Electromagnetic looped network optimal loop opening method and system based on adaptive evolutionary programming

Abstract

The invention discloses an electromagnetic looped network optimal loop opening method and system based on adaptive evolutionary programming, and relates to the technical field of electromagnetic looped network loop opening, and the method comprises the steps: determining an initial combination scheme of a power supply node and a load node, calculating an electrical distance, building an initial population, and carrying out the adaptive random variation of a power grid individual, generating a partition scheme and independently operating each partition to form a new network, performing load flow calculation, N-1-1 verification and short-circuit current verification, if the conditions are met, calculating the optimal power supply capacity and the active power loss, forming a relative target of a multi-target function, performing random variation and adaptive value calculation on parent and child populations, and calculating the optimal power supply capacity and the active power loss; excellent individuals are selected to form a new parent population, and optimal individuals are reserved. According to the method, the operation stability and economy of the power grid are improved, the concept of comprehensive evaluation is introduced into the electromagnetic looped network loop opening method, the optimal loop opening model is established, the multi-objective evolutionary programming algorithm is adopted for solving, the optimal partition scheme is determined, and the open-loop operation of the power grid is achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of electromagnetic ring network decoupling, and in particular to an electromagnetic ring network optimal decoupling method and system based on adaptive evolutionary programming. Background Art

[0002] As the construction scale of China's ultra-high voltage AC and DC transmission projects continues to expand, transmission lines of different voltage levels are connected through transformers to form a tight ring grid structure. With the continuous growth of system load and the continuous adjustment and improvement of the grid structure, the problems caused by the electromagnetic ring network are becoming increasingly severe. The short-circuit current level of the grid continues to rise, and the chain failures caused by high and low voltage line faults have brought huge threats to the stable and safe operation of the grid. Therefore, how to correctly open the electromagnetic ring network and ensure the safety and reliability of the open-loop operation of the electromagnetic ring network has become one of the research hotspots in the development of ultra-high voltage transmission projects in recent years.

[0003] The traditional electromagnetic ring network de-looping method relies on the de-looping experience of power grid technicians to initially determine the open-loop point, generate a candidate open-loop solution based on the open-loop point, and then conduct simulation analysis in combination with the actual power grid structure and case verification to obtain the best open-loop solution. However, this method currently relies too much on human experience, is highly subjective, and has a complex calculation process.

[0004] The present invention establishes an optimal ring-breaking model of an electromagnetic ring network, and adopts an adaptive evolutionary programming algorithm to solve it, thereby determining an optimal independent partitioning scheme for a power grid and realizing hierarchical and partitioned operation of the power grid. Summary of the invention

[0005] In view of the above-mentioned existing problems, the present invention aims to provide an optimal de-looping method for an electromagnetic ring network based on adaptive evolutionary programming, which achieves comprehensive optimization with short-circuit current limiting effect, active network loss, power supply capacity and number of partitions as indicators to realize the de-looping of the electromagnetic ring network.

[0006] In order to solve the above technical problems, an optimal de-ringing method for electromagnetic ring network based on adaptive evolutionary programming is proposed, including:

[0007] According to the network structure and operation characteristics of the actual power grid, the initial combination scheme of power supply nodes and power load nodes is determined, and the electrical distances between power supply nodes and between power supply and load nodes are calculated at the same time; any power grid is taken as an independent individual of the initial population, and the independent individual is subjected to adaptive random mutation to generate a total of M partitioning schemes, where each partitioning scheme contains the number of partitions and the power supply nodes in each partition, forming an initial parent population; based on the partitioning scheme of the random individuals in the parent population, the combination scheme of the power supply nodes and load nodes in each partitioning scheme is determined by the distance of the electrical distance, the connecting lines between the partitions are disconnected, and each partition operates independently to form a new network; the new network is subjected to power flow calculation and it is determined whether the power flow calculation converges. When the power flow converges, a verification is performed; otherwise, a constraint penalty function is used to assign a maximum number to the objective function, and the fitness value of the random individual is calculated; the new network is subjected to N-1-1 verification and short-circuit current verification. If both verifications meet the requirements, If there is no constraint condition, a relative target is generated. Otherwise, a constraint penalty function is used to assign a maximum number to the target function and calculate the fitness value of the random individual; the optimal power supply capacity of the random individual in each partition scheme and the optimal active network loss of the random individual are calculated to form the relative target of the multi-objective function; a new network is repeatedly formed to complete the calculation of the fitness values ​​of all random individuals in the parent population; the K individuals of the parent population are randomly mutated to generate K new individuals of the child population; a new network is repeatedly formed to complete the calculation of the fitness values ​​of all random individuals in the child population; the parent population and the child population are formed into a new population of 2K random individuals, and K individuals are selected according to the size of the fitness value to form a new parent population, including 2 / 3K excellent individuals and 1 / 3K poor individuals; the best individual in the current iteration is retained to determine whether all iterations are completed. When all iterations are completed, the best individual and scheme are output. When all iterations are completed, a new network is re-formed for the next iteration.

