Power distribution network control optimization method based on resource regulation and control

Abstract

The invention discloses a power distribution network control optimization method based on resource regulation and control, and the method comprises the steps: carrying out the measurement of uncertain factors of a supply side and a demand side of a power distribution network, and constructing a first constraint condition; secondly, constructing a second constraint condition for a three-phase imbalance condition of the power distribution network; constructing a target function by taking the minimum annual comprehensive operation cost of the power distribution network as a target, and constructing a resource regulation and control optimization model according to the target function and a second constraint condition; and carrying out linear conversion on the resource regulation and control optimization model, solving the resource regulation and control optimization model according to the first constraint condition, obtaining a power distribution network regulation and control strategy, and carrying out regulation and control on the power distribution network. Modeling is carried out according to the characteristic of three-phase imbalance of the power distribution network, the uncertainty of the model is described, a more accurate regulation and control strategy capable of reducing the comprehensive operation cost and the voltage out-of-limit risk of the power distribution network is obtained, and the operation benefit and the operation stability of the power distribution network are improved.

Description

Technical Field

[0001] The present application belongs to the field of distribution network resource planning, and specifically relates to a distribution network control optimization method based on resource regulation. Background Art

[0002] In order to ease environmental pressure and achieve green development, distributed power sources represented by wind power and photovoltaic power are widely connected to the distribution network. However, the increase in the penetration rate of distributed power sources has changed the flow direction of the distribution network from the traditional one-way flow to the two-way flow, resulting in voltage over-limit and increased peak-to-valley difference in the system. Therefore, changing the control means of the traditional distribution network and realizing high-precision real-time operation optimization of the distribution network is the construction task of the new power system.

[0003] Existing literature has proposed to rely on control measures such as SVC, SOP, ESS to improve system power distribution and prevent voltage over-limit, but most of them only consider the influence of SOP, ESS or SVC single elements or both, without fine-grained modeling for the characteristics of three-phase imbalance in the distribution network, and do not fully utilize the ability of control resources to improve voltage imbalance. Therefore, the proposed control measures cannot effectively solve the problem of voltage over-limit caused by three-phase imbalance in the distribution network in actual situations. Summary of the invention

[0004] This application proposes a distribution network control optimization method based on resource regulation, models the characteristics of three-phase imbalance in the distribution network, and uses Wasserstein distance to describe the uncertainty of the model, so as to obtain a more accurate regulation strategy that can reduce the comprehensive operating cost and voltage over-limit risk of the distribution network, thereby improving the operating efficiency and stability of the distribution network.

[0005] A first aspect of the present application provides a distribution network control optimization method based on resource regulation, the method comprising:

[0006] Based on the historical data of the distribution network, the Wasserstein distance algorithm is used to calculate the uncertainty factors on the supply and demand sides of the distribution network to construct the first constraint condition;

[0007] According to the historical data of the distribution network, a second constraint condition is constructed for the three-phase imbalance of the distribution network and the operation status of the electrical equipment;

[0008] An objective function is constructed with the goal of minimizing the annual comprehensive operating cost of the distribution network, and a resource regulation optimization model is constructed based on the objective function and the second constraint condition;

[0009] The resource control optimization model is linearly transformed by a continuous linearization algorithm, and the resource control optimization model after the linear transformation is solved according to the first constraint condition to obtain a distribution network control strategy;

[0010] The distribution network resources are regulated according to the distribution network regulation strategy.

[0011] The above scheme first uses the Wasserstein distance algorithm to construct the first constraint condition to express the uncertainty of the distribution network on the demand side and the supply side, reduce the impact of the model's uncertainty factors on the solution results, and improve the accuracy of subsequent model calculations; then constructs constraints for the three-phase imbalance of the distribution network, realizes refined modeling, and constructs a resource regulation optimization model with the goal of minimizing the annual comprehensive operating cost of the distribution network; uses a continuous linearization algorithm to convert nonlinear data in the resource regulation optimization model into linear data, reduces the complexity of subsequent calculations, and then uses the first constraint condition to perform model conversion to make the uncertainty in the resource regulation optimization model measurable, and obtain an accurate distribution network regulation strategy; uses the distribution network regulation strategy to regulate the distribution network, reduces the risk of voltage exceeding the limit in the distribution network, and improves the operating efficiency and stability of the distribution network.

[0012] In a possible implementation method of the first aspect, based on the acquired historical data of the distribution network, the uncertainty factors on the supply side and the demand side of the distribution network are calculated using the Wasserstein distance algorithm to construct a first constraint condition, specifically:

[0013] According to the historical data of the distribution network, the uncertainty of the distributed energy output on the supply side is measured using the Wasserstein distance algorithm to obtain the supply side uncertainty constraint;

[0014] According to the historical data of the distribution network, the uncertainty of the user demand on the demand side is measured using the Wasserstein distance algorithm to obtain the demand side uncertainty constraint;

[0015] The first constraint condition is constructed based on the supply-side uncertainty constraints and the demand-side uncertainty constraints.

[0016] The above scheme uses the Wasserstein distance algorithm to measure the uncertainty of distributed energy output on the supply side and the uncertainty of user demand on the demand side, making the uncertainty of the distribution network measurable and calculable, providing data support for the subsequent accurate calculation of the distribution network control strategy.

