Power distribution network harmonic monitoring device configuration method and system considering node disturbance sensitivity

Abstract

The present application relates to a kind of power distribution network harmonic monitoring device configuration method and system considering node disturbance sensitivity, construct traceability comprehensive error model, under the condition of harmonic source constraint, sensitivity concentration constraint, sensitivity dispersion constraint and monitoring device configuration quantity constraint, determine multiple first configuration scheme, sensitivity concentration constraint is obtained by the sensitivity matrix between each node;It is solved by binary particle swarm optimization algorithm, and second configuration scheme is obtained;Traceability comprehensive error is calculated by traceability comprehensive error model, and the second configuration scheme corresponding to minimum traceability comprehensive error is determined as optimal configuration scheme, system includes preliminary configuration scheme determination module, configuration scheme optimization solving module and optimal configuration scheme evaluation module.Compared with prior art, the present application reduces redundant monitoring node under the premise of guaranteeing harmonic information observability, realizes the reasonable balance between monitoring device quantity and monitoring capacity, to realize harmonic perception minimization configuration.

Description

Technical Field

[0001] This invention relates to the field of power technology, and in particular to a method and system for configuring a harmonic monitoring device for a distribution network that takes into account the sensitivity of node disturbances. Background Technology

[0002] With profound changes in the load and power supply structures of urban power grids, a large number of photovoltaic, energy storage, electric vehicle charging facilities, and various power electronic devices have been connected to the distribution network. This has led to a significant increase in nonlinear loads in the system, resulting in a continuous increase in the number and more complex distribution of harmonic sources within the distribution network. Harmonic monitoring is the foundation for harmonic source tracing and mitigation. However, due to the large number of nodes and wide coverage of the distribution network, achieving full-network harmonic status awareness through large-scale deployment of power quality monitoring devices would incur high construction and maintenance costs. Therefore, it is necessary to optimize the number and location of monitoring devices to effectively acquire harmonic information from the distribution network within a limited number of devices.

[0003] However, existing methods for configuring harmonic monitoring devices typically rely on prior information such as the known location of harmonic sources or the level of harmonic emission. In practical engineering applications, power grid companies often only have knowledge of the basic network structure and line impedance parameters of the distribution network when planning monitoring devices. The specific location, number, and level of harmonic sources within the system are usually difficult to obtain accurately. This makes it difficult for traditional methods to effectively generate representative harmonic samples and accurately characterize the harmonic correlation between nodes under the coupling effect of multiple harmonic sources. Furthermore, the configuration of monitoring devices can easily lead to problems such as overly concentrated local node deployment and insufficient overall coverage. This results in high redundancy and poor economic efficiency in the configuration of monitoring devices, making it difficult to meet the needs of harmonic monitoring and subsequent harmonic source tracing analysis in the distribution network.

[0004] Patent application CN118534245A discloses a harmonic monitoring method, a harmonic sensitivity analysis method, a device, and a computer. The method includes: constructing a distribution network model; adding harmonic sources to the nodes in the distribution network model and acquiring the harmonic currents of each node; the nodes include monitoring nodes and conventional nodes; calling a preset harmonic sensitivity analysis model to quantify the influence of the harmonic currents of each conventional node on the measured harmonic current values ​​of the monitoring nodes; based on the influence of the harmonic currents of each conventional node on the measured harmonic current values ​​of the monitoring nodes, obtaining the sensitivity constraints of the distribution network model and determining the optimal configuration model for harmonic monitoring; and obtaining the optimal harmonic monitoring nodes based on the optimal configuration model for harmonic monitoring and the line impedance data of the distribution network model, further obtaining the configuration scheme of the monitoring device. However, it still has shortcomings such as the reliance on neural network fitting sensitivity leading to an unintuitive physical mechanism, and the use of a random initialization strategy in the optimization process lacking explicit spatial dispersion constraints, making it difficult to ensure a globally optimal balance between the economic efficiency of the monitoring layout and the sensing effectiveness under complex operating conditions. Summary of the Invention

[0005] The purpose of this invention is to overcome the above-mentioned defects of the prior art and provide a method and system for configuring a harmonic monitoring device for a distribution network that takes into account both sensing capability and economy, optimizes resource allocation and reduces costs, while also achieving high monitoring accuracy and observability.

[0006] The objective of this invention can be achieved through the following technical solutions: The configuration method for harmonic monitoring devices in distribution networks that take into account node disturbance sensitivity includes: A comprehensive error model for source tracing is constructed, integrating harmonic current amplitude error, phase angle error, and harmonic source location deviation, to quantitatively evaluate the source tracing effectiveness of the entire network's harmonic status; The node harmonic current monitoring values ​​were processed by variance sensitivity analysis to construct a sensitivity matrix characterizing the degree of influence of disturbances between nodes; Candidate nodes are initially selected based on the number of node branches, and sensitivity concentration constraints and sensitivity dispersion constraints are established using the sensitivity matrix. Multiple first configuration schemes are determined under the constraints of sensitivity concentration constraints, sensitivity dispersion constraints, harmonic source constraints, and the number of monitoring devices. Using the first configuration scheme as the initial value, multiple second configuration schemes are obtained by iterative optimization using the binary particle swarm optimization algorithm; The comprehensive error of each second configuration scheme is calculated using the source tracing comprehensive error model, and the scheme with the minimum error is determined as the optimal configuration scheme.

