A reactive power partitioning method for power systems considering source-load power fluctuation

Abstract

The application discloses a power system reactive power partitioning method considering source and load power fluctuation, and belongs to the field of power system reactive power voltage control and voltage stability analysis. The method comprises the following steps: constructing an operation condition set containing power electronic source output fluctuation and load fluctuation, checking the convergence of alternating current power flow of each operation condition, screening all feasible operation conditions with converged power flow to form a source and load fluctuation scene set; based on each operation scene, constructing a reactive power transmission distribution factor matrix; establishing a reactive power transmission distance matrix between PQ nodes and a comprehensive reactive power transmission distance matrix; clustering the comprehensive reactive power transmission distance matrix to obtain a power system reactive power partitioning result and evaluating the result. The application reflects the change of power system reactive power support capability after high proportion of power electronic sources are connected and the influence of source and load fluctuation on the reactive power transmission relationship, improves the adaptability of the reactive power partitioning result to complex operation scenes, and provides a basis for new type power system reactive power voltage partitioning control and reactive power compensation planning.

Description

Technical Field

[0001] This invention relates to the field of reactive power voltage control and voltage stability analysis in power systems, and specifically to a reactive power partitioning method for power systems that takes into account source-load power fluctuations. Background Technology

[0002] The voltage level at power system nodes is closely related to reactive power, and reasonable reactive voltage control is an important measure to ensure the safe and stable operation of the power grid. Because reactive power has strong local balance characteristics, long-distance reactive power transmission can easily lead to increased voltage loss and reduced reactive power support efficiency. Therefore, in actual power grid operation and planning, it is usually necessary to divide the system into several reactive power control zones based on factors such as grid structure, voltage level, reactive power support capacity, and operating mode, in order to carry out zoned voltage control, reactive power compensation device configuration, and reactive power support resource coordination.

[0003] With the large-scale integration of new energy power generation, flexible DC transmission, and power electronic loads, the power system's source-load structure and operating characteristics have undergone significant changes. On the one hand, the output of power electronic sources such as wind power and photovoltaics is significantly affected by meteorological conditions, exhibiting volatility and uncertainty. On the other hand, load levels and distribution change over time, leading to variations in system power flow distribution, node voltage, and line reactive power flow under different operating conditions. The high proportion of power electronic sources also alters the start-up methods and reactive power support distribution of traditional synchronous generators, making the system's reactive power support capacity, reactive power transmission paths, and voltage control relationships more strongly correlated with operating conditions. Therefore, reactive power zoning based solely on a single typical section or fixed operating mode is insufficient to fully reflect the changes in reactive power transmission relationships under fluctuating source-load conditions, and the resulting zoning results are not sufficiently adaptable to complex operating scenarios.

[0004] Existing reactive power zoning methods typically construct inter-node relationships based on information such as voltage-reactive power sensitivity, electrical distance, or network topology. Some methods also consider wind power fluctuations or source-load timing characteristics, and correct electrical distances through multiple power flow states or zoning for different scenarios. However, these methods mostly describe inter-node coupling relationships from the perspective of node voltage response or network structure, making it difficult to directly reflect the transmission characteristics of reactive power injection disturbances at PQ nodes in the line. At the same time, scenario-by-scenario zoning or zoning adjustments are difficult to form a unified zoning result applicable to a certain range of source-load fluctuations, and existing evaluation indicators rarely directly measure the level of reactive power transmission across zones. Summary of the Invention

[0005] The purpose of this invention is to overcome the problems of insufficient adaptability of existing reactive power zoning methods to source-load fluctuations and difficulty in directly characterizing the impact of reactive power disturbances at PQ nodes on reactive power flow at line ports. This invention provides a reactive power zoning method for power systems that takes into account source-load power fluctuations. To adapt to the wide range and frequent changes in source-load fluctuations in power systems with a high proportion of renewable energy, and considering diverse operating conditions under source-load fluctuations, this invention analyzes the mapping relationship between reactive power injection disturbances at PQ nodes and changes in reactive power flow at line ports. It constructs a reactive power transmission distribution factor matrix characterizing the reactive power flow response characteristics of lines, and further constructs a reactive power transmission distance matrix and a comprehensive reactive power transmission distance matrix between PQ nodes. The reactive power inter-regional transmission rate is used to evaluate the clustering and zoning results, thereby improving the adaptability of the reactive power zoning results to multiple operating modes and complex source-load fluctuation scenarios.

[0006] To achieve the above objectives, the present invention adopts the following technical solution:

[0007] A reactive power zoning method for a power system that takes into account source-load power fluctuations includes the following steps:

[0008] Step S1: Obtain the basic network structure data of the power system to be partitioned, determine all PQ nodes in the system as nodes to be partitioned, and exclude PV nodes and balancing nodes from being partitioned; construct a source-load operating condition set that includes power electronic power output fluctuations and load fluctuations.

[0009] Step S2: Filter the source load operating conditions according to the application scenario type of the source load operating conditions to form a source load fluctuation scenario set;

[0010] Step S3: For each feasible operating condition in the source load fluctuation scenario set, taking PQ node as the reactive power disturbance node, calculate the impact of unit reactive power injection disturbance of PQ node on reactive power flow of system line port, and obtain the reactive power transmission distribution factor matrix under the corresponding operating condition; the columns of the reactive power transmission distribution factor matrix correspond one-to-one with the nodes to be partitioned.