[0017] In a possible implementation method of the first aspect, the supply-side uncertainty constraint and the demand-side uncertainty constraint are specifically:

[0018] The supply-side uncertainty constraint and the demand-side uncertainty constraint are specifically expressed as follows:

[0019]

[0020] Where, d W1 d W∞ are the Wasserstein metric under the Wasserstein radius and the Wasserstein metric under the infinite norm Wasserstein radius, R t (ξ) is the real distribution of the historical data of the distribution network, Φ is ξ and The joint distribution of is a data set containing all historical data of distribution networks, ξ is sample data, κ1 is the sphere radius of Wasserstein distance under 1-norm, κ ∞ is the radius of the ball of infinite norm Wasserstein distance, t is the running time, T d is the set of running time, P is the empirical distribution, R is the real distribution, is the empirical distribution of the historical data of the distribution network, β1 is the confidence of the 1-norm Wasserstein distance, β ∞ is the confidence level of the infinite-norm Wasserstein distance.

[0021] In a possible implementation method of the first aspect, a second constraint condition is constructed according to the historical data of the distribution network and for the three-phase imbalance of the distribution network and the operation status of the electrical equipment, specifically:

[0022] According to the historical data of the distribution network, a three-phase power flow equation of the distribution network is obtained, and according to the three-phase power flow equation, a power flow constraint is constructed taking into account the three-phase imbalance of the distribution network; wherein the power flow constraint includes a distribution network power flow constraint, a voltage amplitude constraint, a line thermal limit constraint and a voltage imbalance constraint;

[0023] Establishing electrical equipment constraints on the installation and operation of electrical equipment in the distribution network based on the historical data of the distribution network;

[0024] Constructing node-branch constraints for voltage and current amplitudes of branches and nodes in the distribution network according to the distribution network historical data;

[0025] The second constraint condition is constructed based on the power flow constraint, electrical equipment constraint and node-branch constraint.

[0026] The above scheme constructs a power flow constraint that takes into account the three-phase imbalance of the distribution network based on the three-phase power flow equation of the distribution network; then constructs constraints on the installation and operation of electrical equipment in the distribution network, and constructs node-branch constraints for the voltage and current amplitudes of branches and nodes in the distribution network, providing a basis for achieving refined modeling.

[0027] In a possible implementation method of the first aspect, the distribution network power flow constraint, the voltage amplitude constraint, the line thermal limit constraint and the voltage imbalance constraint are specifically:

[0028] The distribution network flow constraint is specifically expressed as:

[0029]

[0030] Where P i,α is the active power of the line flow, Q i,α is the reactive power of the line flow, N + is a node set, V i α , are the voltage at node i and node j respectively, is the real part of the node admittance, is the angle difference between the α phase in node i and the β phase in node j, is the imaginary part of the node admittance;

[0031] The voltage amplitude constraint is specifically expressed as follows:

[0032]

[0033] In the formula, is the voltage amplitude of phase α at node i at time t, is the upper limit of the voltage amplitude of phase α at node i at time t, t is time t, and T is the time set;

[0034] The circuit thermal limit constraint is specifically expressed as follows:

[0035]

[0036] In the formula, is the current amplitude of phase αα on line ij at time t, is the upper limit of the current amplitude of phase αα on line ij at time t, is the admittance of phase αα on line ij, is the voltage amplitude of phase α at node i at time t, is the voltage amplitude of phase α at node j at time t, is the phase angle difference between the α phase of node i and the αα phase of node i at time t, is the upper limit of the current amplitude of phase αα on line ij, and L is the set of distribution network branches;

[0037] The voltage imbalance constraint is specifically expressed as:

[0038]

[0039] In the formula, is the negative sequence voltage of node i at time t, is the positive sequence voltage of node i at time t, is the upper limit of voltage unbalance in the distribution network, is the phase a voltage of node i at time t, is the phase b voltage of node i at time t, is the phase c voltage of node i at time t, e i120 is the Euler formula, representing cos120+isin120, e i240 is the Euler formula, which represents cos240+isin240. is the zero-sequence voltage of node i at time t.

[0040] In a possible implementation method of the first aspect, an objective function is constructed with the goal of minimizing the annual comprehensive operating cost of the distribution network, specifically:

[0041] According to the unit capacity cost and unit power cost of the electrical equipment in the distribution network, the investment cost of the electrical equipment is calculated in units of years to construct an annual investment cost function;

[0042] According to the loss cost caused by the three-phase imbalance in the distribution network, the operation and maintenance cost of the electrical equipment is calculated in units of years, and an annual operation and maintenance cost function is constructed;

[0043] According to the annual investment cost function and the annual operation and maintenance cost function, the objective function is constructed with the goal of minimizing the annual comprehensive operation cost of the distribution network.

[0044] The above scheme first constructs the objective function of the model from two aspects: the investment cost of electrical equipment and the loss cost caused by three-phase imbalance. It fully considers the factors that affect the changes in the comprehensive operating costs of the distribution network and realizes refined modeling from the perspective of constructing the objective function.

[0045] In a possible implementation method of the first aspect, the objective function is specifically:

[0046] The objective function is specifically expressed as:

[0047] minF=C INV +C OPE ;

[0048] Where, F is the annual comprehensive operating cost, C INV is the annual investment cost function, C OPE is the annual operation and maintenance cost function.

