Voltage Balance Control Method Based on Source-Load Polarity Switching in Bipolar DC Distribution Network

Abstract

The present invention relates to the technical field of DC distribution network voltage balance, and specifically relates to a voltage balance control method based on source-load polarity switching in a bipolar DC distribution network, including: establishing a multi-objective optimization model with the sum of switching times and the sum of voltage unbalances as the objective functions; calculating preliminary switching schemes for different objective functions based on the multi-objective optimization model; solving the multi-objective optimization model based on the preliminary switching schemes in combination with the genetic algorithm to obtain alternative switching schemes; determining the optimal switching scheme that can balance the values of the two objective functions from the alternative switching schemes based on the fuzzy membership function method, and then implementing the polarity switching of the unipolar source-load in the bipolar DC distribution network based on the optimal switching scheme. The voltage balance control method of the present invention can reduce the sum of source-load polarity switching times and can effectively suppress the voltage imbalance throughout the day, thereby improving the effect of voltage balance control based on source-load polarity switching in the bipolar DC distribution network.

Description

Technical Field

[0001] The present invention relates to the technical field of DC distribution network voltage balance, and particularly to a voltage balance control method based on source-load polarity switching in a bipolar DC distribution network. Background Art

[0002] Compared with an AC distribution network, a DC distribution network has advantages such as fewer conversion links, lower line losses, larger power supply capacity, and higher power supply reliability. The topological structure of a DC distribution network can be divided into a single-pole type and a bipolar type. Different from a single-pole DC distribution network, a bipolar DC distribution network has characteristics such as multiple voltage levels, multiple power supply circuits, and reliable grounding. Its power supply method is more flexible and its power supply reliability is higher.

[0003] In a bipolar DC distribution network, alternating current passes through an AC-DC converter and a voltage balancer in sequence and is converted into direct current in the bipolar DC distribution network. Both the power source and the load can be connected to the distribution network in the form of bipolar or single-pole, and the single-pole connection is further divided into positive-pole connection and negative-pole connection. Uneven polarity distribution of each single-pole source-load in the bipolar DC distribution network will cause voltage imbalance problems in the network. Voltage imbalance in a bipolar DC distribution network will increase the losses of the power grid and affect economic benefits.

[0004] The applicant has found that switching the polarity of the single-pole source-load in the distribution network and making it evenly distributed in the network helps to suppress voltage imbalance in the power grid. However, when designing the source-load polarity switching control method in a bipolar DC distribution network in the existing solutions, only the load polarity switching that can be achieved with fewer switching times at a single moment is considered to improve the voltage imbalance generated by the bipolar DC distribution network, and the switching strategy that takes the power source as the switching object and the damage to economic benefits caused by excessive switching times and excessive voltage imbalance degree throughout the day are not considered, that is, the impact brought by the sum of source-load polarity switching times and the voltage imbalance throughout the day cannot be effectively taken into account, resulting in poor control effect of source-load polarity switching in the bipolar DC distribution network. Therefore, how to design a method that can reduce the sum of source-load polarity switching times and can effectively suppress the voltage imbalance throughout the day is a technical problem that needs to be solved urgently. Summary of the Invention

[0005] Aiming at the deficiencies of the above-mentioned existing technologies, the technical problem to be solved by the present invention is: how to provide a voltage balance control method based on source-load polarity switching in a bipolar DC distribution network, which can reduce the sum of source-load polarity switching times and can effectively suppress the voltage imbalance throughout the day, so as to improve the control effect of voltage balance based on source-load polarity switching in the bipolar DC distribution network.

[0006] To solve the above technical problems, the present invention adopts the following technical solutions:

[0007] A voltage balance control method based on source-load polarity switching in a bipolar DC distribution network, comprising:

[0008] S1: Establish a multi-objective optimization model with the sum of switching times and the sum of voltage unbalance degrees as the objective functions;

[0009] S2: Calculate preliminary switching schemes for different objective functions based on the multi-objective optimization model;

[0010] S3: Solve the multi-objective optimization model based on the preliminary switching scheme combined with the genetic algorithm to obtain alternative switching schemes;

[0011] S4: Determine the optimal switching scheme that can balance the values of the two objective functions from the alternative switching schemes based on the fuzzy membership function method, and then implement the polarity switching of the single-pole source-load in the bipolar DC distribution network based on the optimal switching scheme.

[0012] Preferably, in step S1, the objective functions of the multi-objective optimization model are expressed by the following formula:

[0013] min F(X m )={f1(X m ),f2(X m )};

[0014]

[0015]

[0016] Where: F(X m ) represents the objective function of the multi-objective optimization model; X m represents the vector of the total switching scheme; f1(X m ) is SVUF(X m ), representing the sum of voltage unbalance degrees; f2(X m ) is SNSA(X m ), representing the sum of switching times; T represents the number of time periods; M represents the number of all nodes; VUF i,j (X i ) represents the voltage unbalance degree of node j in time period i; and represent the positive voltage and negative voltage of node j in time period i; NSA i (X i ,X i-1 ) represents the sum of switching times of all single-pole nodes from time period i-1 to time period i; X i and X i-1A vector representing the switching scheme within time period i and time period i-1; S represents the number of all single-pole nodes, where each single-pole node has only one single-pole power source or single-pole load; x i,s and x i-1,s represents the polarity state of the single-pole source and load at the single-pole node s within time period i and time period i-1, taking 1 for the positive pole and 0 for the negative pole.