[0011] Step S4: Based on the reactive power transmission distribution factor matrix under each feasible operating condition, construct the reactive power transmission distance matrix under the corresponding operating condition; the reactive power transmission distance matrix, as well as the subsequent comprehensive reactive power transmission distance matrix and reactive power partitioning results, are all constructed with the node to be partitioned as the object; the reactive power transmission distance matrices corresponding to multiple feasible operating conditions are combined to obtain the comprehensive reactive power transmission distance matrix.

[0012] Step S5: Using the comprehensive reactive power transmission distance matrix as the distance matrix between PQ nodes, cluster the PQ nodes to obtain the reactive power partitioning result of the power system to be partitioned.

[0013] Step S6: Identify cross-regional lines according to the reactive power zone to which the nodes at both ends of the line belong, calculate the reactive power cross-regional transmission rate, and evaluate the reactive power zone results based on the reactive power cross-regional transmission rate.

[0014] Further, in step S1, the source-load operating conditions are constructed through a planning application operating condition selection method or an operating application operating condition generation method; wherein, the planning application operating condition selection method includes: acquiring historical operating section data of the power system to be partitioned within a preset historical time interval, and selecting historical operating sections that can reflect the power electronic power supply output fluctuation and load fluctuation from the historical operating section data as the source-load operating condition set; the operating application operating condition generation method includes: taking the current operating point of the power system to be partitioned as the benchmark operating condition, and generating a candidate source-load operating condition set that includes power electronic power supply output fluctuation and load fluctuation within a preset source-load fluctuation range.

[0015] Furthermore, the method for generating operating conditions includes: generating multiple candidate source-load operating conditions based on the current operating point, the preset power electronic power supply output fluctuation range, the preset load fluctuation range, and system power balance constraints; wherein, the candidate source-load operating conditions include power electronic power supply output changes, load changes, and power balance adjustment results corresponding to the power electronic power supply output changes and load changes.

[0016] Further, in step S2, when the source load operating conditions are constructed through the planning application operating condition selection method, the source load operating condition set composed of the selected historical operating sections that meet the preset conditions is used as the source load fluctuation scenario set; when the source load operating conditions are constructed through the operating application operating condition generation method, AC power flow calculation is performed on the generated candidate source load operating conditions respectively, and the candidate source load operating conditions that have converged power flow and meet the preset operating constraints are included as feasible operating conditions in the source load fluctuation scenario set; the number of feasible operating conditions in the source load fluctuation scenario set is not less than the preset minimum number of scenarios.

[0017] Further, in step S3, the reactive power transmission distribution factor matrix includes a reactive power transmission distribution factor matrix at the line from end and a reactive power transmission distribution factor matrix at the line to end; the reactive power transmission distribution factor at the line from end is used to characterize the change in reactive power flow at the line from end when a unit reactive power injection disturbance occurs at the PQ node, and the reactive power transmission distribution factor at the line to end is used to characterize the change in reactive power flow at the line to end when a unit reactive power injection disturbance occurs at the PQ node.

[0018] Furthermore, the reactive power transmission distribution factor matrix is ​​analytically obtained through linearization of the AC power flow equations. For any feasible operating condition, the phase angles of the node voltages (excluding the slack node) and the voltage magnitudes of the PQ nodes are used as state variables. The AC power flow equations are linearized for this feasible operating condition, and a unit reactive power injection disturbance matrix for the PQ nodes is constructed to obtain the sensitivity matrix of the state variables to the reactive power injection disturbances at the PQ nodes. Then, based on the partial derivative matrices of the reactive power flow at the from and to ends of the line with respect to the state variables, the reactive power transmission distribution factor matrix at the from end of the line and the reactive power transmission distribution factor matrix at the to end of the line are obtained.

[0019] Further, in step S4, the construction of the reactive power transmission distance matrix between PQ nodes includes: calculating the reactive power transmission coupling strength between PQ nodes based on the impact of the unit reactive power injection disturbance of PQ nodes on the reactive power flow of the adjacent branch port of another PQ node; normalizing and symmetricizing the reactive power transmission coupling strength to obtain the symmetrical reactive power transmission coupling strength; and converting the symmetrical reactive power transmission coupling strength into the reactive power transmission distance between PQ nodes to obtain the reactive power transmission distance matrix between PQ nodes under the corresponding feasible operating conditions.

[0020] Furthermore, the reactive power transmission distance matrix between PQ nodes corresponding to multiple feasible operating conditions is integrated, including weighted summation of the reactive power transmission distance matrix between PQ nodes under each feasible operating condition according to preset weights to obtain the integrated reactive power transmission distance matrix; the preset weights include any one of equal weights, operating probability weights, or voltage stability risk weights.

[0021] Further, in step S5, the comprehensive reactive power transmission distance matrix is ​​used as the distance matrix between PQ nodes, and a preset clustering method is used to cluster the PQ nodes to be partitioned, and the reactive power partitions of PQ nodes are obtained according to the preset number of reactive power partitions; the preset clustering method is a clustering method that can take the distance matrix as input and output node category labels.