[0049] In a possible implementation method of the first aspect, the resource regulation optimization model is linearly transformed by a continuous linearization algorithm, and the resource regulation optimization model after the linear transformation is solved according to the first constraint condition to obtain a distribution network regulation strategy, which is specifically:

[0050] Determining nonlinear constraints in the resource regulation optimization model; wherein the nonlinear constraints include power flow constraints;

[0051] The nonlinear constraint condition is linearly transformed by a continuous linearization algorithm to obtain a resource control optimization model after linear transformation;

[0052] According to the first constraint condition, the uncertainty factor in the resource regulation optimization model after the linear transformation is measured to obtain the resource regulation optimization model after the model transformation;

[0053] The CCG algorithm is used to perform target decomposition and iterative solution on the resource control optimization model after model conversion to obtain the distribution network control strategy.

[0054] The above scheme first converts the nonlinear constraints in the model into a linear format that is easy to calculate, thereby reducing the computational complexity. Then, the model is converted using the first constraint that can quantify the uncertainty of the distribution network, so that the uncertainty in the model can also be measured, further reducing the computational complexity and improving the computational accuracy. Finally, the CCG algorithm is used for iterative solution to obtain a more accurate distribution network control strategy.

[0055] In a possible implementation method of the first aspect, the CCG algorithm is used to perform target decomposition and iterative solution on the resource control optimization model after model conversion to obtain a distribution network control strategy, which is specifically:

[0056] The optimization target of the resource control optimization model after model conversion is decomposed into the main target and the secondary target by using CCG algorithm;

[0057] Iteratively solving the converted resource regulation optimization model;

[0058] By solving the main objective and updating the lower limit value of the iterative solution;

[0059] By solving the sub-objective and updating the upper limit value of the iterative solution;

[0060] According to the upper limit value and the lower limit value, judging whether to end the iterative solution with a preset convergence threshold;

[0061] When the iterative solution is completed, the distribution network control strategy is obtained.

[0062] The above scheme uses the CCG algorithm to divide the calculation target into a primary target and a secondary target, and updates the threshold of the iterative solution through the primary target and the secondary target, so as to achieve convergence within a smaller number of iterations and improve the solution efficiency.

[0063] In a possible implementation method of the first aspect, resource regulation of the distribution network is performed according to the distribution network regulation strategy, specifically:

[0064] Determine the access nodes, access power and access capacity of electrical equipment according to the distribution network control strategy;

[0065] According to the access nodes, access power and access capacity, the location and capacity of electrical equipment in the distribution network are regulated.

[0066] The above scheme is based on the distribution network control strategy, and controls the access nodes, access power and access capacity of electrical equipment in the distribution network, so as to reduce the maintenance cost and investment cost of electrical equipment through site selection and capacity determination, while reducing the risk of voltage exceeding the limit at the distribution network nodes and branches, and realizing stable operation of the distribution network. BRIEF DESCRIPTION OF THE DRAWINGS

[0067] In order to more clearly illustrate the technical solution of the present application, the drawings required for use in the implementation manner will be briefly introduced below. Obviously, the drawings described below are only some implementation manners of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0068] Figure 1 It is a specific flow chart of a distribution network control optimization method based on resource regulation provided in a certain embodiment of the present application;

[0069] Figure 2 It is a distributed energy node access diagram of a distribution network according to a distribution network control optimization method based on resource regulation provided in a certain embodiment of the present application. DETAILED DESCRIPTION

[0070] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0071] It should be understood that the step numbers used in this article are only for the convenience of description and are not intended to limit the order in which the steps are executed.

[0072] Embodiment 1

[0073] Changing the traditional means of distribution network regulation and realizing high-precision real-time operation optimization of distribution network is the construction task of new power system. However, existing studies rarely consider the impact of electrical equipment such as distributed power sources, energy storage and compensators on the operation of distribution network, and rarely model the characteristics of three-phase imbalance in distribution network, resulting in the application effect of the constructed distribution network regulation model in actual situations is not satisfactory, and it is difficult to reduce the probability of voltage over-limit through regulation. Therefore, considering the risk of three-phase imbalance when constructing the distribution network regulation optimization model can greatly improve the operation stability of the distribution network.

[0074] like Figure 1 As shown, Figure 1 A specific flow chart of a distribution network control optimization method based on resource regulation is provided for a certain embodiment of the present application. The distribution network control optimization method based on resource regulation of this embodiment includes steps S1 to S5, which are described in detail as follows:

[0075] Step S1, based on the acquired historical data of the distribution network, the Wasserstein distance algorithm is used to calculate the uncertainty factors on the supply side and the demand side of the distribution network to construct a first constraint condition.

[0076] In this step, a large amount of historical data of the distribution network is first obtained. Based on the historical data of the distribution network, the Wasserstein distance algorithm is used to measure the uncertainty of the distributed energy output on the supply side and the uncertainty of the user demand on the demand side, so as to make these uncertain factors concrete and construct the first constraint condition.

[0077] The Wasserstein distance algorithm is a method to measure the distance between two probability distributions. From a probability perspective, the Wasserstein distance can be understood as the minimum cost required to transform one probability distribution into another.

