Power distribution network cluster division method and system considering uncertainty and comprehensive performance

Abstract

The invention discloses a power distribution network cluster division method and system considering uncertainty and comprehensive performance, and belongs to the technical field of power system planning and optimization, and the method comprises the steps: employing probability distribution to describe distributed energy, load demands and distribution microgrid interaction power randomness features contained in a power distribution network, and generating a plurality of scenes; at least one of the structural indexes and the functional indexes representing the cluster performance indexes is selected, the multiple cluster performance indexes are fused to form a cluster comprehensive performance index, and the cluster comprehensive performance index serves as a power distribution network cluster division model considering the cluster comprehensive performance; carrying out load flow calculation on the power distribution network according to various scenes representing uncertainty, and substituting a load flow calculation result into the power distribution network cluster division model considering cluster comprehensive performance; and solving the power distribution network cluster division model to obtain a power distribution network cluster division result. According to the invention, the autonomous capability of the cluster is improved from the interior of the cluster, and the problems of serious network loss, voltage out-of-limit and the like caused by large-scale transmission of electric energy are reduced.

Description

Technical Field

[0001] The present invention belongs to the technical field of power system planning and optimization, and relates to a method for dividing a multi-microgrid active distribution network cluster suitable for distribution network planning and operation analysis, and more specifically, to a method and system for dividing a multi-microgrid active distribution network cluster taking into account uncertainty and comprehensive cluster performance. Background Art

[0002] With the continuous advancement of the construction of new power systems, the characteristics of distribution networks are gradually changing from passive to active, unidirectional to bidirectional, and deterministic to random. Its planning and operation goals have also evolved from improving power supply security capabilities to promoting multi-energy integration, complementarity, and multi-aggregation interaction. At the same time, the high proportion of renewable energy and multiple microgrids connected to the active distribution network have further aggravated the complexity and uncertainty of the system, bringing many challenges to the safe and economic operation of the distribution network, such as changes in power flow distribution, power reverse transmission, increased system losses, and node voltage exceeding the limit.

[0003] In a distribution network system containing multiple microgrids, dividing and integrating access units such as distributed power sources, loads, and microgrids to form clusters with dispatching control capabilities and autonomous capabilities is a solution to achieve and ensure the orderly, reliable, and efficient access of large-scale distributed power sources, microgrids, etc. to the distribution network. Therefore, clustering the active distribution network containing multiple microgrids can reasonably reorganize the distributed resources, microgrids, loads, and other resources within the cluster, forming strong coupling within the cluster and weak coupling between clusters, so that the clusters can operate independently and coordinate with each other. In addition, the cluster division results can provide an important reference for solving the problem of high-proportion new energy consumption, reactive power shortage and voltage crossing caused by grid connection, as well as the site selection and capacity planning of energy storage systems and reactive power compensation equipment, and play a guiding role in the active-reactive planning of active distribution networks containing multiple microgrids.

[0004] In the cluster division of active distribution networks containing multiple microgrids, cluster division indicators are crucial, which directly determine the optimization effect of the cluster, and then affect the operation efficiency, stability and implementation effect of active-reactive coordination planning of the distribution network. Cluster division indicators are generally divided into two categories, namely structural and functional indicators. Structural indicators focus on the coupling relationship in geographical space or electrical distance, that is, the node relationship within the cluster is close, while the relationship between clusters is sparse; while functional indicators focus on the collaboration ability of each node within the cluster, that is, reflecting the regulation purpose of the cluster.

[0005] Generally speaking, each cluster after cluster division needs to meet the following principles: (1) Logical principle, that is, there are no overlapping nodes between clusters, and each node in the cluster must meet the physical connectivity requirements; (2)Under the division of functional indicators, each cluster must have a certain self - coordination ability. The existing cluster indicators consider the comprehensive combination of structural indicators and functional indicators including active / reactive power balance, but fail to take into account the flexibility supply - demand level within the cluster, that is, the flexibility supply - demand matching degree index is not considered, and the impact of multi - microgrids on the active distribution network is not fully considered during cluster division.

[0006] In addition, during cluster division, the output values of wind power and photovoltaic power, load, and the interactive power values of the distribution microgrid under typical daily scenarios are usually selected, ignoring the inherent uncertainty of these factors. However, compared with conventional energy sources, the output of new energy sources such as wind power and photovoltaic power and the load side have significant randomness and volatility; for the distribution network with multi - microgrids connected, the interactive power between the distribution network and the microgrid also has uncertainty. Therefore, it is particularly important to analyze and handle the uncertainty of wind power, photovoltaic power, and the interactive power between the microgrid and the distribution network during cluster division.

[0007] In summary, there is an urgent need to provide a method and system for dividing clusters of active distribution networks with multi - microgrids that take into account uncertainty and cluster comprehensive performance. Summary of the Invention

[0008] To solve the deficiencies in the existing technology, the present invention provides a method and system for dividing clusters of active distribution networks with multi - microgrids that take into account uncertainty and cluster comprehensive performance.

[0009] The present invention adopts the following technical solutions.

[0010] In the first aspect of the present invention, a method for dividing clusters of a distribution network that takes into account uncertainty and comprehensive performance is provided. The distribution network is an active distribution network set with multi - microgrids, and the method for dividing clusters of the distribution network includes the following steps: Use probability distributions to characterize the stochastic characteristics of distributed energy sources, load demands, and distribution - microgrid interactive power included in the distribution network, generating multiple scenarios to represent the uncertainty of the active distribution network with multi - microgrids; Select at least one from the structural indicators and functional indicators that characterize the cluster performance indicators, and fuse multiple cluster performance indicators to form a cluster comprehensive performance indicator as a model for dividing clusters of the distribution network that takes into account cluster comprehensive performance; For multiple scenarios representing uncertainty, perform power flow calculations on the distribution network, and substitute the power flow calculation results into the model for dividing clusters of the distribution network that takes into account cluster comprehensive performance; Solve the model for dividing clusters of the distribution network to obtain the result of dividing clusters of the distribution network.