[0007] Furthermore, the determination of multiple first configuration schemes for installing monitoring devices in the distribution network includes: Calculate the number of branches connected to each node in the distribution network, sort the branches from largest to smallest, and select nodes with more branches than the preset number of branches as candidate nodes. For each candidate node, under the constraints of sensitivity concentration, sensitivity dispersion, harmonic source, and number of monitoring devices, the node corresponding to the installation of the monitoring device is determined, and the node with the installed monitoring device is labeled with the first label, and the remaining nodes are labeled with the second label, thus obtaining the sub-configuration scheme; The source traceability comprehensive error of each sub-configuration scheme is calculated based on the source traceability comprehensive error model, and the sub-configuration scheme with the smallest source traceability comprehensive error is determined as the first configuration scheme.

[0008] Furthermore, the process of using the first configuration scheme as the initial value and iteratively optimizing using the binary particle swarm optimization algorithm to obtain multiple second configuration schemes includes: Based on the first configuration scheme, initialize the particle positions, and initialize the individual optimal position and the global optimal position; The particle's velocity and position are iteratively updated based on the current particle velocity, individual optimal position, and global optimal position. When the number of iterations reaches a preset number and the number of monitoring devices installed reaches a preset number, a second configuration scheme is obtained.

[0009] Furthermore, the source tracing comprehensive error model is expressed by the following calculation formula: in, For the first in the distribution network The overall harmonic source tracing error corresponding to the configuration scheme of each monitoring device This represents the total number of monitoring device configuration schemes in the power distribution network. The weight of the harmonic current amplitude error. For harmonic current amplitude error, The weight of the phase angle error of the harmonic current. This is for harmonic current phase angle error. The weight of the location deviation of the harmonic source. The location deviation of the harmonic source.

[0010] Furthermore, the sensitivity matrix is ​​determined by the following method: Calculate the harmonic current monitoring value of each node in the distribution network, where the harmonic current monitoring value is the sum of the harmonic currents generated at each node by all harmonic sources in the distribution network under the effect of grid impedance coupling; The harmonic current monitoring values ​​of each node are processed by the variance sensitivity analysis method to obtain the total variance and partial variance of the node, and the sub-sensitivity matrix of each node is calculated based on the total variance and partial variance. By integrating the sub-sensitivity matrices of multiple nodes, the sensitivity matrix between each node in the distribution network is obtained.

[0011] Furthermore, the sensitivity concentration constraint is obtained through the sensitivity matrix between various nodes in the distribution network, which is used to characterize the harmonic response relationship between nodes. The calculation formula is as follows: in, For the first in the distribution network The overall sensitivity of each node to all nodes in its neighborhood. For the first j The numbers of the adjacent nodes of each node. This is the overall sensitivity value of the node. For the firstj A set of node numbers in the neighboring region of a given node. For the first in the distribution network j Nodes in the set The sensitivity of each node within, and The specific values ​​are obtained through the sensitivity matrix. n The number of nodes connected to each branch. For the first in the distribution network j The number of branches connected to each node q This represents the column number in the sensitivity matrix. To determine the number of monitoring devices introduced, N To determine the number of nodes to be input.

[0012] Furthermore, the sensitivity dispersion constraint is used to avoid the monitoring devices being installed at too many nodes, resulting in uneven coverage of the distribution network nodes. Its calculation formula is as follows: in, For nodes in the distribution network k The set of node numbers in the neighboring region. For the set of all nodes in the distribution network, K This is the set of node numbers for which monitoring devices are installed in the power distribution network.

[0013] Furthermore, the harmonic source constraint is expressed by the following calculation formula: in, The error for tracing the source of the maximum harmonics in the distribution network. For the first in the distribution network The comprehensive harmonic source tracing error corresponding to the configuration scheme of each monitoring device; The constraint on the number of monitoring devices is expressed by the following formula: in, This represents the minimum number of monitoring devices required among all configuration options in the distribution network. For the maximum number of monitoring devices, For the first The number of monitoring devices in each configuration scheme.

[0014] Furthermore, the harmonic current monitoring value is expressed by the following calculation formula: in, For the first in the distribution network Harmonic current monitoring values ​​at each node For constant terms, For nodes in the distribution network For nodes The influence function of the harmonic current monitoring value, For nodes in the distribution network and nodes Under the combined effect of the nodes The influence function of the harmonic current monitoring value, For nodes in the distribution network Under the combined effect of the nodes The influence function of the harmonic current monitoring value, This represents the total number of nodes in the distribution network.

[0015] A system for configuring harmonic monitoring devices for distribution networks that takes into account node disturbance sensitivity includes: The source tracing comprehensive error model construction module is used to construct a source tracing comprehensive error model, which integrates harmonic current amplitude error, phase angle error, and harmonic source positioning deviation. The sensitivity matrix construction module is used to process the node harmonic current monitoring values ​​through variance sensitivity analysis and construct a sensitivity matrix that characterizes the degree of influence of disturbances between nodes. The preliminary configuration scheme determination module is used to initially select candidate nodes based on the number of node branches, and to establish sensitivity concentration constraints and sensitivity dispersion constraints using the sensitivity matrix. Under the constraints of sensitivity concentration constraints, sensitivity dispersion constraints, harmonic source constraints and the number of monitoring devices, multiple first configuration schemes are determined. The configuration scheme optimization and solution module is used to iteratively optimize multiple second configuration schemes using the binary particle swarm algorithm, with the first configuration scheme as the initial value. The optimal configuration scheme evaluation module is used to calculate the comprehensive error of each second configuration scheme using the source tracing comprehensive error model, and to determine the scheme with the minimum error as the optimal configuration scheme.

