0093] This solution uses a personalized diagonal matrix to adjust the update direction of the dual variable in fast quasi-Newton ADMM, rather than the steepest gradient descent / ascent direction in traditional ADMM, to improve the overall convergence performance.

[0094] A distributed decision-making device for distribution network energy management, an electronic device and a computer-readable storage medium provided by the present invention also solve the problems proposed in the background art part. Description of the Drawings

[0095] The specification drawings constituting a part of this application are used to provide a further understanding of the present invention. The schematic embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation to the present invention. In the drawings:

[0096] Figure 1 It is a flowchart of a distributed decision-making method for distribution network energy management according to an embodiment of the present invention;

[0097] Figure 2 It is a schematic diagram of a distributed decision-making method for distribution network energy management according to an embodiment of the present invention;

[0098] Figure 3 It is a schematic diagram of the electricity price and ambient temperature changes within 24 hours in an embodiment of the present invention;

[0099] Figure 4 It is a schematic diagram of the total power demand of the IEEE 123-node unbalanced radial distribution network within 24 hours in the prior art;

[0100] Figure 5 It is a schematic diagram of the minimum bus voltage value within 24 hours in the improved IEEE 123-node unbalanced radial distribution network in the prior art;

[0101] Figure 6 It is a schematic diagram of the total power demand of the improved IEEE 123-node unbalanced radial distribution network within 24 hours in an embodiment of the present invention;

[0102] Figure 7 It is a schematic diagram of the minimum bus voltage value within 24 hours in the improved IEEE 123-node unbalanced radial distribution network in an embodiment of the present invention;

[0103] Figure 8 It is a schematic diagram of the total load of air conditioners and electric vehicles within 24 hours in an embodiment of the present invention;

[0104] Figure 9 It is a schematic diagram of the comparison of air conditioner results in an embodiment of the present invention; among them, (a) the a-phase power of node 1 at the 15th hour, (b) the a-phase power of node 47 at the 15th hour;

[0105] Figure 10 It is a schematic diagram of the comparison of electric vehicle results in an embodiment of the present invention; among them, (a) the c-phase power of node 31 at the 15th hour, (b) the a-phase power of node 35 at the 15th hour;

[0106] Figure 11 It is a schematic diagram of the comparison of the a-phase voltage results of node 114 at the 15th hour in an embodiment of the present invention;

[0107] Figure 12 It is a structural block diagram of a distributed decision-making device for distribution network energy management according to an embodiment of the present invention;

[0108] Figure 13 It is a structural block diagram of an electronic device according to an embodiment of the present invention. Detailed implementation manners

[0109] The present invention will be described in detail below with reference to the accompanying drawings and in conjunction with embodiments. It should be noted that, without conflict, the embodiments in the present application and the features in the embodiments can be combined with each other.

[0110] The following detailed descriptions are all exemplary descriptions, aiming to provide further detailed descriptions of the present invention. Unless otherwise specified, all technical terms adopted by the present invention have the same meaning as commonly understood by those of ordinary skill in the technical field to which the present application belongs. The terms used in the present invention are only for describing specific embodiments, and are not intended to limit the exemplary embodiments according to the present invention.

[0111] Embodiment 1

[0112] As Figure 1 shown, a distributed decision-making method for distribution network energy management includes steps S100 to S500, specifically as follows:

[0113] S100. Obtain a pre-constructed unbalanced distribution network model.

[0114] Step S100 specifically includes:

[0115] Given a radial distribution network with N + 1 nodes, the distribution network includes unbalanced three phases. Let be the node index set, where 0 is the head node index, is the index set of all non-head nodes. The distribution network consists of N feeders, and each feeder connects adjacent node pairs. Each feeder can be single-phase, two-phase, or three-phase. For each node , let represent the node directly connected to node j before node j along the radial network. Let represent the set of all nodes strictly located after node along the radial network.

[0116] Let represent the set of all different feeders in the distribution network. is the time period set, that is, the energy management time set.