[0011] Preferably, the representation of the uncertainty of the active distribution network with multi - microgrids specifically includes: Determine the probability distribution models of distributed energy sources, load demands, and distribution - microgrid interactive power; Randomly sample the probability distribution models of distributed energy, load demand, and interactive power of the distribution and microgrid to generate a set number of scenarios representing different operating states. Cluster the scenarios representing different operating states to reduce the set number to multiple typical representative scenarios.

[0012] Preferably, the representation of the uncertainties of the active distribution network with multiple microgrids specifically includes: Determine the probability distribution models of wind power, photovoltaic power, load demand, and interactive power of the distribution and microgrid by analyzing historical data or based on physical models. Use the Monte Carlo method to randomly sample the probability distribution models to generate scenarios with different sets of approved numbers. Apply the K-means clustering algorithm to cluster the generated Monte Carlo scenarios with a set number of groups. According to the results of the K-means clustering, select the representative cluster centers as the final scenario set.

[0013] Preferably, the selection of at least one from the structural indicators and functional indicators representing the cluster performance indicators includes: Select the modularity index as the structural indicator, and select at least one of the active power balance degree, reactive power balance degree, and flexibility supply-demand matching degree index as the functional indicator.

[0014] Preferably, based on the modularity index of the electrical distance of the constructed active and reactive power sensitivities, the active and reactive power balance degree indices, and the flexibility supply-demand matching degree index, establish the objective function of the distribution network cluster division model considering the comprehensive cluster performance according to the weight fusion.

[0015] Preferably, the construction of the modularity index of the electrical distance of the active and reactive power sensitivities includes: calculating the active power voltage sensitivity matrix and the reactive power voltage sensitivity matrix, using the elements of the sensitivity matrix to calculate the electrical distance between each node in the distribution network, constructing the electrical distance matrix, obtaining the maximum value as the edge weight connecting two nodes, and calculating the modularity ρ based on the edge weight as the modularity index. The construction of the active and reactive power balance degree indices includes: establishing the active power balance degree index based on the summation of the net loads of each cluster, and establishing the reactive power balance degree index based on the ratio of the maximum value and the demand value of the reactive power supply within the cluster. The construction of the flexibility supply-demand matching degree index includes: calculating the cluster flexibility demand value based on the upward and downward flexibility demand values of the cluster, and establishing the flexibility supply-demand matching degree index.

[0016] Preferably, the fusion of multiple cluster performance indicators to form the cluster comprehensive performance indicator, as the distribution network cluster division model considering the cluster comprehensive performance, includes: The modularity index, active power balance index, reactive power balance index, and flexibility supply-demand matching index are multiplied by their respective weights and then summed, with the maximum sum as the objective function. Among them, the sum of the weights of the modularity index, active power balance index, reactive power balance index, and flexibility supply-demand matching index is 1.

[0017] Preferably, the solution of the distribution network cluster division model to obtain the distribution network cluster division result includes: Input the power source type, output, and load conditions of each node, as well as the flexibility demand response situation into the model, encapsulate various cluster division methods as a population, and solve it using the genetic algorithm to obtain the distribution network cluster division result.

[0018] Preferably, the distribution network cluster division method further includes: Analyze the result of cluster division, verify the effectiveness of the proposed cluster division method, and use the scheme that passes the verification as the cluster division scheme; The analysis of the result of cluster division includes: partially or completely setting the weight coefficients of each index to zero, and combining the new energy penetration rate to judge the operation state of the distribution network system according to the value of the comprehensive performance index.

[0019] The second aspect of the present invention provides a distribution network cluster division system considering uncertainty and comprehensive performance, which operates according to the distribution network cluster division method considering uncertainty and comprehensive performance described in the first aspect, including: An uncertainty processing module, which is used to characterize the randomness characteristics of distributed energy, load demand, and distribution network-microgrid interaction power included in the distribution network using probability distributions, and generate multiple scenarios representing the uncertainty of the active distribution network with multiple microgrids; A power flow calculation module, which is used to perform power flow calculations on the distribution network for multiple scenarios representing uncertainty; A cluster comprehensive performance evaluation module, which is used to select at least one from the structural index and functional index representing the cluster performance index, and fuse multiple cluster performance indexes to form a cluster comprehensive performance index, and form a distribution network cluster division model; A solver, which is used to solve the distribution network cluster division model to obtain the distribution network cluster division result.

[0020] Compared with the prior art, the beneficial effects of the present invention at least include: The method and system for clustering the active distribution network with multiple microgrids of the present invention fully consider the dominant factors of uncertainty in the active distribution network with multiple microgrids compared with the usual clustering ideas, and propose a method for dealing with its uncertainty. In addition, the present invention proposes a comprehensive index considering the flexibility of supply-demand matching degree, which gives full play to the structural, active and reactive coordination capabilities and flexibility response capabilities within the cluster, helps to improve its autonomous ability from within the cluster, and reduces problems such as serious network losses and voltage over-limit caused by large-scale power transmission of electric energy, and can provide a theoretical basis and technical support for the active-reactive power planning of the active distribution network with multiple microgrids. Description of the Drawings