[0008] As a preferred solution of the electromagnetic ring network optimal de-ringing method based on adaptive evolutionary programming described in the present invention, the electrical distance includes the voltage and reactive power sensitivity between nodes, and the connection between voltage and reactive power regulation represents the closeness of the connection between nodes. The element d in the electrical distance matrix D ij for:

[0009]

[0010] Among them, B LL is the sensitivity matrix, which is replaced by the imaginary part of the node admittance matrix; the voltage / reactive sensitivity is used to represent the electrical distance between nodes;

[0011] When partitioning the power nodes, extract the rows and columns corresponding to the source nodes in the matrix D and get the matrix D again. GWhen partitioning the load, the relative electrical distance is used to measure the distance between the load node and the power node:

[0012] Extract the rows corresponding to the load nodes and the columns corresponding to the power supply nodes in the electrical distance matrix D to obtain the absolute electrical distance matrix D between the load nodes and the power supply nodes. hg , and then calculate the relative electrical distance l through simple transformation ij :

[0013]

[0014] Among them, min(D hg (i,:)) represents the matrix D hg The minimum value of the i-th row in , that is, the minimum distance between the load nodes in the i-th row.

[0015] As a preferred solution of the electromagnetic ring network optimal de-looping method based on adaptive evolutionary programming described in the present invention, the adaptive random variation includes:

[0016] Assume the initial population size is K, the maximum number of partitions is k, and input the original data;

[0017] Randomly generate a random number y of 0 or 1. When y=0, the current data is split and mutated; when y=1, the data is combined and mutated;

[0018] Randomly select an element in each power node number sequence, split and mutate it into 2 elements, and at the same time k increases by 1;

[0019] Randomly select two elements from each power node number sequence as the power number of a new partition, delete the previous two partition power numbers, and at the same time subtract 1 from k;

[0020] Store the individuals that have completed the mutation into the offspring population;

[0021] Determine whether all mutations are completed; when all mutations are completed, the process ends; when all mutations are not completed, re-enter the original data and perform mutation on the next individual.

[0022] As a preferred solution of the electromagnetic ring network optimal de-ringing method based on adaptive evolutionary programming described in the present invention, wherein: the forming of a new network includes determining the combination scheme of power supply nodes and load nodes in each partition scheme by the distance of electrical distance, partitioning the power supply and partitioning the load;

[0023] The power partitioning includes finding D G The smallest non-diagonal element in the power partition, record the row and column corresponding to the smallest non-diagonal element, record them as i and j; check the number of power nodes P in the determined power partition i , find DG The smallest P in row i i-1 elements, as a partition; delete D G P i The rows and columns corresponding to the power nodes are deleted repeatedly until D G is an empty set;

[0024] The load partitioning includes finding D hg The column j corresponding to the largest non-diagonal element in the last row of D is determined; whether the active capacity in the current partition is greater than the active load of the current load node, if so, the current load is moved to the power partition and the last row is deleted; if it does not exceed the active load of the current load node, the current element is removed and D is searched again. hg The column corresponding to the largest non-diagonal element in the last row of D is repeatedly checked to see whether the active capacity in the current partition is greater than the active load of the current load node until D hg Is an empty set.

[0025] As a preferred solution of the electromagnetic ring network optimal de-ringing method based on adaptive evolutionary programming described in the present invention, wherein: the N-1-1 check and short-circuit current check include dividing the substations in the power grid into three functional stations, namely hub stations, transfer stations and terminal stations, disconnecting the busbars and transformers of the hub stations;

[0026] Conduct grid topology analysis and determine network structure;

[0027] Use the Newton-Raphson method to calculate the power flow and determine whether the power flow calculation has converged. If it has converged, check whether it meets the N-1 index. If it has not converged, mark it as not passing the check.

[0028] The verification determines whether the N-1 index is satisfied; if the N-1 index is satisfied, the verification is marked as passed; if the N-1 index is not satisfied, the verification is marked as failed;

[0029] When the mark passes the verification, it is determined whether all the verifications are completed. When all the verifications are completed, the process ends; when all the verifications are not completed, the power grid topology analysis is repeated to perform power grid flow calculation.