[0022] Further, in step S6, the reactive power partition to which a node belongs is determined based on the reactive power partitioning results. For PV nodes and balancing nodes that are not considered as nodes to be partitioned, auxiliary assignment is performed while keeping the reactive power partitioning results of PQ nodes unchanged: they are preferentially assigned to the reactive power partition to which the PQ node directly connects; when they are connected to PQ nodes in multiple different reactive power partitions, they are assigned to the reactive power partition with the largest sum of reactive power transmission volume of the corresponding connecting lines; when they are not directly connected to PQ nodes, their reactive power partition is determined layer by layer based on the adjacent nodes whose assignments have been determined, according to the above principles. Cross-regional lines are identified based on whether the reactive power partitions to which the nodes at both ends of the line belong are the same; the average absolute value of the reactive power flow at both ends of the line is taken as the reactive power transmission volume of the corresponding line, and the ratio of the sum of the reactive power transmission volume of cross-regional lines to the sum of the reactive power transmission volume of all branches in the system is taken as the reactive power cross-regional transmission rate. The smaller the reactive power cross-regional transmission rate, the weaker the cross-regional reactive power transmission, and the better the reactive power partitioning results.

[0023] Furthermore, the node to be partitioned is a PQ node, the columns of the reactive power transmission distribution factor matrix correspond one-to-one with the PQ node to be partitioned, and the reactive power transmission distance matrix between PQ nodes, the comprehensive reactive power transmission distance matrix, and the reactive power partitioning result are all constructed with the PQ node to be partitioned as the object; PV nodes and balancing nodes are not considered as nodes to be partitioned.

[0024] The present invention also provides a computer-readable storage medium storing instructions that, when executed, enable the above-described reactive power partitioning method for power systems taking into account source-load power fluctuations.

[0025] Compared with the prior art, the present invention has the following beneficial effects:

[0026] First, this invention constructs multiple feasible operating conditions that include power output fluctuations and load fluctuations of electronic power sources. Operating conditions can be selected from historical operating sections or candidate operating conditions can be generated around the current operating point based on planning or operational applications. These conditions are then filtered through power flow convergence and preset operating constraints to form a set of source-load fluctuation scenarios. Compared to dividing scenarios solely based on a single wind power probability model or typical daily source-load time series, the resulting scenario set can more specifically characterize the feasible operating states of the power system to be partitioned, improving the adaptability of the partitioning results to source-load fluctuation scenarios.

[0027] Second, this invention constructs a reactive power transmission distribution factor matrix based on the reactive power flow changes at line ports caused by reactive power injection disturbances at PQ nodes and applies it to reactive power partitioning. Compared to partitioning methods based solely on node voltage response, electrical distance, or line static weights, this method more directly reflects the distribution range and response strength of reactive power disturbances along network line ports, making the partitioning criteria more consistent with local reactive power balance and transmission characteristics.

[0028] Third, this invention integrates the reactive power transmission distance under multiple operating conditions and uses the reactive power cross-regional transmission rate to evaluate the zoning results, which is conducive to forming a unified zoning scheme with strong reactive power connection within the zone and weak reactive power transmission between zones. Attached Figure Description

[0029] Figure 1 A flowchart of a power system reactive power partitioning method considering source-load power fluctuations provided in an embodiment of the present invention. Detailed Implementation

[0030] The present invention will be further illustrated below with reference to the accompanying drawings and specific embodiments. It should be understood that the following specific embodiments are for illustrative purposes only and are not intended to limit the scope of the invention. It should be noted that the terms "front," "rear," "left," "right," "up," and "down" used in the following description refer to directions in the accompanying drawings, and the terms "inner" and "outer" refer to directions toward or away from the geometric center of a specific component, respectively.

[0031] The following is combined Figure 1 The present invention will be further described with reference to specific embodiments.

[0032] 1. Testing System and Basic Data

[0033] This embodiment uses the improved IEEE 118-node test system as an example to illustrate the reactive power zoning method for power systems considering source-load power fluctuations described in this invention. Based on the IEEE 118-node test system, this system selects some primary generator nodes as power electronic power supply access nodes and equates them to PQ nodes. This characterizes the operating characteristics of power electronic power supplies under given active power output and reactive power control modes, where the node voltage is no longer directly maintained by the synchronous excitation system. The power electronic power supply access nodes include nodes 25, 49, 54, 59, 61, 65, 66, 80, and 89.

[0034] Determination of the node to be partitioned: In this embodiment, the node to be partitioned is the PQ node; PV nodes and balancer nodes are not considered as nodes to be partitioned. The system branch set is denoted as... The set of PQ nodes to be partitioned is denoted as The number of system branches is denoted as The number of PQ nodes to be partitioned is denoted as .

[0035] Step S1: Construct source load operating conditions.

[0036] This embodiment constructs source-load operating conditions using an application operating condition generation method. The improved IEEE 118-node test system baseline operating point is used as the baseline operating condition, and a set of power electronic power output coefficients and a set of load level coefficients are set.

[0037] The set of power output coefficients for power electronic sources is as follows:

[0038]

[0039] The set of load level coefficients is as follows:

[0040]

[0041] The power output coefficient and load level coefficient of the power electronic power source are combined to form a set of candidate source-load operating conditions:

[0042]

[0043] in, This is a set of candidate source load operating conditions. This represents the operating condition for the s-th candidate source load. The power output coefficient of the power electronic power supply under the operating condition of the s-th candidate source load; is the load level coefficient under the operating condition of the s-th candidate source load.

[0044] In this embodiment, the number of candidate source load operating conditions is:

[0045]

[0046] For the s-th candidate source load operating condition, the active power output of the j-th power electronic power source is determined according to its installed capacity and the power electronic power source output coefficient:

[0047]

[0048] in, The active power output of the j-th power electronic power source under the operating condition of the s-th candidate source load; Let be the active power installed capacity of the j-th power electronic power source.