[0117] For any time period and each feeder , the unbalanced linearized power flow model of the distribution network can be expressed as:

[0118] (1a)

[0119] (1b)

[0120] (1c)

[0121] Where: kRepresents a time period; Represents k The three - phase active power of the time - period feeder ; Represents k The time - period feeder The three - phase reactive power; Represents k The magnitude of the square of the three - phase voltage at node j in the time period; Represents k The active power at node j in the time period; Represents k The reactive power at node j in the time period; ; ; ; And ; Represents the corresponding phase, , Represents the three - phase phase. 、 Respectively represent k The active and reactive powers between nodes j and m in the time period, Represents k The voltage of node i in the time period; 、 Respectively represent the resistance and reactance between nodes i and j, Is a column vector representing the phase relationship between the three phases. Is The per - unit value matrix of the three - phase resistance of Is The per - unit value matrix of the three - phase resistance of , Is The conjugate transpose of , , Is the multiplication operator. All terms are represented in per - unit values According to the three - phase phase For sorting.

[0122] The unbalanced linearized power - flow model of the distribution network can be written in matrix - vector form:

[0123] , (2)

[0124] In the formula, 、 And Are all standard incidence matrices of the distribution network; , , ; Represents a time period;​ represents the power flow voltage of the distribution network in a time period the transpose of the matrix representing the node connection relationship the vector representing the connection relationship between the head node and the remaining nodes represents k the reference voltage value in a time period 、 respectively represent k the active power flow and reactive power flow in a time period and are both block diagonal matrices represents b p (N) the resistance between node N and represents b p (N) the reactance between node N and b p (N) is the node before node N and directly connected to node N

[0125] S200. Obtain the pre - constructed user electrical equipment model

[0126] In step S200, the user electrical equipment model includes: the utility function of each electrical appliance of the user, and the constraint set corresponding to the actual power consumption of each electrical appliance; where the types of electrical appliances include thermostatic control load electrical appliances and electric vehicles

[0127] Specifically, let be the common name of the user be the preference and structural characteristics be the phase j be the node where the user is located. The user is connected to node through a single - phase line with phase as the location of . The set of electrical appliances operated by each user is . For each electrical appliance of user , let and respectively represent the actual power consumption and reactive power consumption in the time period set , and K represents the total number of time periods in the time period set. Assume that the electrical appliance operates under a constant power factor model:

[0128] (3)​

[0129] Among them, represents a time period, represents the actual power consumption of an electrical appliance , represents the power factor;

[0130] Each electrical appliance has the following characteristics:

[0131] 1) The utility function of each electrical appliance of the user is , and the utility function represents the degree of satisfaction obtained by the user and is a function of its actual power consumption.

[0132] 2) The constraint set for the electrical appliance . This solution involves two types of electrical appliances.

[0133] (1) The first type includes the thermostatic control load electrical appliances (i.e., air conditioners) of the customer with respect to the ambient temperature.

[0134] This solution uses to represent the set of thermostatic control load electrical appliances (air conditioners) of the customer . For each electrical appliance , is used to represent the initial room temperature of the user in the time period set , represents the room temperature of the user at time period, ℉ represents degrees Fahrenheit, represents the ambient temperature during the time period . During the time period , the user utility is defined as the deviation between the maximum thermal comfort that the user can achieve and the user discomfort. The user discomfort is represented by the difference between the actual room temperature and the room temperature at which the user achieves the maximum thermal comfort.

[0135] The user utility can be expressed as follows:

[0136] (4)

[0137] Among them, is a conversion factor, which can be written as:

[0138] (5)

[0139] Among them, K represents the total number of time periods in the time period set ; represents the time period set of the room temperature; represents the time period set under which the user obtains the room temperature with the maximum thermal comfort.