[0017] Preferably, in step S1, the constraint conditions of the multi-objective optimization model are represented by the following formula:

[0018]

[0019] In the formula: I br (X m ) represents the vector of line current; represents the vector of the maximum allowable value of line current; represents the vector of the minimum allowable value of line-to-ground voltage; V br (X m ) represents the vector of line-to-ground voltage; represents the vector of the maximum allowable value of line-to-ground voltage; represents the vector of the minimum allowable value of source and load terminal voltage; V pnb (X m ) represents the vector of source and load terminal voltage; represents the vector of the maximum allowable value of source and load terminal voltage; VUF(X m ) represents the vector of voltage unbalance; VUF max represents the vector of the maximum allowable value of voltage unbalance;

[0020] The constraint conditions are abbreviated as the following formula:

[0021] C1 ≤ C(X m ) ≤ C2;

[0022] In the formula: C(X m ) represents the vector of line current, line-to-ground voltage, source and load terminal voltage, and voltage unbalance; C1 represents the vector of the minimum allowable value of line-to-ground voltage and source and load terminal voltage, where the element values corresponding to line current and voltage unbalance in C(X m ) are 0; C2 represents the vector of the maximum allowable value of line current, line-to-ground voltage, source and load terminal voltage, and voltage unbalance.

[0023] Preferably, in step S2, two preliminary switching schemes need to be calculated: the purpose of the first preliminary switching scheme is to calculate the minimum sum of switching times when the sum of voltage unbalances is at its minimum; the purpose of the second preliminary switching scheme is to calculate the minimum sum of voltage unbalances when the sum of switching times is at its minimum.

[0024] Preferably, for the first preliminary switching scheme:

[0025] The optimization model for time period i is represented by the following formula:

[0026]

[0027]

[0028] In the formula: the subscript pre indicates that the solution result of this variable will be used as the preliminary solution result of the first preliminary switching scheme; the subscript c indicates that the dimension of the constraint condition vector is reduced to adapt to the dimension of a single time period;

[0029] Use the element y in the vector Y with T dimensions i to control each element in the switching scheme obtained by solving the optimization model for the above time period i whether to take the inverse during calculation, and optimize it; where the value of y i is 1 or 0;

[0030] The optimization model of the first preliminary switching scheme is represented by the following formula:

[0031]

[0032] s.t.C1 ≤ C[X m,first (Y)] ≤ C2;

[0033] In the formula: X m,first represents the first preliminary switching scheme; y i is the element in Y corresponding to time period i; and represent the source-load polarities of the single-pole node s corresponding to time period i and time period i - 1 in the preliminary solution result; the subscript first indicates that the solution result of this variable will be used as the first preliminary switching scheme.

[0034] Preferably, the optimization model of the second preliminary switching scheme is represented by the following formula:

[0035]

[0036] s.t.C1 ≤ C(X m,second ) ≤ C2;

[0037] In the formula: the subscript "second" indicates that the solution result of this variable will be used as the second preliminary switching scheme.

[0038] Preferably, in step S3, before solving the multi-objective optimization model, it is first adjusted to the following form:

[0039] min F(X m,mul )={f1(X m,mul ),f2(X m,mul )};

[0040] s.t.C1≤C(X m,mul )≤C2;

[0041] In the formula: the subscript "mul" indicates that this variable is a variable in the multi-objective optimization model.

[0042] Preferably, the multi-objective optimization model is solved through the following steps:

[0043] S301: Use the two calculated preliminary switching schemes as two initial individuals in the initial population of the genetic algorithm, and the other initial individuals in the initial population are randomly generated;

[0044] S302: Obtain offspring by performing selection, mutation, and crossover operations on the individuals in the population;

[0045] S303: Calculate the objective function values and feasibility of the offspring, then add the offspring to the original individuals and retain l optimal new offspring according to the distances and ranks among the individuals; where, l represents the specified quantity;

[0046] S304: Repeat steps S302 to S303 until the iteration meets the set conditions, and output the Pareto front; where, each point on the Pareto front represents a different alternative switching scheme.

[0047] Preferably, in step S4, the optimal switching scheme is determined through the following steps:

[0048] S401: Denote the maximum value of the first objective function in the multi-objective optimization model on the Pareto front as f1 max , the minimum value as f1 min , the maximum value of the second objective function on the Pareto front as f2 max , the minimum value as f2 min ;

[0049] S402: Denote the two objective function values represented by each point on the Pareto front as f1 q and f2 q ;

[0050] S403: According to f1 defined abovemax 、f1 min 、f2 max 、f2 min 、f1 q and f2 q Calculate the satisfaction degree of each point on the Pareto front, and then select the alternative switching scheme corresponding to the point with the highest satisfaction degree as the optimal switching scheme.