[0049] Under constant power factor control, the reactive power output of the j-th power electronic power source is:

[0050]

[0051] in, The reactive power output of the j-th power electronic power source under the operating condition of the s-th candidate source load; k j This is the ratio of reactive power to active power determined by the preset power factor of the j-th power electronic power source. Besides constant power factor control, the power electronic power source can also employ constant reactive power control, constant voltage control, or reactive-voltage droop control.

[0052] For the s-th candidate source load operating condition, the active load and reactive load of the ith load node are respectively:

[0053]

[0054] in, and These represent the active and reactive loads of the i-th load node under the operating condition of the s-th candidate source load; and These represent the active and reactive loads of the i-th load node under the baseline operating conditions.

[0055] The difference in system active power caused by changes in the output of the power electronic power source and load changes is borne by the synchronous generator units. The active power difference borne by the synchronous generator units under the operating condition of the s-th candidate source-load is:

[0056]

[0057] in, The active power difference undertaken by the synchronous generator unit under the operating condition of the s-th candidate source load; For the set of load nodes; A collection of power electronic power supplies; This represents the active power output of the j-th power electronic power source under the baseline operating conditions.

[0058] In this embodiment, the active power difference is allocated among the synchronous generator units according to the baseline active power output ratio. The allocation weight for the g-th synchronous generator unit is:

[0059]

[0060] in, The active power difference allocation weight for the g-th synchronous generator unit; A collection of synchronous generator units participating in power balancing; This is the baseline active power output of the g-th synchronous generator unit.

[0061] The unlimited active power output of the g-th synchronous generator unit after proportional allocation is:

[0062]

[0063] After considering the upper and lower limits of the active power output of the synchronous generator units, the active power output of the g-th synchronous generator unit is:

[0064]

[0065] in, The active power output of the g-th synchronous generator unit under the operating condition of the s-th candidate source load; and These are the lower limit and upper limit of the active power output of the g-th synchronous generator unit, respectively.

[0066] Step S2: Form a source load fluctuation scenario set.

[0067] AC power flow calculations were performed on the 35 candidate source load operating conditions. If the power flow of a candidate source load operating condition converged and met the preset operating constraints, it was included as a feasible operating condition in the source load fluctuation scenario set; if the power flow of a candidate source load operating condition did not converge or did not meet the preset operating constraints, it was removed.

[0068] The source load fluctuation scenario set is denoted as:

[0069]

[0070] In this embodiment, all 35 candidate source-load operating conditions have power flow convergence and satisfy operating constraints. Therefore, the number of feasible operating conditions in the source-load fluctuation scenario set is:

[0071]

[0072] And satisfy:

[0073]

[0074] in, The number of feasible operating conditions in the source load fluctuation scenario set; Set the minimum number of scenes.

[0075] Step S3: Calculate the reactive power transmission distribution factor matrix.

[0076] For the s-th feasible operating condition in the source-load fluctuation scenario set, the PQ node to be partitioned is taken as the reactive power disturbance node. The impact of the unit reactive power injection disturbance of the PQ node on the reactive power flow of the system line port is calculated, and the reactive power transmission distribution factor matrix of the line from end and the reactive power transmission distribution factor matrix of the line to end are obtained.

[0077] For the line and PQ node The reactive power transmission distribution factors at the from end and the to end of the line under the s-th feasible operating condition are defined as follows:

[0078]

[0079]

[0080] in, This represents the change in reactive power flow at the from end of line l when a unit reactive power injection disturbance occurs at node k of PQ under the s-th feasible operating condition. This represents the change in reactive power flow at the to end of line l when a unit reactive power injection disturbance occurs at node k of PQ under the s-th feasible operating condition. and These are the reactive power flows at the from and to ends of line l under the s-th feasible operating condition, respectively. Let be the reactive power injection amount of node k in PQ under the s-th feasible operating condition.

[0081] Calculate for all branches and all PQ nodes to be partitioned separately. and ,get:

[0082]

[0083] ;

[0084] in, Let be the reactive power transmission distribution factor matrix at the from end of the line under the s-th feasible operating condition; Let be the reactive power transmission distribution factor matrix at the to end of the line under the s-th feasible operating condition.

[0085] This embodiment uses a linearized analytical method based on AC power flow equations to calculate the reactive power transmission distribution factor matrix. For the s-th feasible operating condition, the phase angles of the node voltages, excluding the slack node, are... and PQ node voltage amplitude As state variables, linearizing the AC power flow equations yields the sensitivity matrix of the state variables to reactive power injection disturbances at nodes P and Q:

[0086]

[0087] in, Let be the AC power flow Jacobian matrix under the s-th feasible operating condition; This is the sensitivity matrix of the node voltage phase angle to reactive power injection disturbances at the PQ node; This is the sensitivity matrix of the voltage amplitude at node PQ to reactive power injection disturbances at node PQ; This is the zero matrix corresponding to the active power equation; This is the identity matrix corresponding to the reactive power injection disturbance at the PQ node.

[0088] Based on the partial derivative matrix of reactive power flow at the line port with respect to the state variables, we obtain:

[0089]

[0090]

[0091] in, and These are the reactive power flow vectors of the entire system's lines at the from and to ends under the s-th feasible operating condition.

[0092] Step S4: Construct the reactive power transmission distance matrix and the integrated reactive power transmission distance matrix between PQ nodes.