[0016] Compared with the prior art, the present invention has the following beneficial effects: 1. Breaking through the limitations of prior information and accurately characterizing harmonic response relationships: By constructing and introducing a sensitivity matrix, this invention can effectively characterize the harmonic response relationships and correlations between nodes under the coupling effect of multiple harmonic sources. This makes the evaluation of the contribution of node harmonic information more scientific and frees it from dependence on the exact location of power grid harmonic sources.

[0017] 2. Optimize device spatial layout and eliminate monitoring blind spots and local redundancy: This invention innovatively introduces "sensitivity concentration constraint" and "sensitivity dispersion constraint" simultaneously. The concentration constraint ensures that the device is preferentially arranged in the high-sensitivity area where the richest harmonic information can be obtained; the dispersion constraint effectively avoids excessive clustering of monitoring nodes and ensures the uniformity of node coverage across the entire network.

[0018] 3. Balancing monitoring effectiveness and construction costs to achieve the minimum configuration with "optimal cost-effectiveness": This invention not only pursues monitoring effectiveness but also introduces "constraints on the number of monitoring devices." While meeting technical observation requirements, it reduces redundant devices by selecting key monitoring nodes and choosing the minimum number of monitoring devices. This significantly reduces the basic construction investment and subsequent operation and maintenance costs of the monitoring system, achieving a reasonable balance between the number of monitoring devices and monitoring capabilities.

[0019] 4. Construct a multi-dimensional source tracing evaluation system to ensure final monitoring accuracy: This invention abandons the single-dimensional evaluation method and constructs a comprehensive source tracing error model that covers "harmonic current amplitude error", "harmonic current phase angle error" and "harmonic source positioning deviation". Using this model as the final judgment standard, the scheme with the smallest comprehensive error is selected, which fundamentally ensures the high accuracy and high reliability of this configuration scheme in actual power grid harmonic source tracing applications.

[0020] 5. Reduced computational complexity while maintaining engineering applicability and global optimality: This invention employs a two-stage efficient configuration strategy of 'candidate selection + optimization solution'. First, it utilizes network structure information such as branch connection relationships to initially select candidate nodes, narrowing the optimization space. Then, it uses these initial nodes as high-quality initial solutions to introduce a binary particle swarm optimization algorithm for iteration, effectively solving the 'curse of dimensionality' problem in optimizing large-scale distribution network configuration schemes. Attached Figure Description

[0021] Figure 1 This is a flowchart of the method of the present invention; Figure 2 This is a block diagram of the system modules of the present invention. Detailed Implementation

[0022] The present invention will now be described in detail with reference to the accompanying drawings and specific embodiments. These embodiments are based on the technical solution of the present invention and provide detailed implementation methods and specific operating procedures. However, the scope of protection of the present invention is not limited to the following embodiments.

[0023] Example 1 With the construction of new power systems and the integration of numerous power electronic devices into distribution networks, the proportion of nonlinear loads in these networks is continuously increasing, making harmonic problems increasingly prominent. Photovoltaic inverters, energy storage converters, electric vehicle charging devices, and various frequency converters can all generate harmonic currents of varying degrees during operation. These harmonics propagate and superimpose in the power grid, easily causing voltage distortion, power quality degradation, equipment overheating, and even protection malfunctions, adversely affecting the safe and stable operation of the distribution network. Therefore, real-time monitoring and analysis of distribution network harmonics is a crucial foundation for harmonic source tracing, harmonic mitigation, and power quality management.

[0024] In actual power grid operation, power quality monitoring devices are typically installed at key nodes to continuously monitor node voltage and current harmonics, thereby obtaining information on the system's harmonic operating status. Due to the large number and complex structure of distribution network nodes, installing monitoring devices at all nodes would result in high equipment investment and maintenance costs. Therefore, rationally determining the installation location and number of monitoring devices while ensuring monitoring effectiveness, and achieving optimized configuration of monitoring devices, is an important research topic in the current construction of distribution network harmonic monitoring systems.

[0025] Current research on optimizing the configuration of harmonic monitoring devices in distribution networks typically focuses on the distribution network topology, line parameters, and harmonic propagation characteristics. By establishing harmonic response or sensitivity relationships between nodes, monitoring nodes are selected to achieve effective observation of the system's harmonic state.

[0026] However, existing methods often focus on improving the observability of nodal harmonics or reducing the number of monitoring devices during the optimization process, but they do not adequately consider the correlation and redundancy information between different nodal harmonic responses. This can easily lead to situations where monitoring devices are deployed too centrally in some areas, resulting in high redundancy of monitoring information, while other areas have insufficient monitoring coverage. This results in low efficiency of monitoring resource utilization and increases system construction costs, making it difficult to balance monitoring effectiveness and economy.

[0027] In view of this, this embodiment, addressing the actual needs of harmonic source tracing in distribution networks, abandons the traditional single-dimensional evaluation method. The core idea is to construct a comprehensive source tracing error model that uniformly models harmonic current amplitude error, phase angle error, and harmonic source location deviation, using this as the global optimization objective. In the optimization path, a sensitivity matrix between nodes is constructed to characterize the physical propagation characteristics of harmonic disturbances, ensuring that the configuration process not only pursues optimal computational accuracy but also satisfies physical information richness and coverage uniformity.

[0028] This embodiment provides a method for configuring a harmonic monitoring device for a distribution network that takes into account the sensitivity of node disturbances, in order to solve the technical problems existing in the above-mentioned related technologies.