[0051] Preferably, in step S403, the satisfaction degree of each point on the Pareto front is calculated by the following formula:

[0052]

[0053] In the formula: u q represents the satisfaction degree of evaluating each point on the Pareto front; Q represents the number of points on the Pareto front.

[0054] Compared with the prior art, the voltage balance control method based on source-load polarity switching in the bipolar DC distribution network of the present invention has the following beneficial effects:

[0055] The present invention takes the unipolar power supply and the unipolar load together as the switchable objects, establishes a multi-objective optimization model with the sum of switching times and the sum of voltage unbalance degrees as the objective functions, and combines the genetic algorithm to solve the multi-objective optimization model to obtain the alternative switching schemes, so that the optimal switching scheme that takes into account both the sum of switching times and the sum of voltage unbalance degrees can be obtained. Furthermore, the sum of source-load polarity switching times can be reduced and the voltage unbalance throughout the day can be effectively suppressed, thereby improving the effect of voltage balance control based on source-load polarity switching in the bipolar DC distribution network and providing an effective way to more economically improve the voltage balance situation of the bipolar DC distribution network.

[0056] The objective function of the multi-objective optimization model in the present invention has the characteristic of high non-linearity, making it difficult to solve with traditional solution algorithms. At the same time, the number of variables in the multi-objective optimization model is the product of the number of time periods and the number of single-pole source loads, which is relatively large. If the method of randomly generating the initial population using the existing genetic algorithm is adopted, the initial individuals may either not satisfy the security constraints or result in a relatively large value of the objective function, thereby making it difficult for the genetic algorithm to obtain a better solution through subsequent operations such as selection, mutation, and crossover in a short period of time. Therefore, when using the genetic algorithm to solve the multi-objective optimization model in the present invention, the preliminary switching schemes for different objectives are calculated through the multi-objective optimization model, and then the preliminary switching schemes are added to the initial population, so that the initial individuals in the initial population can all satisfy the security constraints and limit the value of the objective function, thereby being able to promote the convergence progress and obtain a better solution in a short period of time, which is beneficial to solving the optimal switching scheme, being able to reduce the sum of the source-load polarity switching times and effectively suppress the voltage imbalance throughout the day, and thus being able to improve the effect of voltage balance control based on source-load polarity switching in a bipolar DC distribution network.

[0057] The present invention determines the optimal switching scheme that can balance the values of the two objective functions from the alternative switching schemes based on the fuzzy membership function method. The points on the Pareto front have made trade-offs in the values of the two objective functions. Different points have a larger value in a certain objective function, and correspondingly, a smaller value in the other objective function. Therefore, screening the schemes on the Pareto front can intuitively reflect the mutual restriction of the two objective functions, making it possible to better achieve reducing the sum of the source-load polarity switching times and effectively suppressing the voltage imbalance throughout the day, and thus being able to further improve the effect of voltage balance control based on source-load polarity switching in a bipolar DC distribution network. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] In order to make the objectives, technical solutions, and advantages of the invention clearer, the present invention will be further described in detail below with reference to the accompanying drawings, where:

[0059] Figure 1 is the logic block diagram of the voltage balance control method based on source-load polarity switching in a bipolar DC distribution network;

[0060] Figure 2 is the network structure diagram of a bipolar DC distribution network;

[0061] Figure 3 is the schematic diagram of the Pareto front. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0062] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. The components of the embodiments of the present invention usually described and illustrated in the accompanying drawings here can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely represents selected embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts fall within the scope of protection of the present invention.

[0063] It should be noted that similar reference numerals and letters denote similar items in the following drawings. Therefore, once an item is defined in one drawing, it does not require further definition and explanation in subsequent drawings. In the description of the present invention, it should be noted that the terms "center", "upper", "lower", "left", "right", "vertical", "horizontal", "inner", "outer", etc. indicate the orientation or positional relationship based on the orientation or positional relationship shown in the drawings, or the orientation or positional relationship in which the inventive product is customarily placed during use. This is only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and thus should not be construed as a limitation of the present invention. In addition, the terms "first", "second", "third", etc. are only used for descriptive distinction and should not be construed as indicating or implying relative importance. In addition, terms such as "horizontal" and "vertical" do not mean that the components are required to be absolutely horizontal or hanging, but can be slightly inclined. For example, "horizontal" only means that its direction is more horizontal relative to "vertical", and does not mean that the structure must be completely horizontal, but can be slightly inclined. In the description of the present invention, it should also be noted that unless otherwise clearly defined and limited, the terms "set", "installed", "connected", "coupled" should be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be directly connected or indirectly connected through an intermediate medium, and it can be the internal communication of two elements. For those of ordinary skill in the art, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0064] The following will be further described in detail through specific embodiments:

[0065] Embodiment:

[0066] In this embodiment, a voltage balance control method based on source-load polarity switching in a bipolar DC distribution network is disclosed.