[0140] In this solution, a linearized thermal model is adopted to simulate the indoor temperature dynamics:

[0141] (6)

[0142] In the formula, and are positive values, representing the thermal inertia coefficient and the power conversion coefficient respectively; represents the rated power of the air conditioner equipment; represents each time period k of the common duration;

[0143] (7)

[0144] Specific expressions for each item:

[0145]

[0146]

[0147]

[0148] Among them, represents the rated power of the air conditioner equipment; K represents the total number of time periods in the time period set ; T represents the transpose. 、 、 and are all the compact forms corresponding to the relevant variables after the formula (6) is converted into a compact form, and have no practical meaning.

[0149] Substitute the formula (6) into the formula (5), through of the following properties, this solution can use to represent the utility function of the constant temperature control load electrical appliance:

[0150] (8)

[0151] Among them, is of the transpose;

[0152] In addition, for each electrical appliance , Constraint set of the constant temperature control load electrical appliance is expressed as follows:

[0153] (9)

[0154] Among them, k represents the time period, 、 are respectively the lower limit and the upper limit of the actual power consumption of the electrical appliance .

[0155] (2) The second category: Electric vehicles. This solution uses to represent the set of electric vehicles of the user .

[0156] For each vehicle , use to represent the power state of the user's electric vehicle at the start of the time period set , represents the power state of the user's electric vehicle at the end of the stage, and represents the desired power state at the end of the time period set , that is, the target value of the power state of the vehicle under the time period set .

[0157] This solution uses a quadratic concave function as the utility function of the electric vehicle to represent the total satisfaction obtained under the user's random charging scheme , as shown below:

[0158] (10)

[0159] and are the utility function parameters of the electric vehicle, is the Hessian matrix of formula (10) with respect to , is the utility function parameter of the electric vehicle; under the condition of satisfying the charging power limit, the charging amount of the user's electric vehicle should be equal to the at the end of the time period set , and the constraint can be expressed as:

[0160] (11a)

[0161] (11b)

[0162] Among them, is the capacity of the electric vehicle, and each time period kCommon duration Measured in hours, is the charging efficiency of the user's electric vehicle; represents the rated capacity of the electric vehicle. Additionally, for each vehicle , can be expressed as:

[0163] (12)

[0164] S300. Based on the unbalanced distribution network model and the user electrical equipment model, construct an energy management optimization model managed by an independent distribution system operator; wherein, the energy management optimization model includes an objective function for maximizing the net benefit of all users and constraint conditions corresponding to the objective function.

[0165] Specifically, the goal of the independent distribution system operator is to maximize the social welfare of all customers by flexibly coordinating the electricity usage of users under the constraints of the distribution network and local users. The electrical appliance set of each user in this solution includes two electrical appliance subsets: ;

[0166] Then, in step S300, the energy management optimization model managed by the independent distribution system operator is expressed as:

[0167] (13a)

[0168] (13b)

[0169] (13c)

[0170] (13d)

[0171] (13e)

[0172] (13f)

[0173] In the formula, represents the active power of the corresponding phase at node i; represents the active power of the constant power device; represents the reactive power of the corresponding phase at node i; represents the reactive power of the constant power device; represents the set of users at node i; represents the power factor; represents the voltage upper limit; represents the voltage lower limit; Represents the maximum active power; represents the user, Denote the user set; Denote the electrical appliance, Denote the set of electrical appliances of the user, For the user of each electrical appliance the utility function; Denote the set of time periods the actual electricity consumption. Denote the time period; Denote the power flow voltage of the distribution network; Denote the transpose of the matrix representing the node connection relationship, Denote the vector representing the connection relationship between the head node and the remaining nodes, Denote the reference voltage value; , respectively denote the active power flow and the reactive power flow; and are both the standard incidence matrices of the distribution network; is the constraint set of the electrical appliance; For each time period the common duration; Denote the rated capacity of the electric vehicle; Denote the node , Denote the index set of all non-head nodes in the distribution network; Denote the corresponding specific phase, Denote the three-phase phase.