[0021] Figure 1 Schematic flow chart of the method for clustering the active distribution network with multiple microgrids considering uncertainty and cluster comprehensive performance proposed by the present invention; Figure 2 Typical structure in the radial distribution network adopted by the present invention; Figure 3 Improved IEEE33 node system; Figure 4 Typical daily scenario data of wind power, photovoltaic power, load and interactive power of the distribution microgrid; Figure 5 Voltage fluctuation data 24 hours before the access of photovoltaic power and wind power under a certain typical daily scenario; Figure 6 Voltage fluctuation data 24 hours after the access of photovoltaic power and wind power under a certain typical daily scenario; Figure 7 Clustering situation of Mode 1; Figure 8 Clustering situation of Mode 2; Figure 9 Clustering situation of Mode 3; Figure 10 Clustering situation of Mode 4; Figure 11 Clustering situation with the number of clusters of Mode 4 being 4; Figure 12 Clustering situation under the new energy penetration rate of 20%; Figure 13 Clustering situation under the new energy penetration rate of 60%. Detailed Implementation Manner

[0022] To make the objectives, technical solutions and advantages of the present invention clearer, the following will clearly and completely describe the technical solutions of the present invention in conjunction with the accompanying drawings in the embodiments of the present invention. The described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the spirit of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the protection scope of the present invention.

[0023] As Figure 1 shown, Embodiment 1 of the present invention provides a method for partitioning an active distribution network cluster with multiple microgrids considering uncertainty and cluster comprehensive performance, including the following steps: Step 1: Process the uncertainty of photovoltaic power, wind power, load, and the interactive power of distribution microgrids to obtain their final scenario power data.

[0024] Preferably but not restrictively, Step 1 specifically includes: Step 1.1, use probability distributions to characterize the stochastic characteristics of distributed energy, load demand, and the interactive power of distribution microgrids, which is used to represent the uncertainty of the active distribution network with multiple microgrids. Preferably but not limited to, by analyzing historical data or based on physical models, determine the probability distribution models of wind power, photovoltaic power, load demand, and the interactive power of distribution microgrids. For characterizing the randomness of wind power, the Weibull distribution is often used to simulate the randomness of wind speed, which is expressed by the following formula:

[0025] In the formula: v is the wind speed; λ is the scale parameter, which controls the expansion range of the distribution; k is the shape parameter, which determines the shape of the distribution.

[0026] The characteristics of the Weibull distribution can well describe the asymmetry of wind speed. The wind speed range is 0~35 (m / s), the scale parameter is 15, and the shape parameter is 2.3.

[0027] For characterizing the randomness of photovoltaic power, the Beta distribution is generally used for modeling. Photovoltaic irradiance (i.e., solar light intensity) is usually considered as a proportional value or probability value between 0 and 1. The Beta distribution can be used to estimate the probability density function at different irradiance levels, so as to understand the distribution law and characteristics of photovoltaic irradiance, and provide support for the design and operation of photovoltaic power generation systems. Specifically, the probability density function of the Beta distribution is expressed by the following formula:

[0028] In the formula: B(α,β) is the Beta function, which is a normalization constant that ensures the total probability of the distribution is 1.

[0029] For the characterization of load demand and the randomness of the interactive power between the distribution network and the microgrid, the normal distribution is generally adopted. The adoption of the normal distribution to characterize the randomness of power load and interactive power is based on multiple practical reasons and theoretical bases. The changes in power load and the interactive power between the distribution network and the microgrid are the result of the combined action of multiple factors, which are essentially independent and have a superimposed effect, making the load data usually approximable by the normal distribution.

[0030] It can be understood that the normal distribution, also known as the Gaussian distribution, is one of the most important distributions in statistics and has the characteristics of a bell-shaped curve. The probability density function of the normal distribution is expressed by the following formula:

[0031] In the formula: x is a random variable; μ is the mean; σ is the standard deviation.

[0032] Step 1.2: Use the Monte Carlo method to perform a large number of random samplings on the above probability distribution to generate 500 different scenarios. Each sampling scenario represents a possible system operating state, including the output of photovoltaic and wind power, load demand, and the scenario data of the interactive power between the microgrid and the distribution network.

[0033] Step 1.3: Apply the K-means clustering algorithm to cluster the generated 500 Monte Carlo scenarios, aiming to aggregate a large number of random scenarios into several representative scenarios. Each cluster represents a class of similar scenarios, such as similar wind speeds, light intensities, load demands, and interactive powers, etc.

[0034] Step 1.4: According to the results of the K-means clustering, select the representative cluster centers as the final scenario set. Calculate the probability of each scenario cluster, and based on the scenario data and scenario probability, obtain the final photovoltaic, wind power, load, and interactive power data between the distribution network and the microgrid.

[0035] As one of the prominent substantial features of the present invention, the combination of the probability distribution models of wind power, photovoltaic, load demand, and the interactive power between the distribution network and the microgrid with the Monte Carlo method and the K-means clustering algorithm for scenario generation and reduction is an effective method for dealing with the dominant factors of uncertainty.

[0036] Specifically, the probability model can effectively describe the stochastic characteristics of distributed energy, load, and interactive power. The Monte Carlo method generates different scenarios through a large number of random samplings to simulate various possible changes of uncertain factors. The K-means clustering algorithm is used to extract representative scenarios from the generated massive scenarios, thereby reducing the computational complexity and improving the optimization efficiency.

[0037] Therefore, as one of the significant improvements brought by the present invention to the prior art, the provided means for handling uncertainty can effectively reduce the consumption of computing resources while ensuring that uncertainty is fully considered, and is particularly suitable for the scheduling and planning problems of complex systems. Especially in the context of multi-microgrids and high-proportion renewable energy access, it can better cope with the uncertainty and volatility of the system.

[0038] Step 2: Perform power flow calculation on the active distribution network with multi-microgrids according to the obtained data and system parameters to obtain the basic data required for cluster division.