[0067] AsFigure 1 As shown in Figure 1 , a voltage balance control method based on source-load polarity switching in a bipolar DC distribution network includes:

[0068] S1: Establish a multi-objective optimization model with the sum of switching times and the sum of voltage unbalance degrees as the objective functions;

[0069] S2: Calculate preliminary switching schemes for different objective functions based on the multi-objective optimization model;

[0070] S3: Solve the multi-objective optimization model based on the preliminary switching schemes combined with the genetic algorithm to obtain alternative switching schemes;

[0071] S4: Determine the optimal switching scheme that can balance the values of the two objective functions from the alternative switching schemes based on the fuzzy membership function method, and then implement the polarity switching of the single-pole source-load in the bipolar DC distribution network based on the optimal switching scheme.

[0072] The present invention takes the single-pole power supply and the single-pole load together as switchable objects, establishes a multi-objective optimization model with the sum of switching times and the sum of voltage unbalance degrees as the objective functions, and combines the genetic algorithm to solve the multi-objective optimization model to obtain alternative switching schemes, so that an optimal switching scheme that takes into account both the sum of switching times and the sum of voltage unbalance degrees can be obtained. Furthermore, it can reduce the sum of source-load polarity switching times and can effectively suppress the voltage unbalance throughout the day, thereby improving the effect of voltage balance control based on source-load polarity switching in the bipolar DC distribution network and providing an effective way to more economically improve the voltage balance of the bipolar DC distribution network.

[0073] The objective functions of the multi-objective optimization model in the present invention have the characteristics of high non-linearity and are difficult to solve with traditional solution algorithms. At the same time, the number of variables in the multi-objective optimization model is the product of the number of time periods and the number of single-pole source-loads, which is relatively large. If the method of randomly generating the initial population using the existing genetic algorithm is adopted, the initial individuals may either not meet the safety constraints or make the value of the objective function larger, resulting in the genetic algorithm being difficult to obtain a better solution through subsequent operations such as selection, mutation, and crossover in a short time. Therefore, when using the genetic algorithm to solve the multi-objective optimization model in the present invention, preliminary switching schemes for different objectives are calculated through the multi-objective optimization model, and then the preliminary switching schemes are added to the initial population, so that the initial individuals in the initial population can all meet the safety constraints and limit the value of the objective function, thereby advancing the convergence progress and obtaining a better solution in a short time, which is beneficial to solving the optimal switching scheme, reducing the sum of source-load polarity switching times and effectively suppressing the voltage unbalance throughout the day, and thus improving the effect of voltage balance control based on source-load polarity switching in the bipolar DC distribution network.

[0074] Based on the fuzzy membership function method, the present invention determines the optimal switching scheme that can balance the values of two objective functions from the alternative switching schemes, so as to better reduce the sum of the number of source-load polarity switches and effectively suppress the voltage imbalance throughout the day, thereby further improving the effect of voltage balance control based on source-load polarity switching in a bipolar DC distribution network.

[0075] The schematic diagram of the network structure of the bipolar DC distribution network is as Figure 2 shown. The alternating current passes through the AC-DC converter and the voltage balancer successively and is converted into direct current in the bipolar DC distribution network. Both the power source and the load can be connected to the distribution network in the form of bipolar or unipolar, and the unipolar connection is further divided into positive-pole connection and negative-pole connection. The uneven polarity distribution of each unipolar source-load in the bipolar DC distribution network will lead to voltage imbalance problems in the network, which can be solved by switching the polarity of the unipolar source-load (i.e., switching between the two connection methods of positive-pole connection and negative-pole connection).

[0076] The switching vector X is used to represent the source-load polarity of the nodes where each unipolar source-load is located. When an element takes the value of 1, it means that the source-load of the corresponding node is at the positive pole, and when it takes the value of 0, it means that the source-load of the corresponding node is at the negative pole. For the convenience of matrix operation, the switching vector X is extended to X exp , and its elements are arranged in the order of the node numbers. The elements corresponding to the nodes with bipolar source-loads take the value of 0, while the elements corresponding to the nodes with unipolar source-loads represent the polarity of the unipolar source-loads according to the same rule as X.

[0077] The relationship between the power of the source-load in the bipolar DC distribution network and X exp is written as the following formula:

[0078]

[0079] In the formula: the subscripts p, n, and b represent the polarities of the positive-pole source-load, negative-pole source-load, and bipolar source-load respectively; P pnb (X exp ) represents the power of the positive-pole, negative-pole, and bipolar source-loads; P p (X exp ) and P n (X exp ) are the power vectors of the positive-pole source-load and negative-pole source-load; E1 is a column vector with all elements being 1; O is a zero matrix; E0 is a column vector with all elements being 0; P uni is the power vector of the unipolar source-load, where the elements corresponding to the bipolar source-loads are 0; P b is the power vector of the bipolar source-load, where the elements corresponding to the unipolar source-loads are 0; The elements in P uni corresponding to the unipolar source-loads and the elements in P b corresponding to the bipolar source-loads are positive when representing the load and negative when representing the power source.