[0078] In order to make the uncertainty of distributed energy output on the supply side, which is uncertain and difficult to describe in detail, and the uncertainty of user demand on the demand side measurable, the Wasserstein distance algorithm is used to calculate the distance between the empirical distribution and the real distribution of the historical data of the distribution network, so as to measure the uncertainty shown by the real distribution in the most approximate way with the empirical distribution, and construct the supply side uncertainty constraint and the demand side uncertainty constraint respectively. Then, based on the supply side uncertainty constraint and the demand side uncertainty constraint, the first constraint condition is constructed to realize the visualization of the uncertainty of distributed energy output on the supply side and the uncertainty of user demand on the demand side.

[0079] First, the confidence of Wasserstein metric and the radius of Wasserstein ball are obtained from the historical data of distribution network, and the Wasserstein distance is calculated to represent the distance between the empirical distribution and the real distribution of the historical data of distribution network. The first constraint condition is constructed. The specific formula is:

[0080]

[0081] Where, d W1 d W∞ are the Wasserstein metric under the Wasserstein radius and the Wasserstein metric under the infinite norm Wasserstein radius, R t (ξ) is the real distribution of the historical data of the distribution network, Φ is ξ and The joint distribution of is a data set containing all historical data of distribution networks, ξ is sample data, κ1 is the sphere radius of Wasserstein distance under 1-norm, κ ∞ is the radius of the ball of infinite norm Wasserstein distance, t is the running time, T d is the set of running time, P is the empirical distribution, R is the real distribution, is the empirical distribution of the historical data of the distribution network, β1 is the confidence of the 1-norm Wasserstein distance, β ∞ is the confidence level of the infinite-norm Wasserstein distance.

[0082] Step S2: constructing a second constraint condition according to the historical data of the distribution network and the three-phase imbalance of the distribution network and the operation status of the electrical equipment.

[0083] In this step, the three-phase power flow equation of the distribution network is obtained according to the historical data of the distribution network; wherein the power flow constraints include distribution network power flow constraints, voltage amplitude constraints, line thermal limit constraints and voltage imbalance constraints.

[0084] Furthermore, the distribution network power flow constraint, taking the common three-phase four-wire distribution network as an example, in order to integrate the neutral line loss into the phase line loss, firstly uses Kron simplification to reduce the original 4×4 network impedance matrix to a size of 3×3. Taking the injected power, node voltage, and node phase as variables, considering the mutual inductance and phase-to-phase capacitance, the distribution network power flow constraint is constructed based on the three-phase power flow equation. The specific expression is:

[0085]

[0086] Where P i,α is the active power of the line flow, Q i,αis the reactive power of the line flow, N + is a node set, V i α , are the voltage at node i and node j respectively, is the real part of the node admittance, is the angle difference between the α phase in node i and the β phase in node j, is the imaginary part of the node admittance.

[0087] The node admittance is defined as:

[0088]

[0089] The voltage amplitude constraint is specifically:

[0090]

[0091] In the formula, for, is, t is, T is; and assuming that the phase voltage of the root node is a fixed standard value and different boundary conditions are feasible, in this embodiment, an assumed boundary condition is usually selected, which is:

[0092]

[0093] The circuit thermal limit constraint is specifically expressed as follows:

[0094]

[0095] In the formula, is the current amplitude of phase αα on line ij at time t, is the upper limit of the current amplitude of phase αα on line ij at time t, is the admittance of phase αα on line ij, is the voltage amplitude of phase α at node i at time t, is the voltage amplitude of phase α at node j at time t, is the phase angle difference between the α phase of node i and the α phase of node i at time t, is the upper limit of the current amplitude of phase αα on line ij, and L is the set of distribution network branches.

[0096] In order to limit the voltage imbalance degree within the standard requirements, it is necessary to derive the positive sequence, negative sequence and zero sequence voltage components, and limit the negative sequence component to obtain the voltage imbalance constraint. The specific expression is:

[0097]

[0098] In the formula, is the negative sequence voltage of node i at time t, is the positive sequence voltage of node i at time t, is the upper limit of voltage unbalance in the distribution network, is the phase a voltage of node i at time t, is the phase b voltage of node i at time t, is the phase c voltage of node i at time t, e i120 is the Euler formula, representing cos120+isin120, e i240 is the Euler formula, which represents cos240+isin240. is the zero-sequence voltage of node i at time t.

[0099] According to the historical data of the distribution network, electrical equipment constraints are constructed for the installation and operation of electrical equipment in the distribution network; wherein the electrical equipment constraints include smart soft switch (SOP) installation constraints, smart soft switch operation constraints, energy storage (ESS) installation constraints, energy storage operation constraints, compensator (SVC) installation constraints, compensator operation constraints and distributed power supply operation constraints.

[0100] The specific expression of the intelligent soft switch installation constraint is:

[0101]

[0102] In the formula, It is a binary variable of the intelligent soft switch of branch ij, and is 1 if the intelligent soft switch is installed in branch ij.

[0103] The specific expression of the intelligent soft switch operation constraint is:

[0104]

[0105] In the formula, is the intelligent soft switch at time t The injected active power of the phase, is the intelligent soft switch at time t The injected reactive power of the phase; is the value of the two ends of the intelligent soft switch at time t Active power loss of the phase; γ SOP is the loss coefficient of the intelligent soft switch.