[0174] (13a) represents the maximum net benefit of all users, (yuan / kWh) is the retail electricity price of the user, and the superscript T represents the transpose; represents the marginal utility of money of the user's electrical appliance (defined as the benefit lost per additional yuan of electricity cost). (13b)-(13e) are the power flow constraints, and (13f) is the electrical appliance device constraint of the user.

[0175] S400. Calculate the second-order information of the energy management optimization model, and perform an equivalent reconstruction on the energy management optimization model according to the second-order information to obtain an equivalent representation model.

[0176] Step S400 specifically includes:

[0177] 1) Based on the energy management of the independent distribution system operator.

[0178] For each user , use to represent the actual electricity consumption of all electrical appliances of the user in the set of time periods as a column vector, and use to represent the column vector Constraints: ; Use and to represent the total utility function and cost function of the user respectively; ; ;

[0179] Define the parameters and as: ; ;

[0180] It can be seen from (13a), (13b), and (13c) that , is an affine function of, and represent the energy management optimization model as the following first intermediate model:

[0181] (14a)

[0182] (14b)

[0183] (14c)

[0184] where represents the voltage column vector of all users during the time period, and represents the power column vector of all users

[0185] Let and be respectively:

[0186] (15a)

[0187] (15b)

[0188] The Hessian matrix is a block diagonal matrix, expressed as:

[0189] (16a)

[0190] The Hessian matrix is a block diagonal matrix, expressed as:

[0191] (16b)

[0192] Introduce an auxiliary variable , so that This scheme sets the parameter is:

[0193] (17)

[0194] wherein, represents the column vector of voltages of all users at time period k after introducing the auxiliary variable, represents the column vector of powers of all users at time period k after introducing the auxiliary variable.

[0195] Let the parameter represent the indicator function of the closed convex set : : ;

[0196] Reconstruct the first intermediate model equivalently into an equivalent representation model:

[0197] (18a)

[0198] (18b)

[0199] wherein, x is the control variable and z is the auxiliary variable.

[0200] Here, and exhibit the following properties:

[0201] a. is a strongly convex function on , and the second-order partial derivative of x , i.e., the Hessian matrix, satisfies , and is a convex set; H is the Hessian matrix of the objective function;

[0202] b. is Lipschitz continuous on with a constant , and there is: ; represents any two values in;

[0203] c. is a closed proper convex function.

[0204] S500. Solve the equivalent representation model by using the fast quasi-Newton ADMM algorithm to obtain a distributed decision-making scheme.

[0205] In some other embodiments, first determine the augmented Lagrangian function of Equation (18) as:

[0206] (19)

[0207] Then, the classical ADMM shown in Algorithm 1 is applied to solve Problem (18) in a distributed manner as shown in Table 1 below:

[0208] Table 1 Solving Problem (18) by the classical ADMM in a distributed manner

[0209]

[0210] As shown in Algorithm 1, for the users in the update , the independent distribution system operator does not need to know the private information of and .

[0211] The convergence rate of Algorithm 1 depends on the condition number of the function . The condition number of can be expressed as: ; where are the maximum and minimum eigenvalues of . Due to the diversity and different characteristics and properties of user appliances, the condition number of

[0212] may be very large, resulting in a very slow convergence rate of Algorithm 1.

[0213] Specifically, step S500 includes the following steps:

[0214] This solution first introduces a personalized positive definite matrix for the customer , and defines ;

[0215] Define the parameter as follows:

[0216] (20a)

[0217] (20b)

[0218] Express the equivalent representation model as the following second intermediate model:

[0219] (21a)

[0220] (21b)

[0221] where , represent the transformed x and z respectively.

[0222] Determine the augmented Lagrangian function of the second intermediate model as follows:

[0223] (22)

[0224] Next, define the parameters , and as follows:

[0225] (23a)

[0226] (23b)

[0227] (23c)

[0228] where: , β represents the Lagrange multiplier.

[0229] As Figure 2 shown, solve according to Algorithm 2 designed in this solution, as shown in Table 2 below:

[0230] Table 2 Fast Quasi-Newton ADMM Solving Problem (18)

[0231]

[0232] It should be noted that before the implementation of Algorithm 2, each customer should send to the corresponding independent distribution system operator.