[0029] like Figure 1 As shown, Figure 1 This is a flowchart of the method of the present invention, see reference. Figure 1 The method includes; S1: Construct a comprehensive error model for tracing the source of the distribution network to quantify the comprehensive deviations in amplitude, phase angle, and positioning; S2: The monitoring values ​​of node harmonic currents are processed by variance sensitivity analysis to construct a sensitivity matrix characterizing the degree of influence of disturbances between nodes; S3: Initially select candidate nodes based on the number of branches connected to the node, and use the sensitivity matrix to extract the disturbance intensity and distribution characteristics of each node as sensitivity concentration constraints and sensitivity dispersion constraints, and then determine multiple first configuration schemes for installing monitoring devices in the distribution network under the constraints and quantity constraints. S4: Using the first configuration scheme as the initial value, multiple second configuration schemes are obtained by iterative optimization using the binary particle swarm algorithm; S5: Calculate the comprehensive error of each second configuration scheme using the source tracing comprehensive error model, and determine the scheme with the minimum error as the optimal configuration scheme.

[0030] Through the above technical solution, under the constraints of harmonic source constraints, sensitivity concentration constraints, sensitivity dispersion constraints, and the number of monitoring devices, based on the source tracing comprehensive error model, multiple first configuration schemes are determined with the goal of minimizing the source tracing comprehensive error. Among these first configuration schemes, a binary particle swarm optimization algorithm is used to process them, resulting in multiple second configuration schemes. The source tracing comprehensive error corresponding to each second configuration scheme is calculated, and the second configuration scheme with the minimum source tracing comprehensive error is determined as the optimal configuration scheme. Specifically, the sensitivity concentration constraint is obtained through the sensitivity matrix between various nodes in the distribution network, aiming to guide monitoring nodes to cluster in high-information areas to ensure that key disturbance characteristics are not lost. The sensitivity dispersion constraint acts as a check, limiting excessive local clustering and ensuring the uniformity of perception across the entire network. Together, they simulate the dual requirements of 'key monitoring' and 'global coverage' in engineering practice. This constraint mechanism based on the sensitivity matrix enables the configuration scheme to accurately characterize the harmonic response relationship between each node without needing to know the precise location of the harmonic source. This allows for the reduction of redundant monitoring nodes while ensuring the observability of harmonic information, achieving a reasonable balance between the number of monitoring devices and monitoring capabilities. Consequently, it improves the economy and effectiveness of the monitoring device configuration scheme and achieves a minimized configuration for harmonic sensing.

[0031] To enable those skilled in the art to better understand the spatiotemporal data prediction method provided in this embodiment, the above steps are illustrated in detail below.

[0032] For example, the source-tracing comprehensive error can be seen as the overall deviation between the network state inferred from monitoring data and the actual state. This error can be used to evaluate the quality of the monitoring device configuration scheme. The harmonic source constraint ensures that when selecting nodes for monitoring device installation in the distribution network, these nodes must preferentially or reasonably cover the locations of harmonic sources. The sensitivity concentration constraint prioritizes the placement of monitoring devices in areas of the distribution network that are most sensitive to harmonic changes and have the largest amount of information. The sensitivity dispersion constraint avoids excessive clustering of monitoring devices in the distribution network; they should be reasonably distributed throughout the network while ensuring sensitivity. The monitoring device quantity constraint ensures that the total number of nodes with monitoring devices in the distribution network cannot exceed the maximum allowed number for the project.

[0033] In this embodiment, a source tracing comprehensive error model is first constructed based on the topology of the distribution network. The source tracing comprehensive error model is expressed by the following calculation formula: in, For the first in the distribution network The overall harmonic source tracing error corresponding to the configuration scheme of each monitoring device This represents the total number of monitoring device configuration schemes in the power distribution network. The weight of the harmonic current amplitude error. For harmonic current amplitude error, The weight of the phase angle error of the harmonic current. This is for harmonic current phase angle error. The weight of the location deviation of the harmonic source. The location deviation of the harmonic source.

[0034] After installing monitoring devices in the distribution network, harmonic monitoring values ​​from multiple nodes can be obtained. Harmonic source tracing algorithms are then used to obtain source estimates, constructing a harmonic source tracing error index system. The estimated values ​​are then compared with the actual harmonic source values ​​set in the simulation. This index system includes harmonic current amplitude error. Harmonic current phase angle error Harmonic source positioning deviation .

[0035] Harmonic current amplitude error This can be expressed by the following calculation formula: in, This refers to the number of simulations required for harmonic source tracing when performing harmonic state estimation simulations on the distribution network topology. To determine the number of harmonic sources when performing harmonic state estimation simulations on the distribution network topology, Harmonic source In the The actual amplitude value during the second simulation. Harmonic source In the The amplitude estimate during the second simulation; Harmonic current phase angle error This can be expressed by the following calculation formula: in, In order to conduct the first step of analyzing the distribution network topology In the simulation of subharmonic state estimation, the maximum value of the phase angle range set for the harmonic source is determined. In order to conduct the first step of analyzing the distribution network topology In the simulation of subharmonic state estimation, the minimum value of the phase angle range set for the harmonic source is determined. Harmonic source In the The actual phase angle value during the second simulation. Harmonic source In the Phase angle estimates during the next simulation; Harmonic source positioning deviation This can be expressed by the following calculation formula: in, Harmonic source In the The results of the harmonic source tracing during the simulation are obtained by performing harmonic state estimation simulation on the distribution network topology in the simulation software.