[0080] First, consider the case where X in Equation (1) exp is a fixed value. In a bipolar DC distribution network, the mismatch vector I mis represents the degree of mismatch between the incoming and outgoing currents at each node.

[0081]

[0082] In the formula: the subscripts +, N, and - represent the polarities of the positive line, neutral line, and negative line, respectively; represents the current flowing from other nodes to each node (this vector does not include the node where the voltage balancer is located); E is the identity matrix; represents the source-load current.

[0083] Expand to where the subscript vb indicates that this vector includes the node where the voltage balancer is located. and can be further expressed in terms of the line-to-ground voltage:

[0084]

[0085]

[0086] In the formula: V +,d , V N,d and V -,d represent the line-to-ground voltages of the three polarities respectively, and the subscript d indicates that this matrix is a diagonal matrix; G is the conductance matrix of the network; is the line-to-ground voltage vector including the voltage balancer node.

[0087] Substitute Equation (1), Equation (3), Equation (4), and the operation of changing the vector dimension into Equation (2), and it can be known that the current mismatch vector is a function of the line-to-ground voltage as a variable. Denote this relationship as I mis (V br ).

[0088] When each line-to-ground voltage is an exact value, the value of each element in I mis should be 0. Therefore, based on the Newton-Raphson algorithm, the following formula can be obtained:

[0089]

[0090] In the formula: k is the iteration number; J is the Jacobian matrix.

[0091] When I misWhen the norm of [the relevant quantity] is less than the set value, the iteration terminates. When X changes, the voltage distribution of the network will change. Therefore, different line-to-ground voltage vectors can be obtained through the above steps. The process of taking the elements in X as variables and performing the above iterative calculation to obtain the line-to-ground voltage can be denoted as V br (X). The relationship between the source-load voltage and X is as follows:

[0092]

[0093] It should be noted that the relevant technical means for calculating the power flow of the bipolar DC distribution network in this embodiment can be implemented with reference to the existing technical means.

[0094] The optimization variable is the switching state x of each monopole node in the distribution network at different time periods within a day i,s , and the x within the time period i i,s is arranged in the order of monopole nodes to obtain the vector X i , and then X i is arranged in chronological order to obtain the vector X m .

[0095] The first objective function is the sum of voltage unbalances, and the second objective function is the sum of switching times. The objective functions of the multi-objective optimization model are represented by the following formula:

[0096] min F(X m ) = {f1(X m ), f2(X m )} (7)

[0097]

[0098]

[0099] In the formula: F(X m ) represents the objective function of the multi-objective optimization model; X m represents the vector of the total switching scheme; f1(X m ) is SVUF(X m ), representing the sum of voltage unbalances; f2(X m ) is SNSA(X m ), representing the sum of switching times; T represents the number of time periods; M represents the number of all nodes; VUF i,j (X i ) represents the voltage unbalance of node j in time period i; and represent the positive voltage and negative voltage of node j in time period i; NSA i (X i , X i-1) represents the sum of the switching times of all single-pole nodes from time period i - 1 to time period i; X i and X i-1 represents the vector of the switching scheme between time period i and time period i - 1; S represents the number of all single-pole nodes, where each single-pole node has only one single-pole power source or single-pole load; x i,s and x i-1,s represents the polarity state of the single-pole source and load at the single-pole node s between time period i and time period i - 1, taking 1 for the positive pole and 0 for the negative pole.

[0100] The switching times from the initial polarity configuration to time period 1 are not included in the SNSA.

[0101] Considering the upper limit of line current, the upper and lower limits of line-to-ground voltage, the upper and lower limits of source and load voltage, and the upper limit of voltage unbalance degree of the network as constraints:

[0102]

[0103] In the formula: I br (X m ) represents the vector of line current; represents the vector of the maximum allowable value of line current; represents the vector of the minimum allowable value of line-to-ground voltage; V br (X m ) represents the vector of line-to-ground voltage; represents the vector of the maximum allowable value of line-to-ground voltage; represents the vector of the minimum allowable value of source and load terminal voltage; V pnb (X m ) represents the vector of source and load terminal voltage; represents the vector of the maximum allowable value of source and load terminal voltage; VUF(X m ) represents the vector of voltage unbalance degree; VUF max represents the vector of the maximum allowable value of voltage unbalance degree.

[0104] For the convenience of representing this constraint, the constraint condition is abbreviated as the following formula:

[0105] C1 ≤ C(X m ) ≤ C2 (11)

[0106] In the formula: C(X m ) represents the vector of line current, line-to-ground voltage, source and load terminal voltage, and voltage unbalance degree; C1 represents the vector of the minimum allowable value of line-to-ground voltage and the minimum allowable value of source and load terminal voltage, where corresponding to C(X m) The element values of the line current and voltage unbalance are 0; C2 represents the vector of the maximum allowable line current, the maximum allowable line-to-ground voltage, the maximum allowable source-load terminal voltage, and the maximum allowable voltage unbalance.