[0039] Step 3: Select at least one index representing cluster performance from the structural index and the functional index, and fuse multiple cluster performance indexes to form a comprehensive cluster performance index. Based on the comprehensive cluster performance index, establish a distribution network cluster division model. Preferably but not limited to, establish a distribution network cluster division model with modularity, active / reactive power balance degree, and flexibility supply-demand matching degree as the comprehensive cluster performance index.

[0040] Preferably but not restrictively, Step 3 specifically includes: Step 3.1: Construct modularity index, active and reactive power balance degree index, and flexibility supply-demand matching degree index respectively.

[0041] Further preferably but not restrictively, Step 3.1 specifically includes: A. Construct a modularity index based on the electrical distance of active and reactive power sensitivities.

[0042] Use the modularity index based on electrical distance as the structural index. The larger the modularity value, the closer the internal connection of the cluster. Apply the modularity proposed by Girvan and Newman here. Different from traditional community division techniques, it is not necessary to determine the number of communities in advance to quantify the strength of the community structure in the network and can solve the problems of complex networks in community division. Use an improved modularity function with electrical distance as the weight to characterize the electrical coupling degree between nodes in the distribution network, which is expressed by the following formula:

[0043] In the formula: e ijis the weight of the edge connecting node i and node j, abbreviated as edge weight, which refers to the electrical distance in the present invention; , which is the sum of all edge weights in the network; represents the sum of all edge weights connected to node i; When node i and node j are in the same cluster , otherwise .

[0044] The electrical distance is used to measure the tightness of electrical coupling between two nodes in the network, and is generally obtained through the sensitivity relationship between voltage and active and reactive power. Specifically, the power flow calculation correction equation in the polar coordinates of the distribution network is expressed by the following formula:

[0045] In the formula: ΔP and ΔQ are the changes in active power and reactive power of the node respectively; Δθ and ΔU are the changes in node voltage phase angle and amplitude respectively; J is the Jacobian matrix of power flow calculation; H, N, M, and L are the corresponding Jacobian submatrices respectively. Expanding the above formula, it is expressed by the following formula:

[0046] After eliminating Δθ, it is expressed by the following formula:

[0047] In the formula: S P , S Q are the active voltage sensitivity matrix and the reactive voltage sensitivity matrix respectively, and the element S Pij , S Qij in the i-th row and j-th column of them are the change values of node i voltage when the active and reactive power changes of node j are unit values respectively.

[0048] Considering the mutual influence between nodes, the electrical distance of each node in the distribution network can be characterized by the following formula:

[0049] In the formula: D ij , D ji are both the electrical distances between nodes i and j; d ij is used to characterize the influence degree of the power change injected into node j on the voltage of node i, where the power includes active power and reactive power.

[0050] Therefore, in the modularity index It is expressed by the following formula:

[0051] In the formula: max(D ij ) is the maximum value in the electrical distance matrix.

[0052] B. Construct the active power balance index.

[0053] In order to enable the internal autonomy of the cluster to be given priority and avoid serious problems of network loss caused by large-scale power transmission, the active power balance index is used to divide the nodes in the distribution network.

[0054] For the divided clusters, within the entire planning or scheduling period T, the net loads of each node in the cluster c i at each scheduling time period t are summed to obtain P net,ci (t), and the maximum cluster net load max(P net,ci (t)) at this time period is calculated. Then the active power balance index is expressed by the following formula:

[0055]

[0056] In the formula: P net,ci is the active power balance value of the cluster c i ; γ P is the active power balance index, C is the number of divided clusters.

[0057] C. Construct the reactive power balance index.

[0058] Regarding the control problem of voltage over-limit, the cluster should also have a certain voltage regulation ability. Therefore, it is necessary to set the reactive power balance index.

[0059] Considering the influence brought by the access of distributed photovoltaic and wind power to the distribution network nodes, among them, only the output active power of wind power is considered, and photovoltaic can output both active power and output or absorb reactive power. If the photovoltaic operates at the rated power factor angle θ PV , then at time t, when the photovoltaic outputs active power P PV,t , the reactive power that the photovoltaic can output or absorb is expressed by the following formula:

[0060] In the formula: Q PV,t is the reactive power output by the photovoltaic.

[0061] Taking the radial distribution network as an example, analyze the impact of large-scale photovoltaic and wind power access to the distribution network on node voltage and line losses. Its typical structure is as Figure 2 shown. In Figure 2 , there are N users distributed along the line, and the reactance between the i-th and j-th users is R j +jX j . The voltage at the beginning of the line is U 0 , the voltage of the i-th user is U i , the load power is expressed as P j +jQ j , the photovoltaic power generation is expressed as P PV +jQ PV , and the wind power generation is expressed as P WG .

[0062] When photovoltaic and wind power are not connected, the voltage difference ΔU j between nodes i and j of the line is expressed by the following formula:

[0063] In the formula: U i , U j are the voltages of nodes i and j respectively; Ω j is the set of all nodes from node j to the end of the line; P k , Q k are the active power and reactive power of each node from node j to the end node of this line respectively.

[0064] It can be seen that when photovoltaic and wind power are not connected, usually the load of the node is positive, then the ΔU j at this time is positive, indicating that the node voltage gradually decreases along with the transmission of the line, that is, the node voltage gradually decreases along the line starting from the root node.

[0065] When photovoltaic and wind power are connected to each node, the voltage difference between nodes i and j at this time is the voltage difference between nodes i and j, which is expressed by the following formula:

[0066] In the formula: In the formula: P PV,k , P WG,k are the active powers of the photovoltaic and wind power connected to node k respectively; Q PV,k is the reactive power of the photovoltaic connected to node k.

[0067] It can be seen from the above formula that when the total output of photovoltaic and wind power at node k exceeds the total load at the node, the voltage at node k along the k line will increase; on the contrary, if the total output is not enough to meet the load demand, the voltage will show a downward trend along the line, and undervoltage will occur when the load is too large; if the total output is just balanced with the load, no voltage drop will occur on this branch.