[0106] The energy storage installation constraint is specifically expressed as:

[0107]

[0108] In the formula, is the energy storage binary variable of node i, which is 1 if node i has energy storage installed.

[0109] The energy storage operation constraint is specifically expressed as:

[0110]

[0111] In the formula, and is a binary variable for energy storage, which is 1 if it is in the charging or discharging state; and The energy storage is the actual charging power and the actual discharging power; and The energy storage is the actual charging coefficient and the actual discharging coefficient; It is time t The stored residual energy of the phase; and It is the upper limit coefficient and lower limit coefficient of energy storage charge state.

[0112] The compensator installation constraint is specifically expressed as:

[0113]

[0114] In the formula, is the compensator binary variable of node i, which is 1 if node i has a compensator installed.

[0115] The compensator operation constraint is specifically expressed as:

[0116]

[0117] In the formula, is the compensator at time t The residual reactive power of the phase; and are the upper and lower limit coefficients of the compensator charge state.

[0118] The distributed generation (DG) operation constraint is specifically expressed as:

[0119]

[0120] In the formula, and is The upper and lower limits of the output active power of the distributed power sources of each phase; and is The upper and lower limits of the output reactive power of the distributed power sources of each phase; is the rated power factor angle of the distributed power source.

[0121] According to the historical data of the distribution network, node-branch constraints are constructed for the voltage and current amplitudes of branches and nodes in the distribution network; wherein the node-branch constraints include node power constraints and node voltage and branch current constraints.

[0122] The node power constraint is specifically expressed as:

[0123]

[0124] In the formula, and is through the branch ij Phase active power and reactive power; and yes The active power and reactive power injected by the phase at node i; and yes Phase active load and reactive load at node i.

[0125] The node voltage and branch current constraints are specifically expressed as follows:

[0126]

[0127] In the formula, and are the upper and lower limits of the voltage amplitude at node i, It is the upper limit of the current flowing through branch ij.

[0128] Step S3, constructing an objective function with the goal of minimizing the annual comprehensive operating cost of the distribution network, and constructing a resource regulation optimization model based on the objective function and the second constraint condition.

[0129] In this step, the investment cost of the electrical equipment is calculated on an annual basis according to the unit capacity cost and unit power cost of the electrical equipment in the distribution network, and an annual investment cost function is constructed. The specific calculation formula is:

[0130]

[0131] Where τ is the equivalent annual cost coefficient of electrical equipment, r is the annual discount rate, and L a is the service life of the electrical equipment a, c ESS,P 、c ESS,E 、c SVC and c SOP They are the unit cost of energy storage power, the unit cost of energy storage capacity, and the unit capacity investment cost of compensators and intelligent soft switches; and They are The actual installed energy storage power, actual installed energy storage capacity, actual installed compensator capacity and actual installed intelligent soft switch capacity of each phase.

[0132] According to the loss cost caused by the three-phase imbalance in the distribution network, the operation and maintenance cost of the electrical equipment is calculated in units of years, and an annual operation and maintenance cost function is constructed. The specific calculation formula is:

[0133] C OPE =C MAI +C LOSS +C DIS +C UB ;

[0134]

[0135] In the formula, C MAI is the maintenance cost of electrical equipment, C LOSS is the power loss cost caused by three-phase imbalance, C DIS is the penalty cost of abandoning solar and wind power, C UB Other costs caused by three-phase imbalance; c ESS,MAI is the maintenance cost of energy storage, c SVC,MAI is the maintenance cost of the compensator, c SOP,MAI is the maintenance cost of the intelligent soft switch; is the actual installed capacity of the energy storage φ phase, is the actual installed capacity of the compensator phase φ, is the actual installed capacity of the intelligent soft switch phase φ; B ESS , B SVC and B SOP It is a node group equipped with energy storage, compensator and intelligent soft switch. c PVG,DIS and c WTG,DIS is the power loss cost coefficient, the penalty cost per unit of wind and solar power reduced; r X is the impedance parameter of the φ phase; is the total current flowing through phase φ at time t in scenario i; Δt is the measurement time interval; and is the active power generated by generators, photovoltaics and wind power; B PG , B PVG and B WTG is the set of nodes with generators, photovoltaic generators and wind turbines; Ω Z and Ω T It is a set of scenes for typical days and times; D Z is the probability of the zth typical day scenario; c u is the cost coefficient, which represents the other cost increase caused by the three-phase imbalance in a single day; N sce is the number of scenes; γi is the sum of the three-phase imbalance in a single day under scenario i; γ t,i is the three-phase imbalance degree at time t in scenario i.

[0136] According to the annual investment cost function and the annual operation and maintenance cost function, the objective function is constructed with the goal of minimizing the annual comprehensive operation cost of the distribution network. The specific expression is:

[0137] minF=C INV +C OPE ;

[0138] Where, F is the annual comprehensive operating cost, C INV is the annual investment cost function, C OPE is the annual operation and maintenance cost function.

[0139] A resource regulation optimization model is constructed according to the objective function and the second constraint condition.

[0140] Step S4, linearly transforming the resource regulation optimization model through a continuous linearization algorithm, and solving the resource regulation optimization model after linear transformation according to the first constraint condition to obtain a distribution network regulation strategy.