[0030] As a preferred solution of the electromagnetic ring network optimal de-ringing method based on adaptive evolutionary programming described in the present invention, wherein: the relative objectives of forming the multi-objective function include that the maximum power supply capacity continuously increases the system load during calculation, and when the power of the line and the transformer is about to exceed the limit, the maximum power supply capacity is considered;

[0031] When calculating the minimum active network loss, the total net injected power of all nodes in the whole network is recorded, expressed as node injected power, and solved using the linearized optimal power flow method:

[0032] The optimized variables are the phase angle increment of the voltage at each node in the whole network and the active output increment of all generators, expressed as Δθ1, Δθ2…Δθ n , ΔP G1 ,ΔP G2 …ΔP Gn ;

[0033] The objective function is to minimize the active network loss and is expressed as:

[0034]

[0035] The inequality constraints include the generator active output limit and the line transmission power limit, namely:

[0036]

[0037] Among them, ΔP Gi =P min -P G , represents the lower limit of the change in the active output of the generator; Indicates the upper limit of the change in the active output of the generator; ΔP Lij =P Lij,min -P Lij , represents the lower limit of the change in line transmission active power; Indicates the upper limit of the change in line transmission active power;

[0038] Line transmission power increment ΔP Lij Expressed as optimization variables:

[0039] ΔP Lij =V i0 V j0 [(-G ij sinθ ij0 +B ij cosθ ij0 )*Δθ i +(G ij sinθ ij0 -B ij cosθ ij0 )*Δθ j ]

[0040] Among them, G ij is the real part of the elements in the node admittance matrix, B ij is the imaginary part of the elements in the node admittance matrix, V i0 ,V j0 are the initial voltages of nodes i and j, respectively, ij0 is the initial phase angle difference between nodes i and j;

[0041] The equality constraints include the active power balance constraint of the node and the phase angle invariance constraint of the balanced node. The injected active power of the node is:

[0042]

[0043] By linearizing the above equation, we get the active power balance constraint, and the equality constraint of the linear programming problem is as follows:

[0044]

[0045] Δθ ref =0

[0046] Among them, V0, θ0 variables are the initial states of the network and are constants.

[0047] As a preferred solution of the electromagnetic ring network optimal de-ringing method based on adaptive evolutionary programming described in the present invention, wherein: the calculation of the fitness value of the random individual includes obtaining the fitness value through a multi-objective relative dominance function, reflecting the superiority or inferiority of the individual, the larger the fitness value of the individual, the more the individual is the optimal solution; the smaller the fitness value of the individual, the greater the gap between the individual and the optimal solution;

[0048] A multi-objective function with minimum network loss, maximum power supply capacity and the largest number of partitions is established. The multi-objective relative dominance function expression is:

[0049]

[0050] Among them, ΔP(x k ), P G (x k )、N(x k ) are respectively the optimal network loss, optimal power supply capacity and optimal number of partitions; min(ΔP), max(P G ), max(N) are the minimum network loss, the maximum power supply capacity, and the maximum number of partition schemes respectively; λ is the penalty factor.

[0051] Another object of the present invention is to provide an optimal de-ringing system for an electromagnetic ring network based on adaptive evolutionary programming. The present invention solves the problem of finding the optimal partitioning scheme of power nodes and load nodes through adaptive evolutionary programming to optimize the operation of the power grid; by reducing active network losses, the power supply capacity of the power grid is improved, thereby improving the economy and reliability of the power grid; through N-1-1 verification and short-circuit current verification, it is ensured that the power grid can remain safe and stable under various operating conditions.

[0052] As a preferred solution of the electromagnetic ring network optimal de-ringing system based on adaptive evolutionary programming described in the present invention, it is characterized by comprising an initialization module, an adaptive mutation module, a network formation and power flow calculation module, a verification module, a multi-objective optimization module, a fitness value calculation module and a population update module;

[0053] The initialization module determines the initial combination scheme of power nodes and load nodes according to the grid structure and operation characteristics, calculates the electrical distance, receives the grid structure data, and outputs the initial population to the adaptive mutation module;

[0054] The adaptive mutation module performs adaptive random mutation on the initial population to generate a new partitioning scheme, receives the parent population data, and outputs the mutated child population to pass to the network formation and power flow calculation module;

[0055] The network formation and power flow calculation module receives the partitioning scheme, forms a new network structure, performs power flow calculation, determines whether the power flow has converged, and performs security verification through the verification module;

[0056] The verification module receives the power flow calculation results, performs N-1-1 verification and short-circuit current verification to ensure the safe and stable operation of the power grid, outputs the verification results, and the qualified data enters the multi-objective optimization module for optimization calculation;

[0057] The multi-objective optimization module receives the verification results, calculates the optimal power supply capacity and active network loss of each partition scheme, forms the relative target of the multi-objective function, and evaluates the fitness of the individual through the fitness value calculation module;

[0058] The fitness value calculation module receives the multi-objective optimization results, calculates the fitness value of the individual through the multi-objective relative dominance function, reflects the quality of the individual, and inputs it into the population update module for population iteration;

[0059] The population update module selects excellent individuals to form a new parent population according to the fitness value, and retains the current optimal individual.

[0060] A computer device includes a memory and a processor, wherein the memory stores a computer program, and is characterized in that when the processor executes the computer program, the steps of the optimal de-ringing method of an electromagnetic ring network based on adaptive evolutionary programming are implemented.