[0093] For the s-th feasible operating condition, let the set of branches with node m of PQ as the from end be . The set of branches with node m of PQ as the to end is The set of adjacent branches of node m in PQ is:

[0094]

[0095] Based on the impact of the unit reactive power injection disturbance at node k of PQ on the reactive power flow of the adjacent branch ports of node m of PQ, calculate the reactive power transmission coupling strength from node k of PQ to node m of PQ:

[0096] ;

[0097] in, Let be the reactive power transmission coupling strength between PQ node k and PQ node m under the s-th feasible operating condition; Let m be the number of adjacent branches of node m in PQ.

[0098] To eliminate the influence of differences in the number of connected branches and local response amplitudes at different nodes, the reactive power transmission coupling strength is normalized:

[0099]

[0100] in, This represents the normalized reactive power transmission coupling strength. To prevent extremely small positive numbers with a denominator of zero.

[0101] The normalized reactive power transmission coupling strength is then symmetrically processed:

[0102]

[0103] in, Let be the symmetric reactive power transmission coupling strength between PQ node m and PQ node k under the s-th feasible operating condition.

[0104] Converting symmetrical reactive power transmission coupling strength into reactive power transmission distance:

[0105]

[0106] The diagonal elements are:

[0107]

[0108] This yields the reactive power transmission distance matrix between PQ nodes under the s-th feasible operating condition:

[0109]

[0110] in, Let be the reactive power transmission distance matrix between PQ nodes under the s-th feasible operating condition; The smaller the value, the stronger the reactive power transmission coupling between PQ node m and PQ node k.

[0111] The reactive power transmission distance matrices between PQ nodes corresponding to each feasible operating condition in the source-load fluctuation scenario set are synthesized to obtain the comprehensive reactive power transmission distance matrix. This embodiment uses an equal-weighting method, and the comprehensive reactive power transmission distance matrix is ​​as follows:

[0112]

[0113] in, This is a comprehensive reactive power transmission distance matrix; Let be the weight corresponding to the s-th feasible operating condition, and satisfy:

[0114]

[0115] In this embodiment, each feasible operating condition is weighted equally:

[0116]

[0117] Step S5: Perform clustering based on the comprehensive reactive power transmission distance matrix.

[0118] Based on the integrated reactive power transmission distance matrix The distance matrix between PQ nodes is used to cluster the PQ nodes to be partitioned. This embodiment uses the average link hierarchy clustering method for illustration, but the present invention is not limited to this clustering method.

[0119] At the beginning of hierarchical clustering, each PQ node to be partitioned is treated as an independent node cluster:

[0120]

[0121] For any two node clusters and Inter-cluster distance is defined using the average link distance:

[0122]

[0123] in, For node clusters With node clusters The average link distance between them; For the integrated reactive power transmission distance matrix The distance element between node i and node j in PQ.

[0124] In each clustering step, the two clusters with the smallest inter-cluster distance are selected for merging:

[0125]

[0126] The new cluster of nodes after the merger is:

[0127]

[0128] Repeat the above process until the number of node clusters reaches the preset reactive power partition number K, and obtain the reactive power partitioning result:

[0129]

[0130] in, This represents the reactive power partitioning results for the power system. These are the reactive power partitions of the obtained K PQ nodes.

[0131] The reactive power partitioning of the K PQ nodes satisfies:

[0132]

[0133] In this embodiment, K can be set to 4, 5, 6, 7, 8, 9 to obtain reactive power partitioning results under different numbers of partitions, and the results of each partition can be evaluated by reactive power cross-zone transmission rate.

[0134] Step S6: Calculate the reactive power cross-regional transmission rate.

[0135] The node's partition is determined based on the reactive power partitioning results. In this embodiment, PV nodes and balancing nodes do not participate in clustering. Before calculating the reactive power cross-regional transmission rate, PV nodes and balancing nodes are preferentially assigned to the partition of the PQ node directly connected to them. If they are connected to PQ nodes in multiple different partitions, they are assigned to the partition with the largest sum of reactive power transmission amounts calculated according to the following method for the corresponding connecting lines. If they are not directly connected to PQ nodes, their partition is determined layer by layer based on the already assigned adjacent nodes according to the above principles. This process is only used for cross-regional line determination and does not change the reactive power partitioning results of PQ nodes obtained in step S5.

[0136] Let the reactive power partition number to which node i belongs be ? The from and to nodes of line l are f(l) and t(l), respectively. When the two ends of the line belong to different reactive power zones, the line is a cross-zone line, and the set of cross-zone lines is:

[0137]

[0138] The average absolute value of the reactive power flow at both ends of the line is taken as the reactive power transmission amount of line l:

[0139]

[0140] in, This represents the reactive power transmission of line l. and These represent the reactive power flow at the from and to ends of line l, respectively.

[0141] The ratio of the sum of reactive power transmission through inter-regional lines to the sum of reactive power transmission through all branches of the system is taken as the reactive power inter-regional transmission rate.

[0142] ;

[0143] in, This represents the reactive power inter-zone transmission rate. The smaller the reactive power inter-zone transmission rate, the weaker the reactive power transmission across zones, and the more conducive the zoning results are to achieving local reactive power balance.

[0144] II. Results and Comparative Verification of Examples

[0145] Following the steps outlined above, reactive power partitioning was performed on the improved IEEE 118-node test system. The reactive power transmission distribution factor matrix and the reactive power transmission distance matrix between PQ nodes were calculated for each of the 35 feasible operating conditions, and a comprehensive reactive power transmission distance matrix was obtained using an equal-weighting method. Then, using this comprehensive reactive power transmission distance matrix as the distance matrix between PQ nodes, clustering was performed under different preset numbers of reactive power partitions to obtain the corresponding reactive power partitioning results.