[0141] In this step, because the resource regulation optimization model takes into account the three-phase power flow equation, and the nonlinearity of the three-phase power flow equation makes the resource regulation optimization model difficult to solve. In order to solve the calculation problem, a continuous linearization algorithm is used here to perform a linear transformation on the resource regulation optimization model.

[0142] The continuous linearization algorithm is a linearization method that approximates three-phase power flow, but since such methods all contain balance assumptions, including that the phase angle difference of different phases is balanced and close to 120°, or the unit voltage amplitude is equal to 1, it is essentially inconsistent with the characteristics of an unbalanced distribution network and will produce unacceptable errors. In order to minimize the error generated, this embodiment uses the CCG algorithm to solve the model, and expresses the solution of the model as a problem based on MILP, which does not require the assumption of strict balance conditions, is insensitive to initialization, and can also find a suboptimal solution while ensuring convergence.

[0143] The first step is to determine the nonlinear constraints in the resource regulation optimization model. First, it is determined that the three-phase power flow equation is nonlinear, so the power flow constraints are also nonlinear, which are specifically reflected in the line thermal limit constraints and voltage imbalance constraints.

[0144] The first step is to continuously linearize the three-phase power flow equation. It is necessary to ignore the time index t, use Taylor series expansion, omit quadratic terms and high-order terms to approximate the active power flow, specifically:

[0145]

[0146] Where V(0), θ(0), and P(0) are the initial variable values.

[0147] Define the Jacobian matrix J containing all partial derivative results. The specific expression is:

[0148]

[0149] In the formula, J PV , J Pθ , J QV and J Qθ They are respectively the matrix of node voltage amplitude to active power injection, the matrix of node voltage phase angle to active power injection, the matrix of node voltage amplitude to reactive power injection and the matrix of node voltage phase angle to reactive power injection.

[0150] The size of the Jacobian matrix is ​​6|N + |×6|N + |.

[0151] According to the Jacobian matrix, the three-phase power flow equation processed by continuous linearization is converted into a matrix form, and the linearization conversion iteration is performed again.

[0152] Finally, the expression of the three-phase power flow equation processed by continuous linearization at the kth iteration is:

[0153]

[0154] Then, the line thermal limit constraint and the voltage imbalance constraint are linearized. For the line thermal limit constraint, assuming that the angle difference between adjacent nodes can be ignored, the nonlinear term in the line thermal limit constraint can be approximated as:

[0155]

[0156] Therefore, the expression of the line thermal limit constraint after linearization is:

[0157]

[0158] For the voltage unbalance constraint, firstly transform the real part of the negative sequence voltage component and the imaginary part Using Taylor series expansion, the approximate expression is:

[0159]

[0160] In the formula, is the initial value.

[0161] The linearized voltage unbalance constraint obtained is approximately expressed as:

[0162]

[0163] According to the linearized three-phase power flow equation and constraints, the resource regulation optimization model after linear transformation is obtained.

[0164] Then, according to the first constraint, the uncertainty factor in the resource control optimization model after linear transformation is measured to obtain a linearized resource control optimization model expressed by Wasserstein distance, which is:

[0165]

[0166] y≥0

[0167] y REG =F3

[0168] G1x ≥ G2-G3y

[0169] J1x=J2-ξ 3t F3-ξ 4t F4

[0170]

[0171] In the formula, is the empirical distribution based on historical power grid data; 1t , 2t , 3t and 4t They are the four uncertainties at time t; is a dataset containing all historical power grid data; β1 and β ∞ is the confidence of Wasserstein metric; κ1 and κ ∞ is the radius of the Wasserstein sphere; R is the true distribution; is a dataset containing all historical data; R t (ξ) is the real distribution of historical power grid data.

[0172] The above formula is the first-stage linear inequality constraint obtained based on the first constraint condition. If the linearity of the model can be guaranteed, it can also be converted into the following formula that is easier to calculate:

[0173]

[0174] Then, the CCG algorithm is used to decompose the target and iteratively solve the resource control optimization model after model conversion. Among them, the CCG algorithm is an algorithm for solving two-stage robust optimization problems. It gradually adds constraints and variables in an iterative manner to solve the optimization problem. It can reach convergence in a smaller number of iterations, thus improving the solution efficiency.

[0175] Furthermore, according to actual needs, the objective function of the resource regulation optimization model is decomposed into the main objective and the secondary objective, and then iteratively solved. The specific process is as follows:

[0176] (1) Set L B =0,U B = +∞, initialize k = 0;

[0177] (2) Solve the main objective and update the lower limit value

[0178] (3) Solve the sub-objective and update the upper limit value

[0179] (4) Check the convergence according to the preset convergence threshold. If U B -L B <=γ, the iteration terminates; otherwise, update k=k+1, add new constraints for calculation and return to (2).

[0180] When the iterative solution is completed, the distribution network control strategy is obtained.

[0181] Step S5: performing resource regulation on the distribution network according to the distribution network regulation strategy.

[0182] In this step, the access node, access power and access capacity of the electrical equipment are determined according to the distribution network control strategy; and the site selection and capacity determination of the electrical equipment in the distribution network are controlled according to the access node, access power and access capacity.