[0061] A computer-readable storage medium having a computer program stored thereon, characterized in that when the computer program is executed by a processor, the steps of the electromagnetic ring network optimal de-ringing method based on adaptive evolutionary programming are implemented.

[0062] Beneficial effects of the present invention: The present invention introduces the concept of comprehensive evaluation into the electromagnetic ring network de-looping method, establishes an optimal de-looping model, and adopts a multi-objective evolutionary programming algorithm to solve it, thereby determining the optimal partitioning scheme and realizing the open-loop operation of the power grid.

[0063] Compared with traditional electromagnetic ring network de-ringing, it has the following advantages:

[0064] The multi-objective evaluation optimization problem is solved by the relative objective dominance strategy, which effectively weighs the weight ratios among multiple objectives. It is easy to calculate and is objective and scientific.

[0065] It can automatically form the optimal partitioning and ring-breaking solution, reduce the reliance on manual experience, and is universal and scientific.

[0066] It can conduct a scientific and comprehensive evaluation of the partition and ring-breaking schemes and select the optimal scheme based on the fitness value. BRIEF DESCRIPTION OF THE DRAWINGS

[0067] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative work, among which:

[0068] Figure 1 An overall flow chart of an optimal de-ringing method for an electromagnetic ring network based on adaptive evolutionary programming is provided as an embodiment of the present invention.

[0069] Figure 2 A random mutation algorithm flow chart of an optimal de-ringing method for an electromagnetic ring network based on adaptive evolutionary programming is provided as an embodiment of the present invention.

[0070] Figure 3 A system solution module diagram of an electromagnetic ring network optimal de-looping system based on adaptive evolutionary programming provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0071] In order to make the above-mentioned purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the drawings of the specification. Obviously, the described embodiments are part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary persons in the art without creative work should fall within the scope of protection of the present invention.

[0072] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0073] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor is it an embodiment that is mutually exclusive with other embodiments, either individually or selectively.

[0074] The present invention is described in detail with reference to schematic diagrams. When describing the embodiments of the present invention, for the sake of convenience, the cross-sectional diagrams showing the device structure will not be partially enlarged according to the general scale, and the schematic diagrams are only examples, which should not limit the scope of protection of the present invention. In addition, in actual production, the three-dimensional dimensions of length, width and depth should be included.

[0075] At the same time, in the description of the present invention, it should be noted that the directions or positional relationships indicated by the terms "upper, lower, inner and outer" are based on the directions or positional relationships shown in the drawings, and are 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 direction, be constructed and operated in a specific direction, and therefore cannot be understood as limiting the present invention. In addition, the terms "first, second or third" are only used for descriptive purposes and cannot be understood as indicating or implying relative importance.

[0076] In the present invention, unless otherwise clearly specified and limited, the terms "install, connect, connect" should be understood in a broad sense, for example: it can be a fixed connection, a detachable connection or an integral connection; it can also be a mechanical connection, an electrical connection or a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal communication of two components. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0077] Example 1, reference Figure 1 , which is the first embodiment of the present invention, and provides an optimal ring-breaking method for an electromagnetic ring network based on adaptive evolutionary programming, comprising:

[0078] S1: According to the network structure and operation characteristics of the actual power grid, determine the initial combination scheme of power supply nodes and power grid load nodes, and calculate the electrical distances between power supply nodes and between power supply and load nodes.

[0079] Furthermore, according to the relationship between voltage and reactive power regulation, the closeness of the connection between nodes is represented by the element d in the electrical distance matrix D.ij for:

[0080]

[0081] Among them, B LL The sensitivity matrix is ​​replaced by the imaginary part of the node admittance matrix, and the voltage / reactive sensitivity is used to represent the electrical distance between nodes;

[0082] When partitioning the power nodes, extract the rows and columns corresponding to the source nodes in the matrix D and get the matrix D again. G When partitioning the load, the relative electrical distance is used to measure the distance between the load node and the power node:

[0083] Extract the rows corresponding to the load nodes and the columns corresponding to the power supply nodes in the electrical distance matrix D to obtain the absolute electrical distance matrix D between the load nodes and the power supply nodes. hg , and then calculate the relative electrical distance l through simple transformation ij :

[0084]

[0085] Among them, min(D hg (i,:)) represents the matrix D hg The minimum value of the i-th row in , that is, the minimum distance between the load nodes in the i-th row.

[0086] S2: Take any power grid as an independent individual of the initial population, and perform adaptive random mutation on the independent individual to generate a total of M partitioning schemes, where each partitioning scheme contains the number of partitions and the power nodes in each partition, forming the initial parent population.