[0068] When photovoltaic and wind power are not connected, the system line loss ΔP loss It is expressed as follows:

[0069] Where: Ω N is the number of nodes in the distribution network.

[0070] When photovoltaic and wind power are connected to each node, the system line loss ΔP loss It is expressed as the following formula:

[0071] Where: Ω N is the number of nodes in the distribution network.

[0072] It can be seen from the above formula that when the total output of photovoltaic and wind power at node k does not exceed the total load at the node, the active power demand at the node is reduced, which can reduce the network loss of the distribution network; on the contrary, if the total output exceeds the total load at the node, the excess power will be transmitted upward, increasing the power flow in the line, thereby increasing the system network loss and may even cause line overload problems.

[0073] Therefore, considering the above-mentioned impacts brought by the large-scale photovoltaic and wind power access to the distribution network nodes, the impact on voltage over-limit is considered in particular. When the voltage of the node is too high, the node needs to increase reactive power demand to reduce the node voltage. Therefore, within the dispatch period T, the most serious moment of voltage over-limit at each node of the distribution network is found, that is, the moment when photovoltaic and wind power have the largest output. At this moment, the reactive power supply within the cluster should meet the reactive power demand as much as possible, while also reducing reactive power transmission across clusters.

[0074] The reactive power balance index setting is expressed by the following formula:

[0075]

[0076] Where: Q ci For cluster c i Reactive power balance; γ Qis the reactive power balance index; Q ci,sup is the maximum value of reactive power supply within the cluster, including the reactive power provided by the node reactive power compensation device and the reactive power that the photovoltaic inverter can provide; Q ci,need is the demand value of reactive power within the cluster.

[0077] It should be noted that Q ci,need not only refers to the normal reactive power demand of the node, but also includes the minimum reactive power demand required at the moment when the node voltage exceeds the limit most severely within the scheduling period T, that is, at the moment when the output of photovoltaic, wind power, etc. is the largest, which is expressed by the following formula:

[0078] In the formula: Q ci,V is the minimum reactive power required to regulate the overvoltage node of cluster c i ; ΔU j is the difference between the voltage value of node i at the moment t when the voltage exceeds the limit most severely and the voltage value when photovoltaic and wind power are not connected; S Q,ii is the reactive power-voltage sensitivity of node i with respect to itself.

[0079] D. Construct a flexibility supply-demand matching index.

[0080] As a supplement to the active power balance degree, the flexibility of the power system focuses on solving the problem of power fluctuation in the system under a certain time scale. The double fluctuations of the power source and load require the invocation of flexibility resources for matching. If the flexibility of the system is insufficient, there will be risks of wind curtailment and load shedding. Quantifying the flexibility of the power system requires determining the flexibility demand.

[0081] The volatility and uncertainty of new energy and load are the main sources of node flexibility demand. Therefore, the node flexibility demand is expressed as the sum of the fluctuation amount of the node net load and the difference between the load and the new energy prediction error under the quantified time scale, which is expressed by the following formula:

[0082] In the formula: F i,t is the flexibility demand of node i at time t; P net,i,t is the net load of node i at time t; e res,i,t 、e l,i,t are the prediction errors of the new energy output and load of node i at time t.

[0083] Furthermore, according to the direction of net load fluctuations, flexibility requirements can be divided into upward flexibility requirements and downward flexibility requirements, which are specifically expressed by the following formula:

[0084] In the formula: F up,i,t is the upward flexibility requirement; F down,i,t is the downward flexibility requirement.

[0085] It can be seen from the above formula that the upward and downward flexibility requirements of the system at any moment must not be less than 0.

[0086] The flexibility of the cluster is defined as follows: In the power balance of the concerned time scale, the cluster optimally allocates various flexibility resources within the group to adapt to the flexibility requirements of power increase or decrease.

[0087] Therefore, the flexibility supply-demand matching degree index of the cluster is expressed by the following formula:

[0088]

[0089]

[0090] In the formula: F up,ci 、F down,ci are the upward and downward flexibility requirement values of cluster c i respectively; F ci is the flexibility requirement value of cluster c i ; γ F is the flexibility supply-demand matching degree.

[0091] Step 3.2: Based on the modularity index, active and reactive power balance indexes, and flexibility supply-demand matching degree index of the electrical distance of the active and reactive power sensitivities constructed in Step 3.1, a target function is established by fusion.

[0092] Based on the above analysis of the partitioning indexes of each cluster, in order to fully exert the active and reactive power autonomy capabilities and flexibility response capabilities of the cluster, a target function ψ is established by comprehensively considering the modularity, active power balance, reactive power balance, and flexibility supply-demand matching degree indexes, which is expressed by the following formula:

[0093] In the formula: λ 1 、λ 2 、λ 3 、λ4 They are the weight coefficients of each index and satisfy λ 1 + λ 2 + λ 3 + λ 4 = 1.

[0094] Among the various indicators, the modularity index of the electrical distance based on active and reactive sensitivities ensures the structure of the system and is the most basic partitioning index, generally having a relatively large weight. The active power balance index and the reactive power balance index are the basis for achieving a certain coordination ability within the cluster and play an important role among the various functional indicators. On the basis of the cluster achieving power balance, the index weight of the flexibility supply-demand matching degree can be fully considered.

[0095] Step 3.3: Solve the objective function established in Step 3.2.