[0183] The method provided in this embodiment models the characteristics of three-phase imbalance in the distribution network, and uses the Wasserstein distance to describe the uncertainty of the model, so as to obtain a more accurate control strategy that can reduce the comprehensive operating cost and voltage over-limit risk of the distribution network. The siting and sizing of electrical equipment in the distribution network are regulated through the control strategy, so as to improve the operating efficiency and operating stability of the distribution network.

[0184] Embodiment 2

[0185] Furthermore, in order to reflect the technical effect of the above method embodiment, this embodiment provides the result of the distribution network control strategy for resource control of the distribution network, mainly taking a typical IEEE 33-node system of a medium voltage 12.66V line as the planning object for explanation.

[0186] For ease of explanation, only the parts related to this embodiment are shown, including:

[0187] Step T1, select two groups of photovoltaic access distribution network examples, the access nodes are node 17 and node 30, as follows Figure 2 As shown, a distributed power source is connected to each of nodes 17 and 30.

[0188] Step T2, set up three distribution network control schemes. Scheme 1 only considers network reconstruction as an optimization method, scheme 2 considers both intelligent soft switches and energy storage access, and scheme 3 is the distribution network control strategy provided by this application.

[0189] The following table shows the planning results of these three schemes and the site selection and capacity determination results of electrical equipment.

[0190] Table 1 Planning results

[0191]

[0192] Table 2 Results of intelligent soft switch site selection and capacity determination

[0193]

[0194]

[0195] Table 3 Energy storage site selection and capacity determination results

[0196]

[0197] Table 4 Compensator site selection and capacity determination results

[0198]

[0199] According to the above table, after comparing the three schemes, it can be found that the coordinated regulation of two of the three, intelligent soft switch, energy storage or compensator, can not only promote the penetration of distributed power sources, but also reduce costs. The joint access of intelligent soft switch, energy storage and compensator adopted in scheme 3 is more effective. This is because intelligent soft switch, energy storage and compensator can improve the operation of active power and reactive power distribution system, which can not only achieve active power regulation, feeder load balancing and network loss reduction, but also provide reactive support and regulate node voltage.

[0200] The method provided in this embodiment proposes three distribution network control schemes, one of which is based on the distribution network control strategy provided in this application, and is coordinated by intelligent soft switches, energy storage and compensators. It can be seen from the site selection and capacity determination results that the joint access of the three electrical equipment is more effective, the calculated annual comprehensive operating cost of the distribution network is the smallest, and the economic benefits obtained are the highest.

[0201] The specific embodiments described above further describe the purpose, technical solutions and beneficial effects of the present application in detail. It should be understood that the above description is only a specific embodiment of the present application and is not intended to limit the scope of protection of the present application. It is particularly pointed out that for those skilled in the art, any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the scope of protection of the present application.

Claims

1. A distribution network control optimization method based on resource regulation, characterized in that: include: Based on the historical data of the distribution network, the Wasserstein distance algorithm is used to calculate the uncertainty factors on the supply and demand sides of the distribution network to construct the first constraint condition; According to the historical data of the distribution network, a second constraint condition is constructed for the three-phase imbalance of the distribution network and the operation status of the electrical equipment; An objective function is constructed with the goal of minimizing the annual comprehensive operating cost of the distribution network, and a resource regulation optimization model is constructed based on the objective function and the second constraint condition; The resource control optimization model is linearly transformed by a continuous linearization algorithm, and the resource control optimization model after the linear transformation is solved according to the first constraint condition to obtain a distribution network control strategy; The distribution network resources are regulated according to the distribution network regulation strategy.

2. The distribution network control optimization method based on resource regulation according to claim 1 is characterized in that: According to the obtained historical data of the distribution network, the Wasserstein distance algorithm is used to calculate the uncertainty factors on the supply side and the demand side of the distribution network, and the first constraint condition is constructed, which is specifically: According to the historical data of the distribution network, the uncertainty of the distributed energy output on the supply side is measured using the Wasserstein distance algorithm to obtain the supply side uncertainty constraint; According to the historical data of the distribution network, the uncertainty of the user demand on the demand side is measured using the Wasserstein distance algorithm to obtain the demand side uncertainty constraint; The first constraint condition is constructed based on the supply-side uncertainty constraints and the demand-side uncertainty constraints.

3. The distribution network control optimization method based on resource regulation according to claim 2 is characterized in that: The supply-side uncertainty constraints and demand-side uncertainty constraints are specifically: The supply-side uncertainty constraint and the demand-side uncertainty constraint are specifically expressed as follows: Where, d W1 ,d W∞ are the Wasserstein metric under the Wasserstein radius and the Wasserstein metric under the infinite norm Wasserstein radius, R t (ξ) is the real distribution of the historical data of the distribution network, Φ is ξ and The joint distribution of is a data set containing all historical data of distribution networks, ξ is sample data, κ1 is the sphere radius of Wasserstein distance under 1-norm, κ ∞ is the radius of the ball of infinite norm Wasserstein distance, t is the running time, T d is the set of running time, P is the empirical distribution, R is the real distribution, is the empirical distribution of the historical data of the distribution network, β1 is the confidence of the 1-norm Wasserstein distance, β ∞ is the confidence level of the infinite-norm Wasserstein distance.