[0087] Furthermore, let the initial population size be K, the maximum number of partitions be k, and input the original data;

[0088] Randomly generate a random number y of 0 or 1. When y=0, the current data is split and mutated; when y=1, the data is combined and mutated;

[0089] Randomly select an element in each power node number sequence, split and mutate it into 2 elements, and at the same time k increases by 1;

[0090] Randomly select two elements from each power node number sequence as the power number of a new partition, delete the previous two partition power numbers, and at the same time subtract 1 from k;

[0091] Store the individuals that have completed the mutation into the offspring population;

[0092] Determine whether all mutations are completed; when all mutations are completed, the process ends; when all mutations are not completed, re-enter the original data and perform mutation on the next individual.

[0093] S3: Based on the partitioning scheme of random individuals in the parent population, the combination scheme of power nodes and load nodes in each partitioning scheme is determined by the electrical distance, the connection lines between the partitions are disconnected, and each partition operates independently to form a new network.

[0094] It should be noted that the combination of power nodes and load nodes in each partition scheme is determined by the electrical distance, and the power partition and load partition are divided;

[0095] The power partitioning includes finding D G The smallest non-diagonal element in the power partition, record the row and column corresponding to the smallest non-diagonal element, record them as i and j; check the number of power nodes P in the determined power partition i , find D G The smallest P in row i i-1 elements, as a partition; delete D G P i The rows and columns corresponding to the power nodes are deleted repeatedly until D G is an empty set;

[0096] The load partitioning includes finding D hg The column j corresponding to the largest non-diagonal element in the last row of D is determined; whether the active capacity in the current partition is greater than the active load of the current load node, if so, the current load is moved to the power partition and the last row is deleted; if it does not exceed the active load of the current load node, the current element is removed and D is searched again. hg The column corresponding to the largest non-diagonal element in the last row of D is repeatedly checked to see whether the active capacity in the current partition is greater than the active load of the current load node until D hg Is an empty set.

[0097] S4: Calculate the power flow of the new network and determine whether the power flow calculation has converged. If the power flow has converged, perform verification. Otherwise, use the constraint penalty function to assign a maximum number to the objective function and calculate the fitness value of the random individual.

[0098] Perform N-1-1 check and short-circuit current check on the new network. If the check satisfies the constraint conditions, generate a relative target. Otherwise, use the constraint penalty function to assign a maximum number to the target function and calculate the fitness value of the random individual.

[0099] The optimal power supply capacity and the optimal active network loss of random individuals in each partition scheme are calculated to form the relative objectives of the multi-objective function.

[0100] Furthermore, the substations in the power grid are divided into three functional stations, namely hub stations, transfer stations and terminal stations, and the busbars and transformers of the hub stations are disconnected;

[0101] Conduct grid topology analysis and determine network structure;

[0102] Use the Newton-Raphson method to calculate the power flow and determine whether the power flow calculation has converged. If it has converged, check whether it meets the N-1 index. If it has not converged, mark it as not passing the check.

[0103] The verification determines whether the N-1 index is satisfied; if the N-1 index is satisfied, the verification is marked as passed; if the N-1 index is not satisfied, the verification is marked as failed;

[0104] When the mark passes the verification, it is determined whether all the verifications are completed. When all the verifications are completed, the process ends; when all the verifications are not completed, the power grid topology analysis is repeated to perform power grid flow calculation.

[0105] It should be noted that the maximum power supply capacity is calculated by continuously adding system loads, and when the line and transformer are about to exceed the power limit, the maximum power supply capacity is considered;

[0106] When calculating the minimum active network loss, the total net injected power of all nodes in the whole network is recorded, expressed as node injected power, and solved using the linearized optimal power flow method:

[0107] The optimized variables are the phase angle increment of the voltage at each node in the whole network and the active output increment of all generators, expressed as Δθ1, Δθ2…Δθ n , ΔP G1 ,ΔP G2 …ΔP Gn ;

[0108] The objective function is to minimize the active network loss and is expressed as:

[0109]

[0110] The inequality constraints include the generator active output limit and the line transmission power limit, namely:

[0111]

[0112] Among them, ΔP Gi =P min -P G , represents the lower limit of the change in the active output of the generator; Indicates the upper limit of the change in the active output of the generator; ΔP Lij =P Lij,min -P Lij , represents the lower limit of the change in line transmission active power; Indicates the upper limit of the change in line transmission active power;

[0113] Line transmission power increment ΔP Lij Expressed as optimization variables:

[0114] ΔP Lij =V i0 V j0 [(-G ij sinθ ij0 +B ij cosθ ij0 )*Δθ i +(G ij sinθ ij0 -B ij cosθ ij0 )*Δθ j ]

[0115] Among them, G ij is the real part of the elements in the node admittance matrix, B ij is the imaginary part of the elements in the node admittance matrix, V i0 ,V j0 are the initial voltages of nodes i and j, respectively, ij0 is the initial phase angle difference between nodes i and j;

[0116] The equality constraints include the active power balance constraint of the node and the phase angle invariance constraint of the balanced node. The injected active power of the node is:

[0117]

[0118] By linearizing the above equation, we get the active power balance constraint, and the equality constraint of the linear programming problem is as follows:

[0119]

[0120] Δθ ref =0

[0121] Among them, V0, θ0 variables are the initial states of the network and are constants.