[0036] Then, under the constraints of harmonic source constraints, sensitivity concentration constraints, sensitivity dispersion constraints, and the number of monitoring devices, multiple first configuration schemes for installing monitoring devices in the distribution network can be determined.

[0037] Harmonic source constraints are expressed by the following formula: in, The error for tracing the source of the maximum harmonics in the distribution network; During the installation of monitoring devices in a power distribution network, the greater the sensitivity of the node corresponding to the installed monitoring device to its neighboring nodes, the richer the harmonic information it can acquire. Therefore, the sum of the sensitivities of a monitoring node to its neighboring nodes can be calculated as an important reference indicator for the optimal configuration of monitoring points, i.e., applying a sensitivity concentration constraint to the nodes with installed monitoring devices. This sensitivity concentration constraint is expressed by the following formula: in, For the first in the distribution network The overall sensitivity of each node to all nodes in its neighborhood. For the first j The numbers of the adjacent nodes of each node. This is the overall sensitivity value of the node. For the first A set of node numbers in the neighboring region of a given node. For the first in the distribution network Nodes in the set The sensitivity of each node within, and The specific values ​​are obtained through the sensitivity matrix. The number of nodes connected to each branch. For the first in the distribution network The number of branches connected to each node This represents the column number in the sensitivity matrix. To determine the number of monitoring devices introduced, To determine the number of nodes to be input; During the installation of monitoring devices in the distribution network, to avoid overly concentrated installations at specific nodes and uneven network coverage, sensitivity dispersion constraints need to be applied to the installed monitoring devices. These sensitivity dispersion constraints are expressed by the following formula: in, For nodes in the distribution network The set of node numbers in the neighboring region. For the set of all nodes in the distribution network, This is the set of node numbers in the power distribution network where monitoring devices are installed. As the number of monitoring devices installed in the distribution network increases, the source tracing error decreases accordingly; however, when the number of installed monitoring devices reaches a certain level, the degree of error reduction may no longer be significant or prominent. Therefore, selecting the configuration scheme with the minimum number of monitoring devices while meeting technical constraints can achieve the optimal cost-effectiveness. The constraint on the number of monitoring devices is expressed by the following formula: in, This represents the minimum number of monitoring devices required among all configuration options in the distribution network. For the maximum number of monitoring devices, For the first The number of monitoring devices in each configuration scheme.

[0038] In this embodiment, after constructing the source tracing comprehensive error model, the goal is to minimize the source tracing comprehensive error. Under the constraints of harmonic source constraints, sensitivity concentration constraints, sensitivity dispersion constraints, and the number of monitoring devices, multiple first configuration schemes for installing monitoring devices in the distribution network can be determined.

[0039] The minimum comprehensive error in tracing the source can be expressed by the following formula: in, The first configuration option is... For the first in the distribution network The overall harmonic source tracing error corresponding to the configuration scheme of each monitoring device.

[0040] In one possible manner, the sensitivity matrix is ​​determined by the following method: Calculate the harmonic current monitoring value of each node in the distribution network, where the harmonic current monitoring value is the sum of the harmonic currents generated at each node by all harmonic sources in the distribution network under the effect of grid impedance coupling; The harmonic current monitoring values ​​of each node are processed by the variance sensitivity analysis method to obtain the total variance and partial variance of the node, and the sub-sensitivity matrix of each node is calculated based on the total variance and partial variance. By integrating the sub-sensitivity matrices of multiple nodes, the sensitivity matrix between each node in the distribution network is obtained.

[0041] It should be understood that, The specific values ​​are obtained through the sensitivity matrix, which is obtained by the following method: First, the harmonic current monitoring value of each node is calculated.

[0042] The harmonic current monitoring value is expressed by the following formula: in, For the first in the distribution network Harmonic current monitoring values ​​at each node For constant terms, For nodes in the distribution network For nodes The influence function of the harmonic current monitoring value, For nodes in the distribution network and nodes Under the combined effect of the nodes The influence function of the harmonic current monitoring value, For nodes in the distribution network Under the combined effect of the nodes The influence function of the harmonic current monitoring value, This represents the total number of nodes in the distribution network.

[0043] In the above formula, each sub-term is a dependent variable concerning the harmonic currents of the nodes it contains. Taking the sub-terms of harmonic current at a node as an example, for the node... Harmonic injection current Integrating, the following relationship exists: in, For nodes m Harmonic injection current, For nodes The maximum value of the harmonic source current is set.

[0044] Therefore, it can be concluded that the terms in the formula for calculating harmonic current monitoring values ​​are orthogonal, and thus the constant term can be obtained. As shown in the following formula: In the formula, For nodes The maximum value of the harmonic source current is set. The harmonic current content coefficient, For nodes h Rated current, For nodes N Harmonic injection current, The parameter domain is composed of the harmonic currents of each node. ; These are the harmonic currents at each node.

[0045] Furthermore, the influence functions of single-node and two-node harmonic sources in the formula for calculating harmonic current monitoring values ​​can be obtained. and As shown below: In the formula, and They represent the exceptions. 、 remove and Harmonic currents outside the nodes; and They represent the exceptions. ,remove and The maximum value of harmonic currents at nodes other than those in the main node. For nodes in the distribution network and nodes Under the combined effect of the nodes The influence function of the harmonic current monitoring value is given. Similarly, the influence relationships corresponding to different combinations of harmonic sources in the formula for calculating the harmonic current monitoring value can be obtained.