4. The distribution network control optimization method based on resource regulation according to claim 1 is characterized in that: The second constraint condition is constructed according to the historical data of the distribution network and the three-phase imbalance of the distribution network and the operation status of the electrical equipment, specifically: According to the historical data of the distribution network, a three-phase power flow equation of the distribution network is obtained, and according to the three-phase power flow equation, a power flow constraint is constructed taking into account the three-phase imbalance of the distribution network; wherein the power flow constraint includes a distribution network power flow constraint, a voltage amplitude constraint, a line thermal limit constraint and a voltage imbalance constraint; Establishing electrical equipment constraints on the installation and operation of electrical equipment in the distribution network based on the distribution network historical data; Constructing node-branch constraints for voltage and current amplitudes of branches and nodes in the distribution network according to the distribution network historical data; The second constraint condition is constructed based on the power flow constraint, electrical equipment constraint and node-branch constraint.

5. The distribution network control optimization method based on resource regulation according to claim 1 is characterized in that: The distribution network power flow constraint, voltage amplitude constraint, line thermal limit constraint and voltage imbalance constraint are specifically: The distribution network flow constraint is specifically expressed as: Where P i,α is the active power of the line flow, Q i,α is the reactive power of the line flow, N + is a node set, are the voltage at node i and node j respectively, is the real part of the node admittance, is the angle difference between the α phase in node i and the β phase in node j, is the imaginary part of the node admittance; The voltage amplitude constraint is specifically expressed as follows: In the formula, is the voltage amplitude of phase α at node i at time t, is the upper limit of the voltage amplitude of phase α at node i at time t, t is time t, and T is the time set; The circuit thermal limit constraint is specifically expressed as follows: In the formula, is the current amplitude of phase αα on line ij at time t, is the upper limit of the current amplitude of phase αα on line ij at time t, is the admittance of phase αα on line ij, is the voltage amplitude of phase α at node i at time t, is the voltage amplitude of phase α at node j at time t, is the phase angle difference between the α phase of node i and the αα phase of node i at time t, is the upper limit of the current amplitude of phase αα on line ij, and L is the set of distribution network branches; The voltage imbalance constraint is specifically expressed as: In the formula, is the negative sequence voltage of node i at time t, is the positive sequence voltage of node i at time t, is the upper limit of voltage unbalance in the distribution network, is the phase a voltage of node i at time t, is the phase b voltage of node i at time t, is the phase c voltage of node i at time t, e i120 is the Euler formula, representing cos120+isin120, e i240 is the Euler formula, which represents cos240+isin240. is the zero-sequence voltage of node i at time t.

6. The distribution network control optimization method based on resource regulation according to claim 1 is characterized in that: The objective function is constructed with the goal of minimizing the annual comprehensive operating cost of the distribution network, specifically: According to the unit capacity cost and unit power cost of the electrical equipment in the distribution network, the investment cost of the electrical equipment is calculated in units of years to construct an annual investment cost function; According to the loss cost caused by the three-phase imbalance in the distribution network, the operation and maintenance cost of the electrical equipment is calculated in units of years, and an annual operation and maintenance cost function is constructed; According to the annual investment cost function and the annual operation and maintenance cost function, the objective function is constructed with the goal of minimizing the annual comprehensive operation cost of the distribution network.

7. The distribution network control optimization method based on resource regulation according to claim 6 is characterized in that: The objective function is specifically: The objective function is specifically expressed as: minF=C INV +C OPE ; Where, F is the annual comprehensive operating cost, C INV is the annual investment cost function, C OPE is the annual operation and maintenance cost function.

8. The distribution network control optimization method based on resource regulation according to claim 1 is characterized in that: The resource control optimization model is linearly transformed by the continuous linearization algorithm, and the resource control optimization model after the linear transformation is solved according to the first constraint condition to obtain the distribution network control strategy, which is specifically: Determining nonlinear constraints in the resource regulation optimization model; wherein the nonlinear constraints include power flow constraints; The nonlinear constraint condition is linearly transformed by a continuous linearization algorithm to obtain a resource control optimization model after linear transformation; According to the first constraint condition, the uncertainty factor in the resource regulation optimization model after the linear transformation is measured to obtain the resource regulation optimization model after the model transformation; The CCG algorithm is used to perform target decomposition and iterative solution on the resource control optimization model after model conversion to obtain the distribution network control strategy.

9. The distribution network control optimization method based on resource regulation according to claim 8 is characterized in that: The CCG algorithm is used to perform target decomposition and iterative solution on the resource control optimization model after model conversion to obtain the distribution network control strategy, which is specifically: The optimization target of the resource control optimization model after model conversion is decomposed into the main target and the secondary target by using CCG algorithm; Iteratively solving the converted resource regulation optimization model; By solving the main objective and updating the lower limit value of the iterative solution; By solving the sub-objective and updating the upper limit value of the iterative solution; According to the upper limit value and the lower limit value, judging whether to end the iterative solution with a preset convergence threshold value; When the iterative solution is completed, the distribution network control strategy is obtained.

10. The distribution network control optimization method based on resource regulation according to claim 1, characterized in that: The resource regulation of the distribution network according to the distribution network regulation strategy is specifically as follows: Determine the access nodes, access power and access capacity of electrical equipment according to the distribution network control strategy; According to the access nodes, access power and access capacity, the location and capacity of electrical equipment in the distribution network are regulated.

Images