[0122] It should also be noted that it is calculated through the multi-objective relative dominance function, which reflects the degree of individual excellence. The larger the individual fitness value, the more it indicates that the individual is the optimal solution; the smaller the individual fitness value, the greater the gap between the individual and the optimal solution.

[0123] A multi-objective function with minimum network loss, maximum power supply capacity and the largest number of partitions is established. The multi-objective relative dominance function expression is:

[0124]

[0125] Among them, ΔP(xk ), P G (x k )、N(x k ) are respectively the optimal network loss, optimal power supply capacity and optimal number of partitions; min(ΔP), max(P G ), max(N) are the minimum network loss, the maximum power supply capacity, and the maximum number of partition schemes respectively; λ is the penalty factor.

[0126] S5: Repeat to form a new network and complete the calculation of the fitness values ​​of all random individuals in the parent population;

[0127] Randomly mutate the K individuals of the parent population to generate K new individuals of the child population;

[0128] Repeatedly form a new network and complete the calculation of the fitness values ​​of all random individuals in the offspring population;

[0129] The parent population and the offspring population are combined into a new population of 2K random individuals, and K individuals are selected according to the fitness value to form a new parent population, including 2 / 3K good individuals and 1 / 3K bad individuals;

[0130] The best individual in the current iteration is retained to determine whether all iterations are completed. When all iterations are completed, the best individual and solution are output. When all iterations are not completed, a new network is re-formed for the next iteration.

[0131] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

[0132] Embodiment 2, the second embodiment of the present invention, is different from the first two embodiments in that:

[0133] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, and other media that can store program codes.

[0134] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute instructions), or in conjunction with such instruction execution systems, devices or apparatuses. For the purposes of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in conjunction with such instruction execution systems, devices or apparatuses.

[0135] More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection with one or more wires (electronic device), a portable computer disk case (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be a paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, deciphering or, if necessary, processing in another suitable manner, and then stored in a computer memory.

[0136] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, a plurality of steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0137] Example 3, reference Figure 2 , which is the third embodiment of the present invention, and provides an electromagnetic ring network optimal de-ringing system based on adaptive evolutionary programming, comprising an initialization module 10, an adaptive mutation module 20, a network formation and power flow calculation module 30, a verification module 40, a multi-objective optimization module 50, a fitness value calculation module 60 and a population update module 70;

[0138] The initialization module 10 determines the initial combination scheme of the power source node and the load node according to the grid structure and operation characteristics, calculates the electrical distance, receives the grid structure data, and outputs the initial population to the adaptive mutation module 20;

[0139] The adaptive mutation module 20 performs adaptive random mutation on the initial population, generates a new partitioning scheme, receives the parent population data, and outputs the mutated child population to the network formation and power flow calculation module 30;

[0140] The network formation and power flow calculation module 30 receives the partitioning scheme, forms a new network structure, and performs power flow calculation to determine whether the power flow is converged, and performs security verification through the verification module 40;

[0141] The verification module 40 receives the power flow calculation results, performs N-1-1 verification and short-circuit current verification to ensure safe and stable operation of the power grid, outputs the verification results, and the qualified data enters the multi-objective optimization module 50 for optimization calculation;

[0142] The multi-objective optimization module 50 receives the verification results, calculates the optimal power supply capacity and active network loss of each partition scheme, forms the relative target of the multi-objective function, and evaluates the fitness of the individual through the fitness value calculation module 60;

[0143] The fitness value calculation module 60 receives the multi-objective optimization results, calculates the fitness value of the individual through the multi-objective relative dominance function, reflects the quality of the individual, and inputs it into the population update module 70 for population iteration;

[0144] The population updating module 70 selects excellent individuals to form a new parent population according to the fitness value, and retains the current best individual.