[0046] Variance-based sensitivity analysis can capture nonlinear relationships and complex interactions between harmonic sources, making it suitable for models containing multiple harmonic sources with nonlinear relationships and mutual influences. Therefore, the harmonic current monitoring value is defined. Total variance and partial variance They are respectively: In the formula, , The maximum value of the harmonic source current set for the node.

[0047] Harmonic current monitoring values The calculation formula is commutated and in the parameter domain Perform squaring and integration, and combine with the total variance. and partial variance The calculation formula can be obtained as follows: in, For nodes h harmonic current i h For nodes j Monitoring values y j The variance contribution, For nodes h and nodes k The combined effect of harmonic currents on y jThe variance contribution, For nodes 1, 2, ..., N - j The combined effect of harmonic currents on y j The variance contribution.

[0048] Define nodes Harmonic current monitoring values ​​for nodes Sensitivity of harmonic source current emission value The expression is as follows: In the formula, Node to be deleted In addition, the calculations obtained by other nodes Partial variance.

[0049] Sensitivity matrix Discretization representation: In the formula, For sample number, The total number of samples; The average value of the node harmonic current monitoring values ​​is the sample average. For the first Node under sample conditions The harmonic current monitoring value; For the first Except for nodes under each sample condition Nodes affected by the disturbance Harmonic current monitoring values.

[0050] pass , and The calculation formula yields the node. For nodes sensitivity ; Calculate the nodes respectively For nodes Sensitivity, construction Nodes of order Sensitivity Matrix .

[0051] For each node By analyzing each node individually, we can obtain the sensitivity matrix of each node, and then combine them to obtain... Node sensitivity matrix of order , The first in Line 1 Column elements represent nodes Harmonic current emission values ​​for nodes The degree of influence of harmonic current monitoring values; the larger the element value, the stronger the node. For nodes The more sensitive it is to harmonic disturbances. (Construction) This will support the optimization and configuration of subsequent monitoring devices.

[0052] In possible ways, the goal is to minimize the comprehensive error of source tracing, and under the constraints of harmonic source constraints, sensitivity concentration constraints, sensitivity dispersion constraints, and the number of monitoring devices, multiple first configuration schemes for installing monitoring devices in the distribution network are determined, including: Calculate the number of branches connected to each node in the distribution network, sort the branches from largest to smallest, and select nodes with more branches than the preset number of branches as candidate nodes. For each candidate node, under the constraints of sensitivity concentration, sensitivity dispersion, and the number of monitoring devices, the node corresponding to the installation of the monitoring device is determined, and the node with the installed monitoring device is labeled with the first label, and the remaining nodes are labeled with the second label, thus obtaining the sub-configuration scheme; The source traceability comprehensive error of each sub-configuration scheme is calculated based on the source traceability comprehensive error model, and the sub-configuration scheme with the smallest source traceability comprehensive error is determined as the first configuration scheme.

[0053] It should be understood that when determining the first configuration scheme, a sub-configuration scheme can first be determined based on the sensitivity concentration constraint, sensitivity dispersion constraint, and monitoring device configuration quantity constraint. Specifically, for each node in the distribution network, the total number of branches connected to each node is calculated. Then, the total number of branches is sorted in descending order, and nodes with a branch number greater than a preset branch number are selected as candidate nodes. For each candidate node, under the sensitivity concentration constraint, sensitivity dispersion constraint, and monitoring device configuration quantity constraint, the node where the monitoring device is installed is selected, and the installed node is labeled with a first label, while the remaining nodes are labeled with a second label. The first label can be a value of 1, and the second label can be a value of 0. This disclosure embodiment does not specifically limit this aspect.

[0054] Then, in the calculation of multiple sub-configuration schemes, the number and location of monitoring devices in each sub-configuration scheme are input into the source tracing comprehensive error model to obtain the source tracing comprehensive error. The multiple source tracing comprehensive errors are compared, and the sub-configuration scheme with the smallest source tracing comprehensive error is determined as the first configuration scheme.

[0055] For example, after obtaining the first configuration scheme, the binary particle swarm optimization (BPSO) algorithm can be used to solve for the second configuration scheme. To obtain the optimal configuration scheme for the monitoring device, the binary particle swarm optimization (BPSO) algorithm is used. BPSO is suitable for solving discrete binary monitoring device optimization problems. By setting particles as binary values ​​(0 or 1), each particle... Representing a potential configuration scheme, particles The number of elements is equal to the number of system nodes. The element value represents whether a monitoring device is installed at that node, with 1 indicating installation and 0 indicating no installation. By tracking the node... Optimal solution for each particle The optimal solution found so far for all particles To update the velocity, and thus update the particles The location.

[0056] In one possible manner, the step of using the first configuration scheme as an initial value and solving it using a binary particle swarm optimization algorithm to obtain the second configuration scheme includes: Based on the first configuration scheme, initialize the particle positions, and initialize the individual optimal position and the global optimal position; The particle's velocity and position are iteratively updated based on the current particle velocity, individual optimal position, and global optimal position. When the number of iterations reaches a preset number and the number of monitoring devices installed reaches a preset number, a second configuration scheme is obtained.

[0057] It should be understood that the specific steps in the implementation process are as follows: Initial value setting: Set the number of monitoring devices installed in the distribution network to [value]. Maximum number of monitoring devices Maximum number of iterations The initial positions of the particle swarm are randomly generated, and the number of each particle element is 1. Same; initial velocity is set as Learning factor and .

[0058] Setting the initial optimal population configuration: To accelerate convergence, the monitoring device configured for the optimal particle swarm optimization scheme provides more comprehensive coverage. The first label is set to 1, and the second label is set to 0. The resulting first configuration is then set as the optimal population configuration. ,and The number of elements 1 in the middle is equal to The number of elements 0 is equal to .