[0096] Cluster partitioning algorithms can generally be divided into methods such as clustering algorithms, optimization algorithms, and community discovery in complex networks. The genetic algorithm is one of the most commonly used optimization algorithms for solving the cluster partitioning problem. The traditional genetic algorithm uses fixed crossover probability and mutation probability. High-quality individuals and low-quality individuals will perform crossover and mutation operations with the same probability. And after the number of iterations increases, the cluster results get closer and closer, and the fixed crossover probability loses its meaning. Therefore, it reduces the global search ability and convergence speed of the algorithm. Therefore, adopting the idea of adaptive operation, setting dynamically changing mutation probability and crossover probability. As the number of iterations increases, the mutation probability should gradually increase to improve the global search ability of the algorithm; the crossover probability should gradually decrease to improve the convergence ability.

[0097] The specific self-adaptive adjustment of crossover and mutation is expressed by the following formula:

[0098] In the formula: p c and p m are the crossover and mutation probabilities respectively; p c,max and p c,min are the maximum and minimum values of the crossover probability respectively; p m,max and p m,min are the maximum and minimum values of the mutation probability respectively; I is the number of iterations; Imax is the maximum number of iterations; f is the larger fitness value of the two individuals for crossover operation; f m is the fitness value of the individual for mutation operation; f avg is the average fitness of the population.

[0099] The specific process of using the improved adaptive genetic algorithm to solve the cluster partitioning problem is as follows Figure 1 shown. The input basic parameters include the power supply types, outputs, and load conditions of each node in the system, as well as the flexibility demand response situation. The final objective function of the cluster partitioning will be used as the fitness function of the improved genetic algorithm to judge the quality of individuals in the population, so as to achieve the survival of the fittest among individuals. When the fitness of the optimal individual reaches the given threshold or the number of iterations reaches the preset value, the algorithm will terminate. The finally obtained partitioning result will be used as the basis for cluster partitioning.

[0100] Step 4: Analyze the results of the cluster partitioning to verify the effectiveness of the proposed cluster partitioning method.

[0101] To verify the influence of different indicators on the cluster partitioning results, the following multiple weight setting modes are constructed, and the respective indicator values after the cluster partitioning results are compared and analyzed to analyze the influence of different functional indicators on the cluster partitioning.

[0102] The specific weight setting modes are as follows Mode 1: λ 1 = 1, λ 2 = λ 3 = λ 4 = 0, that is, only structural indicators Mode 2: λ 1 = 0.5, λ 2 = 0.5, λ 3 = λ 4 = 0, that is, only considering the active power balance degree within the cluster Mode 3: λ 1 = 0.4, λ 2 = λ 3 = 0.3, λ 4 = 0, that is, comprehensively considering the active and reactive power balance degrees within the cluster, but not considering the flexibility supply-demand matching degree Mode 4: λ 1 = 0.4, λ 2 = 0.3, λ 3 = λ 4 = 0.15, that is, comprehensively considering all the proposed indicators

[0103] Considering the influence of different proportions of new energy grid connection of each node on the cluster partitioning results, the cluster partitioning considering all indicators under new energy penetration rates of 20%, 60%, and 100% is set. Analyze the influence of different new energy penetration rates on the cluster partitioning.

[0104] The specific setting modes are as follows Mode 5: The new energy penetration rate is 20% Mode 6: The new energy penetration rate is 60%; Mode 7: The new energy penetration rate is 100%.

[0105] Step 8: Adopt this cluster division scheme through verification.

[0106] It should be noted that Embodiment 1 provides a method for dividing clusters of an active distribution network with multiple microgrids considering uncertainty and cluster comprehensive performance. The steps are carried out according to strict logic and cannot be simply split or omitted to divide the clusters. According to the results of cluster division under all steps, important reference bases for the site selection and capacity determination of active and reactive power sources in the subsequent distribution network and the operation and dispatch of the distribution network can be provided based on the values of active power, reactive power, and flexibility supply-demand matching indexes within the divided clusters.

[0107] Embodiment 2 of the present invention provides a system for dividing clusters of an active distribution network with multiple microgrids considering uncertainty and cluster comprehensive performance, which operates the method for dividing clusters of an active distribution network with multiple microgrids considering uncertainty and cluster comprehensive performance as described in Embodiment 1, including: An uncertainty processing module, which is used to process the uncertainty of power data to obtain the final scenario power data; A power flow calculation module, which is used to perform power flow calculation on the scenario power data and distribution network parameters to obtain cluster division data; A cluster comprehensive performance evaluation module, which is used to establish a distribution network cluster division model based on the basic data of cluster division; A solver, which is used to solve the distribution network cluster division model to obtain the distribution network cluster division result.

[0108] Embodiment 3 of the present invention provides an electronic device, including a memory, a processor, and a computer program stored on the memory and executable on the processor. When the computer program is loaded into the processor, it implements the method for dividing clusters of an active distribution network with multiple microgrids considering uncertainty and cluster comprehensive performance as described in Embodiment 1.

[0109] Embodiment 4 of the present invention provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, it implements the method for dividing clusters of an active distribution network with multiple microgrids considering uncertainty and cluster comprehensive performance as described in Embodiment 1.

[0110] To verify the effectiveness of the cluster division method proposed in the present invention, an improved IEEE33-node system as Figure 3 shown is used for case analysis.

[0111] The system includes 8 photovoltaic nodes, 7 wind power nodes, 5 reactive power compensation device nodes, and is connected to 3 nodes of the microgrid. The installed capacity of photovoltaic in the system is 400kW, the installed capacity of wind power is 200kW, and the maximum exchange power between the microgrid and the distribution network is 150kW. Through the proposed uncertainty processing method, the typical daily scenario data of wind power, photovoltaic, load, and the interactive power between the distribution network and the microgrid are obtained as Figure 4 shown.

[0112] After performing power flow calculation and analysis, the 24-hour voltage fluctuation data before the connection of photovoltaic and wind power and the voltage fluctuation data after the connection are obtained as Figure 5 and Figure 6 shown.