[0233] The differences between Algorithm 1 and Algorithm 2 are shown in Table 3. It can be observed from this that the updates of x, z, and λ in the fast quasi-Newton ADMM are different from those in the classical ADMM. In the fast quasi-Newton ADMM, the update direction of the dual variable λ no longer depends on the steepest ascent / descent direction as in the classical ADMM. Instead, in the fast quasi-Newton ADMM, M is used to update the dual variable in a Newton-like manner, and a fast convergence rate can be expected.

[0234] Table 3 Comparison between ADMM and Fast Quasi-Newton ADMM

[0235]

[0236] In the preferred embodiment, since F is a diagonal positive definite matrix, therefore, in this solution, the Hessian matrix H is set as a symmetric positive definite real matrix, and then the diagonal positive definite matrix F that minimizes the condition number t can be obtained by solving the following semi-definite convex optimization:

[0237] (24a)

[0238] (24b)

[0239] (24c)

[0240] is a diagonal matrix;

[0241] wherein, H is the Hessian matrix of the objective function; .

[0242] Simulation example

[0243] To further prove the effectiveness of the proposed solution, a specific simulation example is given below.

[0244] The independent distribution system operator is responsible for improving the operation of the IEEE 123-node unbalanced radial distribution network. Each test simulates one day, and one day is divided into 24 hours, that is , = 1 h. Three different values of are set in this solution: , and three different values of are assigned to different customers. The goal of the independent distribution system operator is to maximize the social benefits of all users subject to the distribution network constraints and satisfy the distribution network constraints. There are 345 users in the distribution network, and each user consists of two types of flexible devices, including air conditioners and electric vehicles.

[0245] The remaining parameters of air conditioners and electric vehicles are shown in Table 4 below:

[0246] Table 4 Remaining parameters of air conditioners and electric vehicles

[0247]

[0248] The data of electricity price and ambient temperature within one day are as Figure 3 shown. All simulations in this simulation example are carried out in MATLAB R2019b and integrated with the YALMIP toolbox and the IBM ILOG CPLEX 12.9 solver.

[0249] 1) Simulation results

[0250] When the fast quasi-Newton ADMM proposed in this solution is not used, each user independently determines its electricity consumption for one day, that is, each user directly determines the electricity demand by solving without considering the distribution network constraints. The simulation results are as Figure 4 and Figure 5 shown. At this time, the distribution network has situations where the demand peak exceeds the standard and the voltage limit exceeds the standard.

[0251] However, when using the fast quasi - Newton ADMM proposed in this solution to manage the distribution network, the simulation results are as follows Figure 6 , Figure 7 . Through the method proposed in the present invention, no peak value and voltage over - limit problems occur within 24 hours. Therefore, the method proposed in the present invention successfully and efficiently solves the problems of distribution network operation.

[0252] The total load of air conditioners and electric vehicles within 24 hours determined based on this solution is as follows Figure 8 . As can be seen from Figure 8 , there is no electric vehicle and air conditioner load during the periods of 5 - 12h and 19 - 23h. As is known from Figure 3 , the retail prices around 8h and 19h are higher than those in other periods, which may lead to users being reluctant to use air conditioners and electric vehicle loads during the periods of 5 - 12h and 19 - 23h. In addition, the cumulative air conditioner load around the 15th hour is higher than that in other periods. As shown in Figure 3 , this is because the environmental temperature is very high (much higher than ), and the electricity price is relatively low around the 15th hour. In addition, the total electric vehicle load at the 24th hour is about , while the total air conditioner load at the 24th hour is 0. This is because the environmental temperature is relatively low after 20h, and users do not need to use air conditioners to cool the indoor environment. However, the charging behavior of electric vehicles is not affected by the environmental temperature, and the electricity price at the 24th hour is relatively low, which is conducive to electric vehicles charging at the 24th hour.