[0146] To verify the technical effectiveness of the method described in this invention, this embodiment compares the method of this invention with a zoning method based on voltage-reactive power sensitivity under the same source-load operating conditions, the same number of PQ nodes to be partitioned, the same number of preset partitions, and the same reactive power inter-zone transmission rate evaluation index. Specifically, the method of this invention uses a comprehensive reactive power transmission distance matrix based on the reactive power transmission distribution factor for clustering; the comparison method uses a node distance matrix constructed based on voltage-reactive power sensitivity for clustering. Except for the different distance matrix construction methods, the source-load operating conditions, clustering methods, and evaluation indexes of the two methods are consistent. The evaluation results under different numbers of partitions are shown in Table 1.

[0147] Table 1 Comparison of reactive power transmission rates across zones with different numbers of zones

[0148] 4 14.5622% 21.4954% 35 / 35 5 26.1871% 28.1885% 28 / 35 6 22.8653% 25.9581% 33 / 35 7 23.9598% 26.5956% 33 / 35 8 27.3962% 30.8994% 29 / 35 9 26.7069% 34.9797% 35 / 35

[0149] As shown in Table 1, for different values ​​of the number of partitions K (4-9), the average reactive power transmission rate obtained by the method of this invention is lower than that of the comparative method. This indicates that the comprehensive reactive power transmission distance matrix constructed based on the reactive power transmission distribution factor can effectively reflect the reactive power transmission coupling relationship between PQ nodes. Compared with the partitioning method that constructs the distance matrix solely based on node voltage-reactive power response, the method of this invention can characterize the reactive power transmission connection between nodes from the perspective of reactive power flow response at the line port, resulting in a lower level of inter-regional reactive power exchange in the partitioned results.

[0150] In this embodiment, when the preset number of reactive power partitions K=4, the average reactive power cross-regional transmission rate of the method of the present invention is 14.5622%, which is lower than the 21.4954% of the comparative method, and it outperforms the comparative method in all 35 feasible operating conditions. Therefore, the clustering result corresponding to K=4 can be used as a reactive power partitioning result in this embodiment. The reactive power partitioning results when K=4 are shown in Table 2.

[0151] Table 2 Reactive power partitioning results when K=4

[0152] Partition 1 19 5，43，44，45，47，48，49，50，51，52，53，54，57，58，86，106，108，109，117 Partition 2 23 2，16，17，28，30，33，35，37，38，39，41，59，60，61，63，64，65，66，67，68，71，75，118 Partition 3 9 9，11，13，14，20，21，22，23，25 Partition 4 22 3，7，29，78，79，80，81，82，83，84，88，89，93，94，95，96，97，98，101，102，114，115

[0153] As shown in Table 2, under the condition of K=4, the method of the present invention can form a clear reactive power zoning result. Combined with Table 1, it can be seen that the average reactive power cross-zone transmission rate corresponding to this zoning result is low, indicating that the method of the present invention can enhance reactive power transmission coupling within the zone and weaken reactive power exchange between zones, which is beneficial to achieving local reactive power balance and zone control.

[0154] It should be noted that the different numbers of partitions in Table 1 are only used to illustrate that the method of the present invention can form evaluable reactive power partitioning results under different preset numbers of partitions, and do not constitute a limitation on the scope of protection of the present invention. In practical applications, the final number of reactive power partitions can be determined based on the power grid scale, operation and management requirements, reactive power compensation configuration requirements, and reactive power inter-regional transmission rate evaluation results.

[0155] As can be seen from the above implementation results, the method of the present invention can form reactive power zoning results that are adaptable to multiple operating conditions, taking into account the power output fluctuations and load fluctuations of the power electronic power supply. Through the evaluation of reactive power inter-zone transmission rate, it can be seen that the obtained zoning results can reduce the level of reactive power transmission across zones, which is conducive to enhancing reactive power coupling within zones and weakening reactive power exchange between zones, thereby providing a basis for reactive power voltage zoning control, reactive power compensation planning and reactive power support resource allocation in the power system.

[0156] The technical means disclosed in this invention are not limited to those disclosed in the above embodiments, but also include technical solutions composed of any combination of the above technical features.

Claims

1. A reactive power partitioning method for a power system that takes into account source-load power fluctuations, characterized in that, Includes the following steps: Step S1: Obtain the basic network structure data of the power system to be partitioned, determine all PQ nodes in the system as nodes to be partitioned, and exclude PV nodes and balancing nodes from being partitioned; construct a source-load operating condition set that includes power electronic power output fluctuations and load fluctuations. Step S2: Filter the source load operating conditions according to the application scenario type of the source load operating conditions to form a source load fluctuation scenario set; Step S3: For each feasible operating condition in the source load fluctuation scenario set, taking PQ node as the reactive power disturbance node, calculate the impact of unit reactive power injection disturbance of PQ node on reactive power flow of system line port, and obtain the reactive power transmission distribution factor matrix under the corresponding operating condition. The columns of the reactive power transmission distribution factor matrix correspond one-to-one with the nodes to be partitioned; Step S4: Based on the reactive power transmission distribution factor matrix under each feasible operating condition, construct the reactive power transmission distance matrix under the corresponding operating condition; the reactive power transmission distance matrix, as well as the subsequent comprehensive reactive power transmission distance matrix and reactive power partitioning results, are all constructed with the node to be partitioned as the object; The reactive power transmission distance matrices corresponding to multiple feasible operating conditions are combined to obtain the comprehensive reactive power transmission distance matrix. Step S5: Using the comprehensive reactive power transmission distance matrix as the distance matrix between PQ nodes, cluster the PQ nodes to obtain the reactive power partitioning result of the power system to be partitioned. Step S6: Identify cross-regional lines according to the reactive power zone to which the nodes at both ends of the line belong, calculate the reactive power cross-regional transmission rate, and evaluate the reactive power zone results based on the reactive power cross-regional transmission rate.