[0113] In this scenario, it can be seen that the reverse power transmission of active power at the nodes is obvious when the sunlight intensity is strong at noon, and the voltage limit violation is serious. Through calculation, it can be obtained that the penetration rates of photovoltaic and wind power generation are the highest at 14:00, and the voltage limit violation is the most serious at this time. Therefore, this time is selected to calculate the reactive power balance index of cluster division.

[0114] The adaptive genetic algorithm is adopted, with the population size N = 60, the maximum number of iterations Imax = 500, the crossover probability pc = (0.6, 0.8), and the mutation probability pm = (0.01, 0.1). The cluster division results under different index weights are obtained.

[0115] To verify the influence of different indexes on the cluster division results, the following multiple weight setting modes are constructed, and the various index values after the cluster division results are compared and analyzed to analyze the influence of different functional indexes on the cluster division.

[0116] The specific weight setting modes are as follows: Mode 1: λ 1 = 1, λ 2 = λ 3 = λ 4 = 0, that is, only the structural index; Mode 2: λ 1 = 0.5, λ 2 = 0.5, λ 3 = λ 4 = 0, that is, only considering the active power balance degree within the cluster; Mode 3: λ 1 = 0.4, λ 2 = λ 3 = 0.3, λ 4 = 0, that is, comprehensively considering the active and reactive power balance degrees within the cluster, but not considering the flexibility of supply-demand matching degree; Mode 4: λ 1 = 0.4, λ 2 = 0.3, λ 3= λ 4 = 0.15, that is, all the proposed indicators are comprehensively considered.

[0117] The calculation results of the cluster division indicators under different modes are shown in the following table: Table 1 Calculation results of each indicator for cluster division under different modes

[0118] The corresponding nodes for cluster division under different modes are as Figures 7 to 10 shown: As shown in the table, since Mode 1 only considers the structural indicator of modularity, the modularity indicator after cluster division is the best, indicating that the electrical connections between nodes within each cluster are close. However, the active and reactive power balance indicator value after its division is relatively low, indicating that the active and reactive power output coordination within this cluster is not effective, and there is a large-scale power transmission in actual operation; After Mode 2 takes into account both modularity and active power balance, its active power balance indicator has improved compared to Mode 1, but relatively sacrifices the modularity indicator; Mode 3 further considers the reactive power balance on the basis of Mode 2. It can be found that compared with the previous two modes, its reactive power matching degree has been effectively improved, and the matching effect of active power is still good, yet the structure has decreased; After comprehensively considering all indicators, although each indicator after the cluster division of Mode 4 is not the highest compared to the previous 3 modes, the comprehensive indicator is the best among all modes, indicating that after fully considering each indicator, the cluster division can achieve the overall optimal result.

[0119] At the same time, it can be noted that under the action of the comprehensive indicator of Mode 4, the distribution network is divided into 3 clusters, while 4 clusters are generated under the previous 3 modes. In order to eliminate the non-uniformity problem brought by different cluster numbers to the comparison of the above mode indicator values, it is stipulated in advance in Mode 4 to output four clusters, solve the optimal cluster division result, and the obtained indicator values and the cluster division node conditions are as shown in the following table and Figure 11 shown.

[0120] It can be seen that the indicator values when the number of clusters is 4 are not as good as those when the number is 3, but compared with Modes 1, 2, and 3, the overall performance of the division result under the comprehensive indicator is still better than the division results with a single indicator or fewer indicators.

[0121] Table 2 Calculation results of each indicator for different cluster numbers in Mode 4

[0122] The above pattern is the cluster division under the condition that all new energy sources are connected to the grid at each node. Next, consider the impact of different proportions of new energy grid connection at each node on the cluster division results, and set the cluster division considering all indicators under new energy penetration rates of 20%, 60%, and 100%. The result under a new energy penetration rate of 100% is the result under Mode 4.

[0123] The results of the cluster division indicators under different new energy penetration rates are as follows: Table 3 Calculation results of each indicator under different new energy penetration rates

[0124] The corresponding nodes for the cluster division under different new energy penetrations are as Figure 12 , Figure 13 shown. Figure 12 is the cluster division situation under a new energy penetration rate of 20%. Figure 13 is the cluster division situation under a new energy penetration rate of 60%.

[0125] It can be seen from the above results that the indicator values are the best when the new energy penetration rate is 60%. From this, it can be analyzed that the system operation status is the best at this time. When the new energy penetration rate is too low or too high, the impact on the system operation status is more severe. This also shows that to a certain extent, the indicators of cluster division can reflect the operation status of the entire system.

[0126] Compared with the existing technology, the beneficial effects of the present invention at least include: fully considering the uncertainty dominant factors in the active distribution network with multiple microgrids, and proposing a method for dealing with its uncertainty. In addition, the present invention proposes a comprehensive indicator considering the flexibility of supply-demand matching, which gives full play to the structural, active and reactive coordination capabilities, and flexibility response capabilities within the cluster, helps to improve its autonomy from within the cluster, and reduces problems such as serious network losses and voltage over-limits caused by large-scale power transmission, and can provide a theoretical basis and technical support for the active-reactive power planning of the active distribution network with multiple microgrids.

[0127] Finally, 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 them. Although the present invention has been described in detail with reference to the above embodiments, those of ordinary skill in the art should understand that: still can modify the specific implementation manners of the present invention or make equivalent replacements, and any modification or equivalent replacement without departing from the spirit and scope of the present invention shall be covered by the protection scope of the claims of the present invention.