[0253] In addition, this solution also compares the fast quasi - Newton ADMM with the classical ADMM and the centralized optimization algorithm. Taking the 15th hour as an example, the voltage results of some air conditioners and electric vehicles are compared as shown in Figures 9 - 11 .

[0254] It can be obtained from Figures 9 - 11 that the convergence speed of the fast quasi - Newton ADMM is always significantly faster than that of the ADMM. When using the classical ADMM, the results of air conditioners, electric vehicles, and voltage converge after about 100 iterations. If the fast quasi - Newton ADMM proposed in this solution is used, it can converge after about 40 iterations. The method proposed in the present invention utilizes the second - order Hessian matrix information of users to accelerate the convergence speed of the algorithm, and updates the dual variables in a quasi - Newton manner, thereby improving the efficiency of energy management calculation and communication while ensuring the reliability and safety of the distribution network.

[0255] Embodiment 2

[0256] As shown in Figure 12 , based on the same inventive concept as the above - mentioned embodiment, the present invention also provides a distributed decision - making device for distribution network energy management, which is characterized in that it includes:

[0257] A first acquisition module, configured to acquire a pre-constructed unbalanced distribution network model;

[0258] A second acquisition module, configured to acquire a pre-constructed user electrical equipment model;

[0259] A model construction module, configured to construct an energy management optimization model managed by an independent distribution system operator based on the unbalanced distribution network model and the user electrical equipment model; wherein, the energy management optimization model includes an objective function for maximizing the net benefit of all users, and constraint conditions corresponding to the objective function;

[0260] A model conversion module, configured to calculate the second-order information of the energy management optimization model, and perform equivalent reconstruction on the energy management optimization model according to the second-order information to obtain an equivalent representation model;

[0261] A model solving module, configured to solve the equivalent representation model by using a fast quasi-Newton ADMM algorithm to obtain a distributed decision-making scheme.

[0262] Embodiment 3

[0263] As Figure 13 shown, the present invention further provides an electronic device 100 for implementing a distributed decision-making method for distribution network energy management; the electronic device 100 includes a memory 101, at least one processor 102, a computer program 103 stored in the memory 101 and executable on at least one processor 102, and at least one communication bus 104.

[0264] The memory 101 can be used to store the computer program 103. The processor 102 realizes the steps of a distributed decision-making method for distribution network energy management in Embodiment 1 by running or executing the computer program stored in the memory 101 and calling the data stored in the memory 101.

[0265] The memory 101 mainly includes a program storage area and a data storage area. Among them, the program storage area can store an operating system, application programs required for at least one function (such as a sound playback function, an image playback function, etc.); the data storage area can store data created according to the use of the electronic device 100 (such as audio data, etc.). In addition, the memory 101 can include non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, at least one magnetic disk storage device, a flash device, or other non-volatile solid-state storage devices.

[0266] At least one processor 102 may be a Central Processing Unit (CPU), or may also be other general-purpose processors, Digital Signal Processors (DSPs), Application Specific Integrated Circuits (ASICs), Field-Programmable Gate Arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The processor 102 may be a microprocessor or the processor 102 may also be any conventional processor, etc. The processor 102 is the control center of the electronic device 100 and connects various parts of the entire electronic device 100 through various interfaces and lines.

[0267] The memory 101 in the electronic device 100 stores multiple instructions to implement a distributed decision-making method for distribution network energy management. The processor 102 can execute the multiple instructions to implement:

[0268] Obtain a pre-constructed unbalanced distribution network model;

[0269] Obtain a pre-constructed user electrical equipment model;

[0270] Based on the unbalanced distribution network model and the user electrical equipment model, construct an energy management optimization model managed by an independent distribution system operator; wherein, the energy management optimization model includes an objective function for maximizing the net benefit of all users and constraint conditions corresponding to the objective function;

[0271] Calculate the second-order information of the energy management optimization model, and perform equivalent reconstruction on the energy management optimization model according to the second-order information to obtain an equivalent representation model;

[0272] Use the fast quasi-Newton ADMM algorithm to solve the equivalent representation model to obtain a distributed decision-making solution.