2. The reactive power partitioning method for power systems considering source-load power fluctuations according to claim 1, characterized in that, In step S1, the source load operating conditions are constructed through either the planning application operating condition selection method or the operating application operating condition generation method. The method for selecting the planning application conditions includes: acquiring historical operating section data of the power system to be partitioned within a preset historical time interval, and selecting historical operating sections that can reflect the power output fluctuations and load fluctuations of the power electronic power supply from the historical operating section data as the source-load operating condition set; The method for generating operating conditions includes: using the current operating point of the power system to be partitioned as the baseline operating condition, generating a set of candidate source-load operating conditions that includes power electronic power supply output fluctuations and load fluctuations within a preset source-load fluctuation range; specifically, generating multiple candidate source-load operating conditions based on the current operating point, the preset power electronic power supply output fluctuation range, the preset load fluctuation range, and system power balance constraints; wherein, the candidate source-load operating conditions include power electronic power supply output changes, load changes, and power balance adjustment results corresponding to the power electronic power supply output changes and load changes.

3. The reactive power zoning method for power systems considering source-load power fluctuations according to claim 1, characterized in that, In step S2, a set of source load fluctuation scenarios is formed according to the construction method of the source load operating conditions; When the source load operating conditions are constructed through the planning application operating condition selection method, the source load operating condition set composed of the selected historical operating sections that meet the preset conditions is used as the source load fluctuation scenario set. When the source load operating conditions are constructed through the operation application operating condition generation method, AC power flow calculation is performed on each operating condition in the generated candidate source load operating condition set, and the candidate source load operating conditions that have converged power flow and meet the preset operating constraints are included as feasible operating conditions in the source load fluctuation scenario set. The number of feasible operating conditions in the source load fluctuation scenario set is not less than the preset minimum number of scenarios: ； in, This refers to the number of feasible operating conditions in the source load fluctuation scenario set. Set the minimum number of scenes.

4. The reactive power partitioning method for power systems considering source-load power fluctuations according to claim 1, characterized in that, In step S3, the reactive power transmission distribution factor matrix includes the reactive power transmission distribution factor matrix at the line from end and the reactive power transmission distribution factor matrix at the line to end. For any feasible operating condition, let the set of system branches be . The set of PQ nodes to be partitioned is Line l∈ The reactive power flow from and to ends are respectively and PQ node k∈ The reactive power injection amount is Then the reactive power transmission distribution factor at the from end of the line and the reactive power transmission distribution factor at the to end of the line are respectively: ； ； in, This represents the change in reactive power flow at the from end of line l when a unit reactive power injection disturbance occurs at node k of PQ. This represents the change in reactive power flow at the to end of line l when a unit reactive power injection disturbance occurs at node k of PQ. Calculate for all branches and all PQ nodes to be partitioned separately. and The reactive power transmission distribution factor matrix at the from end of the line and the reactive power transmission distribution factor matrix at the to end of the line are obtained as follows: ； ； in, The reactive power transmission distribution factor matrix at the from end of the line. N is the reactive power transmission distribution factor matrix at the to end of the line; B N represents the number of branches in the system. PQ The number of PQ nodes to be partitioned; the k-th column of the reactive power transmission distribution factor matrix at the line from end and the reactive power transmission distribution factor matrix at the line to end respectively represents the reactive power flow response at the line from end and to end of the entire system when a unit reactive power injection disturbance occurs at PQ node k.

5. The reactive power partitioning method for power systems considering source-load power fluctuations according to claim 4, characterized in that, The reactive power transmission distribution factor matrix is ​​obtained analytically by linearizing the AC power flow equations. For any feasible operating condition, the phase angles of the node voltages (excluding the slack node) and the voltage magnitude V of the PQ node are used as state variables. The AC power flow equations are linearized for this feasible operating condition, and the unit reactive power injection disturbance matrix of the PQ node is constructed. The sensitivity matrix of the state variables to the reactive power injection disturbance of the PQ node is then obtained. ； in, The Jacobian matrix of the AC power flow equation at the feasible operating condition; This is the sensitivity matrix of the node voltage phase angle (excluding the slack node) to reactive power injection disturbances at node PQ. This is the sensitivity matrix of the voltage amplitude at node PQ to reactive power injection disturbances at node PQ; It is a zero matrix; This is the unit perturbation matrix corresponding to the PQ node; Based on the partial derivative matrices of the reactive power flow with respect to the state variables at the from and to ends of the line, the reactive power transmission distribution factor matrix at the from end and the reactive power transmission distribution factor matrix at the to end of the line are obtained: ; ; in, and These are the reactive power flow vectors for the from and to ends of the entire system's lines, respectively. This is the phase angle vector of the node voltages, excluding the slack node. The voltage magnitude vector of the PQ node; and These are the partial derivative matrices of the reactive power flow at the from and to ends of the line with respect to the phase angle of the node voltage, respectively. and These are the partial derivative matrices of the reactive power flow at the from and to ends of the line with respect to the voltage magnitude at node PQ, respectively.