Claims

1. A distribution network cluster division method taking into account uncertainty and comprehensive performance, wherein the distribution network is an active distribution network set including multiple microgrids, and the distribution network cluster division method is characterized in that it includes the following steps: The probability distribution is used to characterize the random characteristics of distributed energy, load demand and interactive power of distribution microgrids contained in the distribution network, and multiple scenarios are generated to characterize the uncertainty of active distribution network containing multiple microgrids; At least one of the structural indicators and functional indicators that characterize cluster performance indicators is selected, and multiple cluster performance indicators are integrated to form a cluster comprehensive performance indicator as a distribution network cluster division model taking into account the cluster comprehensive performance; For various scenarios that represent uncertainty, the power flow calculation is performed on the distribution network, and the power flow calculation results are substituted into the distribution network cluster partition model that takes into account the comprehensive performance of the cluster; The distribution network cluster division model is solved to obtain the distribution network cluster division result.

2. The method for dividing distribution network clusters taking into account uncertainty and comprehensive performance according to claim 1, characterized in that: The characterization of uncertainty of active distribution network containing multiple microgrids specifically includes: Determine the probability distribution model of distributed energy, load demand and distribution microgrid interaction power; Randomly sample the probability distribution model of distributed energy, load demand and distribution microgrid interaction power to generate a set number of scenarios representing different operating states; The scenarios representing different operating states are clustered to reduce the set number to a plurality of typical representative scenarios.

3. A distribution network cluster division method taking into account uncertainty and comprehensive performance according to claim 1 or 2, characterized in that: The characterization of uncertainty of active distribution network containing multiple microgrids specifically includes: Determine the probability distribution model of wind power, photovoltaic power, load demand and distribution microgrid interaction power by analyzing historical data or based on physical models; The Monte Carlo method is used to randomly sample the probability distribution model to generate scenarios with different set numbers; Apply K-means clustering algorithm to cluster the generated Monte Carlo scenarios with a set number of groups; According to the results of K-means clustering, representative cluster centers are selected as the final scene set.

4. The method for dividing distribution network clusters taking into account uncertainty and comprehensive performance according to claim 1, characterized in that: The selecting at least one of the structural indicators and functional indicators representing the cluster performance indicators includes: The modularity index is selected as the structural index, and at least one of the active power balance, reactive power balance, and flexibility supply and demand matching index is selected as the functional index.

5. A distribution network cluster division method taking into account uncertainty and comprehensive performance according to claim 1 or 4, characterized in that: Based on the constructed modularity index of electrical distance of active and reactive sensitivity, active and reactive balance index and flexible supply and demand matching index, the objective function of the distribution network cluster partitioning model taking into account the comprehensive performance of the cluster is established according to weight fusion.

6. A distribution network cluster division method taking into account uncertainty and comprehensive performance according to claim 5, characterized in that: The modularity index of the electrical distance of active and reactive sensitivity is constructed by: calculating the active voltage sensitivity matrix and the reactive voltage sensitivity matrix, using the sensitivity matrix elements to calculate the electrical distance between each node in the distribution network, constructing the electrical distance matrix, obtaining the maximum value as the edge weight connecting two nodes, and calculating the modularity ρ based on the edge weight as the modularity index; Constructing active and reactive power balance indicators includes: establishing an active power balance indicator based on the sum of net loads of each cluster, and establishing a reactive power balance indicator based on the ratio of the maximum value of reactive power supply and demand value within the cluster; Constructing a flexibility supply and demand matching index includes: calculating cluster flexibility demand values ​​based on cluster-based upward and downward adjustment of flexibility demand values, and establishing a flexibility supply and demand matching index.

7. A distribution network cluster division method taking into account uncertainty and comprehensive performance according to claim 1 or 6, characterized in that: The method of integrating multiple cluster performance indicators to form a cluster comprehensive performance indicator as a distribution network cluster division model taking into account the cluster comprehensive performance includes: The modularity index, active power balance index, reactive power balance index and flexibility supply and demand matching index are multiplied by their weights respectively and then added, and the maximum sum is taken as the objective function; among them, the sum of the weights of the modularity index, active power balance index, reactive power balance index and flexibility supply and demand matching index is 1.

8. A distribution network cluster division method taking into account uncertainty and comprehensive performance according to claim 7, characterized in that: Solving the distribution network cluster division model to obtain the distribution network cluster division result includes: The power type, output and load conditions of each node, as well as the flexibility demand response conditions, are input into the model. Various cluster division methods are encapsulated as populations and solved using a genetic algorithm to obtain the distribution network cluster division results.

9. A distribution network cluster division method taking into account uncertainty and comprehensive performance according to claim 8, characterized in that: The distribution network cluster division method further includes: The clustering results are analyzed to verify the effectiveness of the proposed clustering method, and the scheme that passes the verification is used as the clustering scheme; The analysis of the cluster division results includes: partially setting the weight coefficients of various indicators to zero or setting them all to non-zero, combining the new energy penetration rate, and judging the operating status of the distribution network system according to the comprehensive performance indicator values.

10. A distribution network cluster division system taking into account uncertainty and comprehensive performance, running a distribution network cluster division method taking into account uncertainty and comprehensive performance according to any one of claims 1 to 9, characterized in that: include: An uncertainty processing module is used to characterize the random characteristics of distributed energy, load demand and distribution microgrid interactive power contained in the distribution network by using probability distribution, and generate multiple scenarios representing the uncertainty of active distribution network containing multiple microgrids; The power flow calculation module is used to characterize various scenarios of uncertainty and perform power flow calculation on the distribution network; A cluster comprehensive performance evaluation module is used to select at least one of the structural indicators and functional indicators that characterize the cluster performance indicators, and fuse multiple cluster performance indicators to form a cluster comprehensive performance indicator to form a distribution network cluster division model; The solver is used to solve the distribution network cluster partitioning model to obtain the distribution network cluster partitioning result.

Images