[0273] Embodiment 4

[0274] If the integrated module / unit of the electronic device 100 is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on such an understanding, to implement all or part of the processes in the above-described embodiment methods of the present invention, it can also be completed by a computer program instructing relevant hardware. The computer program can be stored in a computer-readable storage medium. When the computer program is executed by a processor, the steps of the above-described various method embodiments can be implemented. Among them, the computer program includes computer program code, and the computer program code can be in the form of source code, object code, executable file, or some intermediate form, etc. The computer-readable medium can include: any entity or device capable of carrying the computer program code, recording medium, USB flash drive, mobile hard disk, magnetic disk, optical disc, computer memory, and read-only memory (ROM, Read-Only Memory).

[0275] Those skilled in the art should understand that the embodiments of the present invention can be provided as a method, a system, or a computer program product. Therefore, the present invention can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0276] The present invention is described with reference to the flowcharts and / or block diagrams of methods, apparatuses (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram can be implemented by computer program instructions, and the combination of processes and / or blocks in the flowchart and / or block diagram can also be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing devices generate a device for implementing the functions specified in Figure 1 one process or multiple processes and / or blocks Figure 1 one block or multiple blocks.

[0277] These computer program instructions can also be stored in a computer-readable memory capable of guiding a computer or other programmable data processing devices to work in a specific manner, so that the instructions stored in the computer-readable memory generate a manufactured article including an instruction device, and the instruction device implements the functions specified in Figure 1 one process or multiple processes and / or blocks Figure 1 one block or multiple blocks.

[0278] These computer program instructions can also be loaded onto a computer or other programmable data processing device, so that a series of operation steps are executed on the computer or other programmable device to generate a computer-implemented process, and thus the instructions executed on the computer or other programmable device provide steps for implementing the functions specified in one process Figure 1 one process or multiple processes and / or blocks Figure 1 steps of the functions specified in one block or multiple blocks.

[0279] In the description of this specification, the descriptions referring to terms such as "one embodiment", "example", "specific example", etc. mean that the specific features, structures, materials or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in a suitable manner in any one or more embodiments or examples.

[0280] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to the above embodiments, those of ordinary skill in the art should understand that: modifications or equivalent replacements can still be made to the specific implementation manners of the present invention, and any modifications or equivalent replacements that do not depart from the spirit and scope of the present invention should be covered by the protection scope of the claims of the present invention.