[0107] So far, the optimization model aiming to minimize the sum of voltage unbalances and the sum of switching times is established.

[0108] In the specific implementation process, due to the highly non-linear characteristics of this model, a genetic algorithm is considered for solving. The general genetic algorithm uses the method of randomly generating the initial population, and obtains the result of this time by performing operations such as selection, mutation, and crossover on the individuals in the initial population, and then takes it as the new population and iterates again. The above process repeats until the iteration meets the set conditions (such as the iteration times reach the set upper limit, etc.). It can be seen that the efficiency of the genetic algorithm in generating offspring depends not only on the specific schemes of operations such as selection, mutation, and crossover, but also has a close relationship with the population of the previous generation; tracing back to the source of iteration, that is, it is closely related to the initial population. Therefore, the preliminary switching schemes can be obtained separately for different objectives first.

[0109] Two preliminary switching schemes need to be calculated: the purpose of the first preliminary switching scheme is to calculate the minimum sum of switching times SNSA when the sum of voltage unbalances SVUF is the minimum; the purpose of the second preliminary switching scheme is to calculate the minimum sum of voltage unbalances SVUF when the sum of switching times SNSA is the minimum.

[0110] For the first preliminary switching scheme, first ignore SNSA, wait to obtain the switching scheme corresponding to the minimum SVUF, and then obtain the switching scheme with the minimum SNSA while keeping SVUF unchanged. A single-objective optimization model is constructed separately for each time period in the distribution network and solved using a genetic algorithm.

[0111] Specifically, the optimization model for time period i is represented by the following formula:

[0112]

[0113]

[0114] In the formula: the subscript pre indicates that the solution result of this variable will be used as the pre-solution result of the first preliminary switching scheme; the subscript c indicates that the dimension of the constraint condition vector is reduced to adapt to the dimension of a single time period;

[0115] Use the element y in the vector Y with T dimensions i to control the switching scheme obtained by solving the optimization model for the above time period i for each element Whether to take the inverse during calculation and optimize it; where y i has a value of 1 or 0.

[0116] In this embodiment, when y i takes 0, it means that during the calculation the inverse should be taken (i.e., changing from 1 to 0, or from 0 to 1); when y i takes 1, it means that during the calculation it should remain unchanged.

[0117] The optimization model of the first preliminary switching scheme is expressed by the following formula:

[0118]

[0119] s.t.C1≤C[X m,first (Y)]≤C2 (15)

[0120] In the formula: X m,first represents the first preliminary switching scheme; y i is the element corresponding to the time period i in Y; and represent the source-load polarities of the single-pole node s corresponding to the time period i and the time period i - 1 in the pre-solved result; the subscript first indicates that the solution result of this variable will be used as the first preliminary switching scheme.

[0121] By solving the optimization models in equations (12) and (13) for each time period and the optimization models in equations (14) and (15), the first preliminary switching scheme can be obtained.

[0122] Specifically, the optimization model of the second preliminary switching scheme is expressed by the following formula:

[0123]

[0124] s.t.C1≤C(X m,second )≤C2 (17)

[0125] In the formula: the subscript second indicates that the solution result of this variable will be used as the second preliminary switching scheme.

[0126] SNSA is 0, that is, each X i is the same, so the in equation (16) is expressed as a quantity related to the S-dimensional vector X second .

[0127] In the specific implementation process, before solving the multi-objective optimization model, it is first adjusted to the following form:

[0128] min F(X m,mul ) = {f1(X m,mul ), f2(X m,mul )} (18)

[0129] s.t. C1 ≤ C(X m,mul ) ≤ C2 (19)

[0130] In the formula: the subscript mul indicates that the variable is a variable in the multi-objective optimization model.

[0131] In the present invention, the multi-objective optimization model is solved by a genetic algorithm. However, the number of variables in the multi-objective optimization model is the product of the number of time periods and the number of single-pole source loads, which is relatively large. If the method of randomly generating the initial population by the existing genetic algorithm is adopted, the initial individuals among them may not satisfy the safety constraints or may result in a relatively large value of the objective function, thus making it difficult for the genetic algorithm to obtain a better solution through subsequent operations such as selection, mutation, and crossover in a short time. Therefore, when using the genetic algorithm to solve the multi-objective optimization model in the present invention, the preliminary switching scheme is added to the initial population, so that the initial individuals in the initial population can satisfy the safety constraints and limit the value of the objective function, thereby being able to promote the convergence progress and obtain a better solution in a short time, which is beneficial to solving the optimal switching scheme, can reduce the sum of the source-load polarity switching times, and can effectively suppress the voltage imbalance throughout the day, thus being able to further improve the effect of voltage balance control based on source-load polarity switching in the bipolar DC distribution network.