[0145] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. An optimal de-ringing method for electromagnetic ring network based on adaptive evolutionary programming, characterized by: include, According to the network structure and operation characteristics of the actual power grid, the initial combination scheme of power supply nodes and power grid load nodes is determined, and the electrical distances between power supply nodes and between power supply and load nodes are calculated; Take any power grid as an independent individual of the initial population, and perform adaptive random mutation on the independent individual to generate M partitioning schemes, where each partitioning scheme contains the number of partitions and the power nodes in each partition, forming the initial parent population; Based on the partitioning scheme of random individuals in the parent population, the combination scheme of power nodes and load nodes in each partitioning scheme is determined by the distance of electrical distance, and the connection lines between the partitions are disconnected. Each partition operates independently to form a new network; Perform power flow calculation on the new network and determine whether the power flow calculation converges. If the power flow converges, perform verification. Otherwise, use the constraint penalty function to assign a maximum number to the objective function and calculate the fitness value of the random individual. Perform N-1-1 check and short-circuit current check on the new network. If the check satisfies the constraint conditions, generate a relative target. Otherwise, use the constraint penalty function to assign a maximum number to the target function and calculate the fitness value of the random individual. Calculate the optimal power supply capacity and the optimal active network loss of random individuals in each partition scheme to form the relative objectives of the multi-objective function; Repeatedly form a new network to complete the calculation of the fitness values ​​of all random individuals in the parent population; Randomly mutate the K individuals of the parent population to generate K new individuals of the child population; Repeatedly form a new network and complete the calculation of the fitness values ​​of all random individuals in the offspring population; The parent population and the offspring population are combined into a new population of 2K random individuals, and K individuals are selected according to the fitness value to form a new parent population, including 2 / 3K good individuals and 1 / 3K bad individuals; The best individual in the current iteration is retained to determine whether all iterations are completed. When all iterations are completed, the best individual and solution are output. When all iterations are not completed, a new network is re-formed for the next iteration.

2. The optimal de-looping method for electromagnetic ring network based on adaptive evolutionary programming as claimed in claim 1, characterized in that: The electrical distance includes the voltage and reactive sensitivity between nodes. The connection between voltage and reactive regulation indicates the closeness of the connection between nodes. The element d in the electrical distance matrix D is ij for: Among them, B LL is the sensitivity matrix, which is replaced by the imaginary part of the node admittance matrix; the voltage / reactive sensitivity is used to represent the electrical distance between nodes; When partitioning the power nodes, extract the rows and columns corresponding to the source nodes in the matrix D and get the matrix D again. G When partitioning the load, the relative electrical distance is used to measure the distance between the load node and the power node: Extract the rows corresponding to the load nodes and the columns corresponding to the power supply nodes in the electrical distance matrix D to obtain the absolute electrical distance matrix D between the load nodes and the power supply nodes. hg , and then calculate the relative electrical distance l through simple transformation ij : Among them, min(D hg (i,:)) represents the matrix D hg The minimum value of the i-th row in , that is, the minimum distance between the load nodes in the i-th row.

3. The optimal de-looping method of an electromagnetic ring network based on adaptive evolutionary programming as claimed in claim 2, characterized in that: The adaptive random mutation includes: Assume the initial population size is K, the maximum number of partitions is k, and input the original data; Randomly generate a random number y of 0 or 1. When y=0, the current data is split and mutated; when y=1, the data is combined and mutated; Randomly select an element in each power node number sequence, split and mutate it into 2 elements, and at the same time k increases by 1; Randomly select two elements from each power node number sequence as the power number of a new partition, delete the previous two partition power numbers, and at the same time subtract 1 from k; Store the individuals that have completed the mutation into the offspring population; Determine whether all mutations are completed; When all mutations are completed, the process ends; when all mutations are not completed, the original data is re-entered to mutate the next individual.

4. The optimal de-looping method of an electromagnetic ring network based on adaptive evolutionary programming as claimed in claim 3, characterized in that: The forming of the new network includes determining the combination scheme of power supply nodes and load nodes in each partition scheme according to the electrical distance, partitioning the power supply and partitioning the load; The power partitioning includes finding D G The smallest non-diagonal element in the power partition, record the row and column corresponding to the smallest non-diagonal element, record them as i and j; check the number of power nodes P in the determined power partition i , find D G The smallest P in row i i-1 elements, as a partition; delete D G P i The rows and columns corresponding to the power nodes are deleted repeatedly until D G is an empty set; The load partitioning includes finding D hg The column j corresponding to the largest off-diagonal element in the last row; Determine whether the active capacity in the current partition is greater than the active load of the current load node. If so, move the current load to the power partition and delete the last row. If it does not exceed the active load of the current load node, remove the current element and search for D again. hg The column corresponding to the largest non-diagonal element in the last row of D is repeatedly checked to see whether the active capacity in the current partition is greater than the active load of the current load node until D hg Is an empty set.

5. The optimal de-looping method of electromagnetic ring network based on adaptive evolutionary programming as claimed in claim 4, characterized in that: The N-1-1 check and short-circuit current check include dividing the substations in the power grid into three functional stations, namely hub stations, transfer stations and terminal stations, and disconnecting the busbars and transformers of the hub stations; Conduct grid topology analysis and determine network structure; Use the Newton-Raphson method to calculate the power flow and determine whether the power flow calculation has converged. If it has converged, check whether it meets the N-1 index. If it has not converged, mark it as not passing the check. The verification determines whether the N-1 index is satisfied; if the N-1 index is satisfied, the verification is marked as passed; if the N-1 index is not satisfied, the verification is marked as failed; When the mark passes the verification, it is determined whether all the verifications are completed. When all the verifications are completed, the process ends; when all the verifications are not completed, the power grid topology analysis is repeated to perform power grid flow calculation.