[0059] The particle element movement speed is updated using the following formula: In the formula, and Refers to the first Next and first t The particle corresponding to the next iteration Inner element The speed, i.e., the element Direction and magnitude of movement ; Particle In the t Elements in the next iteration u The corresponding value; A random number within the range [0,1]; Particle Elements in the optimal solution The corresponding value; Elements in the population-optimal solution The corresponding value.

[0060] Update particles The formula for updating inner element values ​​is as follows: The optimal solution is updated by updating the optimal solution of all particles during the iteration process to the population optimal solution. ;particle The optimal solution during the iteration process is updated to .

[0061] Output the optimal configuration scheme, if the number of iterations is... Repeat the above steps if , And the optimal solution Output the optimal solution while satisfying the constraints. The solution process ends; if If the constraints are not met, then Increment the quantity by 1, and increase the number of iterations. To reset, repeat the above steps; if Output the optimal solution The solution process is now complete.

[0062] For example, after solving each first configuration scheme, the corresponding second configuration scheme is obtained. Among multiple second configuration schemes, the traceability comprehensive error of each second configuration scheme can be calculated by the traceability comprehensive error model. By comparing multiple traceability comprehensive errors, the second configuration scheme with the smallest traceability comprehensive error can be determined as the optimal configuration scheme.

[0063] The above technical solution analyzes the harmonic response relationships and correlations between different nodes, evaluates the harmonic information contribution of each node, and selects key monitoring nodes while ensuring the observability of system harmonics. This reduces redundant monitoring devices, improves the efficiency of monitoring information utilization, and achieves a reasonable balance between the number of monitoring devices and monitoring capabilities. A minimum configuration model for distribution network harmonic monitoring devices that balances sensing capabilities and economy is constructed. The optimization process comprehensively considers the observability of harmonic information and the configuration cost of monitoring devices. An optimization algorithm determines the optimal installation location and number of monitoring devices, thereby improving the harmonic state sensing capability of the distribution network under limited monitoring resources and reducing the construction and operation and maintenance costs of the monitoring system.

[0064] Example 2 This embodiment provides a system for configuring a harmonic monitoring device for a distribution network that takes into account the sensitivity of node disturbances, such as... Figure 2 As shown, it includes the following modules: The source tracing comprehensive error model construction module 1 is used to construct the source tracing comprehensive error model, which integrates harmonic current amplitude error, phase angle error and harmonic source positioning deviation, and is used to quantitatively evaluate the source tracing effectiveness of the entire network harmonic status. Sensitivity matrix construction module 2 is used to process the node harmonic current monitoring values ​​through variance sensitivity analysis and construct a sensitivity matrix that characterizes the degree of influence of disturbances between nodes. The preliminary configuration scheme determination module 3 is used to initially select candidate nodes by combining the number of node branches, and to establish sensitivity concentration constraints and sensitivity dispersion constraints using the sensitivity matrix. Under the constraints of sensitivity concentration constraints, sensitivity dispersion constraints, harmonic source constraints and the number of monitoring devices, multiple first configuration schemes are determined. Configuration scheme optimization and solution module 4 is used to obtain multiple second configuration schemes by iteratively optimizing the first configuration scheme as the initial value using the binary particle swarm algorithm. The optimal configuration scheme evaluation module 5 is used to calculate the comprehensive error of each second configuration scheme using the source tracing comprehensive error model, and to determine the scheme with the minimum error as the optimal configuration scheme.

[0065] The rest is the same as in Example 1.

[0066] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for configuring harmonic monitoring devices for distribution networks considering node disturbance sensitivity, characterized in that, include: A comprehensive error model for source tracing is constructed, integrating harmonic current amplitude error, phase angle error, and harmonic source location deviation, to quantitatively evaluate the source tracing effectiveness of the entire network's harmonic status; The node harmonic current monitoring values ​​were processed by variance sensitivity analysis to construct a sensitivity matrix characterizing the degree of influence of disturbances between nodes; Candidate nodes are initially selected based on the number of node branches, and sensitivity concentration constraints and sensitivity dispersion constraints are established using the sensitivity matrix. Multiple first configuration schemes are determined under the constraints of sensitivity concentration constraints, sensitivity dispersion constraints, harmonic source constraints, and the number of monitoring devices. Using the first configuration scheme as the initial value, multiple second configuration schemes are obtained by iterative optimization using the binary particle swarm optimization algorithm; The comprehensive error of each second configuration scheme is calculated using the source tracing comprehensive error model, and the scheme with the minimum error is determined as the optimal configuration scheme.

2. The method for configuring a harmonic monitoring device for a distribution network considering node disturbance sensitivity according to claim 1, characterized in that, The determination of multiple first configuration schemes includes: Calculate the number of branches connected to each node in the distribution network, sort the branches from largest to smallest, and select nodes with more branches than the preset number of branches as candidate nodes. For each candidate node, under the constraints of sensitivity concentration, sensitivity dispersion, harmonic source, and number of monitoring devices, the node corresponding to the installation of the monitoring device is determined, and the node with the installed monitoring device is labeled with the first label, and the remaining nodes are labeled with the second label, thus obtaining the sub-configuration scheme; The source traceability comprehensive error of each sub-configuration scheme is calculated based on the source traceability comprehensive error model, and the sub-configuration scheme with the smallest source traceability comprehensive error is determined as the first configuration scheme.