[0132] It should be noted that the relevant technical means for solving the multi-objective optimization model by the genetic algorithm in this embodiment can be implemented with reference to the existing technical means. The multi-objective optimization model is specifically solved through the following steps:

[0133] S301: Take the two calculated preliminary switching schemes as two initial individuals in the initial population of the genetic algorithm, and the other initial individuals in the initial population are randomly generated;

[0134] S302: Obtain offspring by performing selection, mutation, and crossover operations on the individuals in the population;

[0135] S303: Calculate the objective function value and feasibility of the offspring, then add the offspring to the original individuals and retain l optimal new offspring according to the distance between individuals and the rank of each individual; where l represents the specified quantity;

[0136] S304: Repeat steps S302 to S303 until the iteration meets the set conditions, and output the Pareto front; where, as Figure 3 shown, each point on the Pareto front represents a different alternative switching scheme.

[0137] In this embodiment, the alternative switching scheme refers to the result output by the genetic algorithm after calculation, and the Pareto front is composed of the two objective function values corresponding to each alternative switching scheme. "Individual" and "offspring" refer to the objects involved in the iterative calculation inside the genetic algorithm, and the results obtained after calculation and retention will participate in the calculation of the next round of iteration.

[0138] In the specific implementation process, the Pareto front can be obtained by solving the multi-objective optimization model through the genetic algorithm, and each point on the Pareto front represents a different alternative switching scheme. The following fuzzy membership function method can be used to screen out the scheme that can balance the values of the two objective functions.

[0139] It should be noted that the relevant technical means for screening the optimal switching scheme by using the fuzzy membership function method in this embodiment can be implemented with reference to the existing technical means.

[0140] The optimal switching scheme is determined through the following steps:

[0141] S401: Denote the maximum value of the first objective function in the multi-objective optimization model on the Pareto front as f1 max and the minimum value as f1 min , and the maximum value of the second objective function on the Pareto front as f2 max and the minimum value as f2 min ;

[0142] S402: Denote the two objective function values represented by each point on the Pareto front as f1 q and f2 q ;

[0143] S403: According to the defined f1 max , f1 min , f2 max , f2 min , f1 q and f2 q calculate the satisfaction degree of each point on the Pareto front, and then select the alternative switching scheme corresponding to the point with the highest satisfaction degree as the optimal switching scheme.

[0144] The satisfaction degree of each point on the Pareto front is calculated by the following formula:

[0145]

[0146] In the formula: u q represents the satisfaction degree of evaluating each point on the Pareto front; Q represents the number of points on the Pareto front.

[0147] In fact, the final solution can also be determined by the decision maker according to their different sensitivities to different objective functions. For example, if the decision maker is more sensitive to voltage imbalance, they can choose a solution with more switching times from the alternative solutions, but which can better improve the voltage imbalance situation of the network.

[0148] Based on the fuzzy membership function method, the present invention determines the optimal switching scheme that can balance the values of the two objective functions from the alternative switching schemes. The points on the Pareto front make trade-offs in the values of the two objective functions. Different points have larger values in one objective function, and correspondingly smaller values in the other objective function. Therefore, screening solutions on the Pareto front can intuitively reflect the mutual constraints of the two objective functions, enabling better reduction of the sum of source-load polarity switching times and effective suppression of the voltage imbalance throughout the day, thereby further improving the effect of voltage balance control based on source-load polarity switching in a bipolar DC distribution network.

[0149] 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. Those of ordinary skill in the art should understand that any modifications or equivalent replacements made to the technical solutions of the present invention without departing from the purpose and scope of the present technical solution should be covered by the scope of the claims of the present invention.

Claims

1. A voltage balance control method based on source-load polarity switching in a bipolar DC distribution network, characterized in that, Including: S1: Establish a multi-objective optimization model with the sum of switching times and the sum of voltage unbalances as the objective functions; In step S1, the objective functions of the multi-objective optimization model are expressed by the following formula: min F(X m ) = {f1(X m ), f2(X m )}; Where: F(X m ) represents the objective function of the multi-objective optimization model; X m represents the vector of the total switching scheme; f1(X m ) is SVUF(X m ), representing the sum of voltage unbalances; f2(X m ) is SNSA(X m ), representing the sum of the number of switchings; T represents the number of time periods; M represents the number of all nodes; VUF i,j (X i ) represents the voltage unbalance of node j in time period i; and represent the positive and negative voltages of node j in time period i; NSA i (X i , X i-1 ) represents the sum of the number of switchings of all monopole nodes from time period i - 1 to time period i; X i and X i-1 represent the vectors of the switching schemes in time periods i and i - 1; S represents the number of all monopole nodes, where each monopole node has only one monopole power source or monopole load; x i,s and x i-1,s represent the polarity states of the monopole source and load at monopole node s in time periods i and i - 1, taking 1 for the positive pole and 0 for the negative pole; S2: Calculate preliminary switching schemes for different objective functions based on the multi-objective optimization model; S3: Solve the multi-objective optimization model based on the preliminary switching schemes combined with the genetic algorithm to obtain alternative switching schemes; S4: Determine the optimal switching scheme that can balance the values of the two objective functions from the alternative switching schemes based on the fuzzy membership function method, and then implement the polarity switching of the single-pole source and load in the bipolar DC distribution network based on the optimal switching scheme.