6. The optimal de-looping method of electromagnetic ring network based on adaptive evolutionary programming as claimed in claim 5, characterized in that: The relative objectives of forming the multi-objective function include that the maximum power supply capacity continuously increases the system load during calculation, and when the line and transformer are about to exceed the power limit, the maximum power supply capacity is considered; When calculating the minimum active network loss, the total net injected power of all nodes in the whole network is recorded, expressed as node injected power, and solved using the linearized optimal power flow method: The optimized variables are the phase angle increment of the voltage at each node in the whole network and the active output increment of all generators, expressed as Δθ1, Δθ2…Δθ n , ΔP G1 ,ΔP G2 …ΔP Gn ; The objective function is to minimize the active network loss and is expressed as: The inequality constraints include the generator active output limit and the line transmission power limit, namely: Among them, ΔP Gi =P min -P G , represents the lower limit of the change in the active output of the generator; Indicates the upper limit of the change in the active output of the generator; ΔP Lij =P Lij,min -P Lij , represents the lower limit of the change in line transmission active power; Indicates the upper limit of the change in line transmission active power; Line transmission power increment ΔP Lij Expressed as optimization variables: ΔP Lij =V i0 V j0 [(-G ij sinθ ij0 +B ij cosθ ij0 )*Δθ i +(G ij sinθ ij0 -B ij cosθ ij0 )*Δθ j ] Among them, G ij is the real part of the elements in the node admittance matrix, B ij is the imaginary part of the elements in the node admittance matrix, V i0 ,V j0 are the initial voltages of nodes i and j, respectively, ij0 is the initial phase angle difference between nodes i and j; The equality constraints include the active power balance constraint of the node and the phase angle invariance constraint of the balanced node. The injected active power of the node is: By linearizing the above equation, we get the active power balance constraint, and the equality constraint of the linear programming problem is as follows: Dth ref =0 Among them, V0, θ0 variables are the initial states of the network and are constants.

7. The optimal de-looping method of electromagnetic ring network based on adaptive evolutionary programming as claimed in claim 6, characterized in that: The calculation of the fitness value of the random individual includes obtaining the fitness value by calculating through a multi-objective relative dominance function, which reflects the pros and cons of the individual. The larger the fitness value of the individual, the more the individual is the optimal solution; the smaller the fitness value of the individual, the greater the gap between the individual and the optimal solution. A multi-objective function with minimum network loss, maximum power supply capacity and the largest number of partitions is established. The multi-objective relative dominance function expression is: Among them, ΔP(x k ), P G (x k )、N(x k ) are respectively the optimal network loss, optimal power supply capacity and optimal number of partitions; min(ΔP), max(P G ), max(N) are the minimum network loss, the maximum power supply capacity, and the maximum number of partition schemes respectively; λ is the penalty factor.

8. A system using an optimal de-looping method for an electromagnetic ring network based on adaptive evolutionary programming as claimed in any one of claims 1 to 7, characterized in that: It includes initialization module, adaptive mutation module, network formation and power flow calculation module, verification module, multi-objective optimization module, fitness value calculation module and population update module; The initialization module determines the initial combination scheme of power nodes and load nodes according to the grid structure and operation characteristics, calculates the electrical distance, receives the grid structure data, and outputs the initial population to the adaptive mutation module; The adaptive mutation module performs adaptive random mutation on the initial population to generate a new partitioning scheme, receives the parent population data, and outputs the mutated child population to pass to the network formation and power flow calculation module; The network formation and power flow calculation module receives the partitioning scheme, forms a new network structure, performs power flow calculation, determines whether the power flow has converged, and performs security verification through the verification module; The verification module receives the power flow calculation results, performs N-1-1 verification and short-circuit current verification to ensure the safe and stable operation of the power grid, outputs the verification results, and the qualified data enters the multi-objective optimization module for optimization calculation; The multi-objective optimization module receives the verification results, calculates the optimal power supply capacity and active network loss of each partition scheme, forms the relative target of the multi-objective function, and evaluates the fitness of the individual through the fitness value calculation module; The fitness value calculation module receives the multi-objective optimization results, calculates the fitness value of the individual through the multi-objective relative dominance function, reflects the quality of the individual, and inputs it into the population update module for population iteration; The population updating module selects excellent individuals to form a new parent population according to the fitness value, and retains the current optimal individual.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the electromagnetic ring network optimal de-ringing method based on adaptive evolutionary programming according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of an optimal de-ringing method for an electromagnetic ring network based on adaptive evolutionary programming according to any one of claims 1 to 7 are implemented.

Images