3. The method for configuring a distribution network harmonic monitoring device considering node disturbance sensitivity according to claim 1, characterized in that, The process of using the first configuration scheme as the initial value and iteratively optimizing it using the binary particle swarm optimization algorithm to obtain multiple second configuration schemes includes: Based on the first configuration scheme, initialize the particle positions, and initialize the individual optimal position and the global optimal position; The particle's velocity and position are iteratively updated based on the current particle velocity, individual optimal position, and global optimal position. When the number of iterations reaches a preset number and the number of monitoring devices installed reaches a preset number, a second configuration scheme is obtained.

4. The method for configuring a harmonic monitoring device for a distribution network considering node disturbance sensitivity according to claim 1, characterized in that, The source tracing comprehensive error model is expressed by the following calculation formula: in, For the first in the distribution network The overall harmonic source tracing error corresponding to the configuration scheme of each monitoring device This represents the total number of monitoring device configuration schemes in the power distribution network. The weight of the harmonic current amplitude error. For harmonic current amplitude error, The weight of the phase angle error of the harmonic current. This is for harmonic current phase angle error. The weight of the location deviation of the harmonic source. The location deviation of the harmonic source.

5. The method for configuring a distribution network harmonic monitoring device considering node disturbance sensitivity according to claim 1, characterized in that, The sensitivity matrix was determined using the following method: Calculate the harmonic current monitoring value of each node in the distribution network, where the harmonic current monitoring value is the sum of the harmonic currents generated at each node by all harmonic sources in the distribution network under the effect of grid impedance coupling; The harmonic current monitoring values ​​of each node are processed by the variance sensitivity analysis method to obtain the total variance and partial variance of the node, and the sub-sensitivity matrix of each node is calculated based on the total variance and partial variance. By integrating the sub-sensitivity matrices of multiple nodes, the sensitivity matrix between each node in the distribution network is obtained.

6. The method for configuring a harmonic monitoring device for a distribution network considering node disturbance sensitivity according to claim 1, characterized in that, The sensitivity concentration constraint is obtained through the sensitivity matrix between various nodes in the distribution network, and is used to characterize the harmonic response relationship between nodes. The calculation formula is as follows: in, For the first in the distribution network The overall sensitivity of each node to all nodes in its neighborhood. For the first j The numbers of the adjacent nodes of each node. This is the overall sensitivity value of the node. For the first j A set of node numbers in the neighboring region of a given node. For the first in the distribution network j Nodes in the set The sensitivity of each node within, and The specific values ​​are obtained through the sensitivity matrix. n The number of nodes connected to each branch. For the first in the distribution network j The number of branches connected to each node q This represents the column number in the sensitivity matrix. To determine the number of monitoring devices introduced, N To determine the number of nodes to be input.

7. The method for configuring a distribution network harmonic monitoring device considering node disturbance sensitivity according to claim 1, characterized in that, The sensitivity dispersion constraint is used to avoid the monitoring devices being installed at too many nodes, resulting in uneven coverage of the distribution network nodes. Its calculation formula is as follows: in, For nodes in the distribution network k The set of node numbers in the neighboring region. For the set of all nodes in the distribution network, K This is the set of node numbers for which monitoring devices are installed in the power distribution network.

8. The method for configuring a harmonic monitoring device for a distribution network considering node disturbance sensitivity according to claim 1, characterized in that, The harmonic source constraint is expressed by the following formula: in, The error for tracing the source of the maximum harmonics in the distribution network. For the first in the distribution network The comprehensive harmonic source tracing error corresponding to the configuration scheme of each monitoring device; The constraint on the number of monitoring devices is expressed by the following formula: in, This represents the minimum number of monitoring devices required among all configuration options in the distribution network. For the maximum number of monitoring devices, For the first The number of monitoring devices in each configuration scheme.

9. The method for configuring a distribution network harmonic monitoring device considering node disturbance sensitivity according to claim 5, characterized in that, The harmonic current monitoring value is expressed by the following formula: in, For the first in the distribution network Harmonic current monitoring values ​​at each node For constant terms, For nodes in the distribution network For nodes The influence function of the harmonic current monitoring value, For nodes in the distribution network and nodes Under the combined effect of the nodes The influence function of the harmonic current monitoring value, For nodes in the distribution network Under the combined effect of the nodes The influence function of the harmonic current monitoring value, This represents the total number of nodes in the distribution network.

10. A system for configuring harmonic monitoring devices for distribution networks that takes into account node disturbance sensitivity, characterized in that, include: The source tracing comprehensive error model construction module is used to construct a source tracing comprehensive error model, which integrates harmonic current amplitude error, phase angle error, and harmonic source positioning deviation. The sensitivity matrix construction module is used to process the node harmonic current monitoring values ​​through variance sensitivity analysis and construct a sensitivity matrix that characterizes the degree of influence of disturbances between nodes. The preliminary configuration scheme determination module is used to initially select candidate nodes based on the number of node branches, and to establish sensitivity concentration constraints and sensitivity dispersion constraints using the sensitivity matrix. Under the constraints of sensitivity concentration constraints, sensitivity dispersion constraints, harmonic source constraints and the number of monitoring devices, multiple first configuration schemes are determined. The configuration scheme optimization and solution module is used to iteratively optimize multiple second configuration schemes using the binary particle swarm algorithm, with the first configuration scheme as the initial value. The optimal configuration scheme evaluation module is used to calculate the comprehensive error of each second configuration scheme using the source tracing comprehensive error model, and to determine the scheme with the minimum error as the optimal configuration scheme.

Images