2. The voltage balance control method based on source-load polarity switching in a bipolar DC distribution network according to claim 1, characterized in that: In step S1, the constraint conditions of the multi-objective optimization model are expressed by the following formula: Where: I br (X m ) represents the vector of line current; represents the vector of the maximum allowable value of line current; represents the vector of the minimum allowable value of line-to-ground voltage; V br (X m ) represents the vector of line-to-ground voltage; represents the vector of the maximum allowable value of line-to-ground voltage; represents the vector of the minimum allowable value of source-load terminal voltage; V pnb (X m ) represents the vector of source-load terminal voltage; represents the vector of the maximum allowable value of source-load terminal voltage; VUF(X m ) represents the vector of voltage unbalance; VUF max represents the vector of the maximum allowable value of voltage unbalance; The above constraint conditions are abbreviated as the following formula: C1 ≤ C(X m ) ≤ C2; Where: C(X m ) represents the vectors of line current, line-to-ground voltage, source-load terminal voltage, and voltage unbalance; C1 represents the vector of the minimum allowable values of line-to-ground voltage and source-load terminal voltage, where the element values corresponding to the line current and voltage unbalance in C(X m ) are 0; C2 represents the vector of the maximum allowable values of line current, line-to-ground voltage, source-load terminal voltage, and voltage unbalance.

3. The voltage balance control method based on source-load polarity switching in a bipolar DC distribution network according to claim 1, characterized in that: In step S2, two preliminary switching schemes need to be calculated: The purpose of the first preliminary switching scheme is to calculate the minimum sum of switching times when the sum of voltage unbalances is the minimum value; The purpose of the second preliminary switching scheme is to calculate the minimum sum of voltage unbalances when the sum of switching times is the minimum value.

4. The voltage balance control method based on source-load polarity switching in a bipolar DC distribution network according to claim 3, characterized in that: For the first preliminary switching scheme: The optimization model for time period i is expressed by the following formula: In the formula: The subscript pre indicates that the solution result of this variable will be used as the pre-solution result of the first preliminary switching scheme; The subscript c indicates that the dimension of the constraint condition vector is reduced to adapt to the dimension of a single time period; Use the element y in the vector Y with T dimensions i to control the switching scheme obtained by solving the optimization model for the above time period i for each element in whether to take the inverse during calculation and optimize it; where y i has a value of 1 or 0 The optimization model of the first preliminary switching scheme is expressed by the following formula: Where: X m,first represents the first preliminary switching scheme; y i is the element corresponding to time period i in Y; and represent the source-load polarities of the single-pole node s corresponding to time period i and time period i - 1 in the pre-solved result; the subscript first indicates that the solution result of this variable will be used as the first preliminary switching scheme.

5. The voltage balance control method based on source-load polarity switching in a bipolar DC distribution network according to claim 3, characterized in that: The optimization model of the second preliminary switching scheme is expressed by the following formula: In the formula: The subscript second indicates that the solution result of this variable will be used as the second preliminary switching scheme.

6. The voltage balance control method based on source-load polarity switching in a bipolar DC distribution network according to claim 3, characterized in that: In step S3, before solving the multi-objective optimization model, it is first adjusted to the following form: min F(X m,mul ) = {f1(X m,mul ), f2(X m,mul )}; s.t. C1≤C(X m,mul )≤C2; In the formula: The subscript mul indicates that this variable is a variable in the multi-objective optimization model.

7. The voltage balance control method based on source-load polarity switching in a bipolar DC distribution network according to claim 6, characterized in that: The multi-objective optimization model is solved through the following steps: S301: Use the two calculated preliminary switching schemes as two initial individuals in the initial population of the genetic algorithm, and the other initial individuals in the initial population are randomly generated; S302: Obtain offspring by performing selection, mutation, and crossover operations on the individuals in the population; S303: Calculate the objective function values and feasibility of the offspring, then add the offspring to the original individuals and retain l optimal new offspring according to the distance between individuals and the rank of each individual; where, l represents the specified quantity; S304: Repeat steps S302 to S303 until the iteration meets the set conditions, and output the Pareto front; where, each point on the Pareto front represents a different alternative switching scheme.

8. The voltage balance control method based on source-load polarity switching in a bipolar DC distribution network according to claim 7, characterized in that: In step S4, the optimal switching scheme is determined through the following steps: S401: Denote the maximum value of the first objective function in the multi-objective optimization model on the Pareto front as the minimum value as the maximum value of the second objective function on the Pareto front as the minimum value as S402: Denote the two objective function values represented by each point on the Pareto front as and S403: Calculate the satisfaction degree of each point on the Pareto front according to the above-defined and and then select the alternative switching scheme corresponding to the point with the highest satisfaction degree as the optimal switching scheme.

9. The voltage balance control method based on source-load polarity switching in a bipolar DC distribution network according to claim 8, characterized in that: In step S403, the satisfaction degree of each point on the Pareto front is calculated by the following formula: where: u q represents the satisfaction degree of each point on the evaluated Pareto front; Q represents the number of points on the Pareto front.

Images