6. The reactive power partitioning method for power systems considering source-load power fluctuations according to claim 1, characterized in that, In step S4, the reactive power transmission distance matrix between PQ nodes is constructed as follows: For the s-th feasible operating condition, let the set of branches with node m of PQ as the from end be . The set of branches with node m of PQ as the to end is The set of adjacent branches of node m in PQ is ,and Based on the impact of the unit reactive power injection disturbance at node k of PQ on the reactive power flow of the adjacent branch ports of node m of PQ, calculate the reactive power transmission coupling strength from node k of PQ to node m of PQ: ； in, Let be the reactive power transmission coupling strength between PQ node k and PQ node m under the s-th feasible operating condition; and These are the elements in the reactive power transmission distribution factor matrix of the line from and to ends under the s-th feasible operating condition; Let m be the number of adjacent branches of node m in PQ; The reactive power transmission coupling strength is normalized to obtain the normalized reactive power transmission coupling strength: ； in, The normalized reactive power transmission coupling strength is given under the s-th feasible operating condition. Let PQ be the set of nodes to be partitioned; ε is a very small positive number to prevent the denominator from being zero. By symmetricizing the normalized reactive power transmission coupling strength, we obtain the symmetrical reactive power transmission coupling strength between PQ node m and PQ node k: ； in, Let be the symmetrical reactive power transmission coupling strength between PQ node m and PQ node k under the s-th feasible operating condition; The symmetrical reactive power transmission coupling strength is converted into reactive power transmission distance between PQ nodes: ， ； And order: ； This yields the reactive power transmission distance matrix between PQ nodes under the s-th feasible operating condition: ; in, Let be the reactive power transmission distance matrix between PQ nodes under the s-th feasible operating condition. The number of PQ nodes to be partitioned; is the reactive power transmission distance between node m and node k under the s-th feasible operating condition. The smaller the value, the stronger the reactive power transmission coupling between the two nodes, and the more likely they are to be assigned to the same reactive power partition.

7. The reactive power partitioning method for power systems considering source-load power fluctuations according to claim 1, characterized in that, In step S4, the reactive power transmission distance matrices between PQ nodes corresponding to multiple feasible operating conditions are synthesized, including weighted summation of the reactive power transmission distance matrices between PQ nodes under each feasible operating condition according to preset weights, to obtain the synthesized reactive power transmission distance matrix: ； in, This is a comprehensive reactive power transmission distance matrix; This is a set of source load fluctuation scenarios; Let be the reactive power transmission distance matrix between PQ nodes under the s-th feasible operating condition; The preset weights for the s-th feasible operating condition satisfy: ， ； The preset weights include any one of equal weights, operational probability weights, or voltage stability risk weights; when equal weights are used, the preset weights satisfy the following: ； in, This refers to the number of feasible operating conditions in the source load fluctuation scenario.

8. The reactive power partitioning method for power systems considering source-load power fluctuations according to claim 1, characterized in that, In step S5, the comprehensive reactive power transmission distance matrix is ​​used as the distance matrix between PQ nodes. A preset clustering method is used to cluster the PQ nodes to be partitioned, and K PQ node reactive power partitions are obtained according to the preset number of reactive power partitions K. The preset clustering method is one that takes a distance matrix as input and outputs node category labels; the reactive power partitioning result is expressed as follows: ； in, This is the result of reactive power zoning in the power system. These are the reactive power partitions of the obtained K PQ nodes; The reactive power partitioning of the K PQ nodes satisfies: ， ， ； in, The set of PQ nodes to be partitioned. and These are the reactive power partitions of the a-th and b-th PQ nodes, respectively.

9. The reactive power partitioning method for power systems considering source-load power fluctuations according to claim 1, characterized in that, In step S6, the reactive power partition to which the node belongs is determined based on the reactive power partitioning results, and cross-regional lines are identified based on whether the reactive power partitions to which the nodes at both ends of the line belong are the same. Among them, for PV nodes and balanced nodes that are not to be partitioned, their reactive power partitions are determined according to the network connection relationship between them and the reactive power transmission of the PQ nodes that have been partitioned. The attribution is only used for cross-regional line identification and reactive power cross-regional transmission rate calculation, and does not change the reactive power partitioning result of the PQ nodes. Let the system branch set be The from and to nodes of line l are f(l) and t(l) respectively, and the reactive power partition number of node i is... Then the set of cross-regional routes is: ； The average absolute value of reactive power flow at both ends of the line is taken as the reactive power transmission amount of line l: ； The ratio of the sum of reactive power transmission of inter-regional lines to the sum of reactive power transmission of all branches in the system is taken as the reactive power inter-regional transmission rate: ； in, For cross-regional routes; This represents the reactive power transmission of line l. and These represent the reactive power flow at the from and to ends of line l, respectively. The reactive power cross-zone transmission rate is defined as follows: the smaller the reactive power cross-zone transmission rate, the weaker the reactive power transmission across zones, and the better the reactive power zoning result.

10. A computer-readable storage medium storing instructions, characterized in that, When the instruction is executed, it can realize the power system reactive power partitioning method that takes into account source load power fluctuations as described in any one of claims 1 to 9.

Images