Claims

1. A distributed decision-making method for energy management in a distribution network, characterized in that: include: Get pre-built unbalanced distribution network models; Get pre-built consumer appliance models; Based on the unbalanced distribution network model and the user electrical equipment model, an energy management optimization model managed by an independent distribution system operator is constructed; wherein the energy management optimization model includes an objective function for maximizing the net benefit of all users and constraints corresponding to the objective function, and the energy management optimization model is expressed as: In the formula, represents the active power of the corresponding phase at node i; Indicates that the constant power equipment is active; represents the reactive power of the corresponding phase at node i; Indicates that the constant power equipment is reactive; represents the set of users at node i; Indicates power factor; Indicates the upper voltage limit; Indicates the voltage lower limit; Maximum active power indication; Represents the user, Represents a collection of users; Indicates electrical appliances, Represents the user's electrical appliance collection, For users Each appliance The utility function of Set for time period Actual electricity consumption; is the retail electricity price for the user, and the superscript T indicates transposition; Represents the marginal monetary utility of the user's electrical appliances; Indicates time period; Indicates the power flow voltage of the distribution network; The transpose of the matrix representing the node connectivity, A vector representing the connection relationship between the first node and the remaining nodes, Indicates the reference voltage value; , Respectively represent active power flow and reactive power flow; and All are standard correlation matrices for distribution networks; is the constraint set of the appliance; For each period k the common duration of Indicates the rated capacity of the electric vehicle; Representation Node , Represents the index set of all non-head nodes in the distribution network; Indicates the corresponding specific phase, Indicates the three-phase phase; Calculating the second-order information of the energy management optimization model, and performing equivalent reconstruction on the energy management optimization model according to the second-order information to obtain an equivalent representation model, including: For each user ,use Indicates that all the user's electrical appliances are collected in the time period Actual power consumption Column vector of Represents a column vector Constraints: use and Representing users Total utility and cost functions; Defining parameters and They are: The energy management optimization model is represented as the following first intermediate model: in, represents the voltage column vector of all users in time period k, represents the power column vector of all users in time period k; set up and They are: The Hessian matrix is a block diagonal matrix, expressed as: The Hessian matrix is a block diagonal matrix, expressed as: Introducing auxiliary variables ,make , set the parameter for: in, Indicates that after the introduction of auxiliary variables, all users The voltage column vector of the time period, Indicates that after the introduction of auxiliary variables, all users The power column vector of the time period; Set parameters Represents a closed convex set The indicator function : Equivalently reconstruct the first intermediate model into an equivalent representation model: Among them, x is the control variable and z is the auxiliary variable; The fast quasi-Newton ADMM algorithm is used to solve the equivalent representation model to obtain a distributed decision-making solution, including: For customers Introducing positive definite matrices , define the diagonal positive definite matrix ; Defining parameters as follows: The equivalent representation model is expressed as the following second intermediate model: in, , Represent the transformed x and z respectively; The augmented Lagrangian function of the second intermediate model is determined as: Defining parameters as well as : in, ; The steps for solving the augmented Lagrangian function are as follows: Iterations =0, each user The positive definite matrices Sent to the corresponding independent distribution system operator; ≥0, update iteratively in sequence x , z and Until convergence.

2. The distributed decision-making method for distribution network energy management according to claim 1, characterized in that: In the step of obtaining a pre-built user electrical device model, the user electrical device model includes: The utility function of each electrical appliance of the user and the constraint set corresponding to the actual power consumption of each electrical appliance; wherein the types of electrical appliances include thermostatically controlled load appliances and electric vehicles.

3. The distributed decision-making method for distribution network energy management according to claim 1, characterized in that: Update iteratively x , z and Until convergence, including: S1. By the user renew x : S2. Update by independent distribution system operator z : S3, Update : S4. Update the number of iterations .

4. The distributed decision-making method for distribution network energy management according to claim 3, characterized in that: For customers Introducing positive definite matrices ,definition In the steps of H For a symmetric positive definite real matrix, the diagonal positive definite matrix F that minimizes the condition number t is calculated by solving the following semi-positive definite convex optimization: is a diagonal matrix; in, .

5. A distributed decision-making device for energy management of a distribution network, used to implement the distributed decision-making method for energy management of a distribution network according to claim 1, characterized in that: include: A first acquisition module is used to acquire a pre-built unbalanced distribution network model; A second acquisition module is used to acquire a pre-built user electrical equipment model; A model building module, used to build an energy management optimization model managed by an independent distribution system operator based on the unbalanced distribution network model and the user electrical equipment model; wherein the energy management optimization model includes an objective function for maximizing the net benefit of all users and constraints corresponding to the objective function; A model conversion module, used to calculate the second-order information of the energy management optimization model, and perform equivalent reconstruction on the energy management optimization model according to the second-order information to obtain an equivalent representation model; The model solving module is used to solve the equivalent representation model using a fast quasi-Newton ADMM algorithm to obtain a distributed decision-making solution.

6. An electronic device, characterized in that: It comprises a processor and a memory, wherein the processor is used to execute a computer program stored in the memory to implement the distributed decision-making method for distribution network energy management as claimed in any one of claims 1 to 4.

7. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores at least one instruction, and when the at least one instruction is executed by the processor, the distributed decision-making method for distribution network energy management according to any one of claims 1 to 4 is implemented.

Images