Power flow convergence method and system based on direct current power flow preprocessing and stage decoupling

Abstract

The application provides a power flow convergence method and system based on DC power flow preprocessing and stage decoupling, and belongs to the field of power system analysis and control. The method comprises the following steps: performing multi-stage checking and parameter adjustment on original power flow data, performing preliminary checking on active power balance of a power system, and generating an initial active power setting value set of nodes; performing DC power flow calculation based on the initial active power setting value set of the nodes, obtaining initial phase angles of the nodes, identifying heavy-load lines with power angle differences exceeding a preset threshold, and obtaining an initial phase angle distribution through adjustment; combining a preset initial voltage amplitude with the initial phase angle distribution to perform stage decoupling power flow calculation and adjustment, and obtaining power flow parameters; taking the power flow parameters as initial values to perform full-coupling power flow calculation, and obtaining a final power flow solution meeting a convergence accuracy requirement. The application can intelligently diagnose and locate a physical root cause of a convergence problem, effectively improves convergence and robustness of power flow calculation under complex working conditions through targeted adjustment in a small range.

Description

Technical Field

[0001] This invention belongs to the field of power system analysis and control technology, and particularly relates to a power flow convergence method and system based on DC power flow preprocessing and staged decoupling. Background Technology

[0002] The statements in this section are merely background information related to the present invention and do not necessarily constitute prior art.

[0003] With the deepening of the "dual carbon" goals, the global energy structure is undergoing a profound transformation, and the new power system is showing the core characteristics of large-scale and high-proportion grid connection of new energy sources. New energy power generation technologies such as wind and solar power have become the main source of incremental power generation due to their clean and low-carbon advantages. However, the output of such power sources is significantly affected by natural conditions, exhibiting strong volatility, intermittency, and spatiotemporal uncertainty, which is in stark contrast to the stable output characteristics of traditional synchronous generators.

[0004] This characteristic leads to extremely frequent and drastic dynamic adjustments in power grid operation. In scenarios with significant fluctuations in renewable energy output and a continuous increase in cross-regional power exchange, the distribution of active and reactive power in the power system is often in an extreme spatiotemporal imbalance state. This causes the solution of power flow equations to fall into ill-conditioned regions, increasing the condition number of the Jacobian matrix and decreasing numerical stability, posing a severe challenge to traditional power flow calculation methods. In current power grid operation, extreme conditions such as heavy loads on cross-sections, weak local voltage support, and limited system balance constraints frequently occur, directly causing power flow calculation values ​​to easily deviate from the convergence region, leading to frequent oscillations or even complete non-convergence in numerical iterations. This problem not only affects the efficiency of core operations such as power grid planning and design, operation and scheduling, but may also delay critical decisions such as fault handling and situational awareness due to the inability to effectively solve power flow problems, threatening the safe and stable operation of the power system.

[0005] Existing power flow convergence improvement techniques have significant limitations: On the one hand, most methods only optimize algorithm steps at the numerical iteration level, lacking in-depth exploration of the physical essence of power flow non-convergence under fluctuating operating conditions, making it difficult to accurately locate the weak links causing computational divergence; on the other hand, the fully coupled solution mechanism used in the traditional Newton-Raphson method is prone to interference between corrections in different dimensions due to ill-conditioned Jacobian matrix under extreme conditions of high new energy penetration, resulting in numerical coupling oscillations; while the traditional PQ decomposition method, although decoupled, often suffers from problems when dealing with heavily loaded sections or weak voltage support points due to… P-θ and QU The alternating mapping of residuals between dimensions leads to a circuitous or even divergent convergence path. Summary of the Invention

[0006] To overcome the shortcomings of the prior art, this invention provides a power flow convergence method and system based on DC power flow preprocessing and phased decoupling. It aims to solve the problems that, in the context of new power system environments, the large-scale integration of new energy sources and the uncertainty of their output lead to drastic fluctuations in the operating conditions of the power grid, which in turn cause the nonlinear coupling of the power flow equations to intensify, the Jacobian matrix to become ill-conditioned, and traditional power flow calculation methods to easily deviate from the convergence domain and make it difficult for numerical solutions to converge stably when facing complex operating conditions.

[0007] To achieve the above objectives, one or more embodiments of the present invention provide the following technical solutions:

[0008] The first aspect of this invention provides a power flow convergence method based on DC power flow preprocessing and staged decoupling;

[0009] Power flow convergence methods based on DC power flow preprocessing and staged decoupling include:

[0010] The original power flow data is subjected to multi-level verification and parameter adjustment, and the initial verification of the system active power balance is performed to generate a set of initial values ​​for node active power settings that meet basic physical feasibility.

[0011] DC power flow calculation is performed based on the set of initial values ​​of the active power of the nodes, the initial phase angle of each node is obtained, the heavily loaded lines whose power angle difference exceeds the preset threshold are identified, and the power angle difference of all lines is satisfied by adjusting the local active power resources to obtain the initial phase angle distribution.

[0012] By combining the preset initial voltage amplitude with the initial phase angle distribution, staged decoupled power flow calculation and adjustment are performed to obtain power flow parameters;

[0013] Using the obtained power flow parameters as initial values, a fully coupled power flow calculation is performed to obtain the final power flow solution that meets the convergence accuracy requirements.

[0014] As a further technical solution, the process involves multi-level verification and parameter adjustment of the original power flow data, and preliminary verification of the system's active power balance, generating a set of initial node active power settings that meet basic physical feasibility, including:

[0015] The original power flow data is subjected to network topology verification and device parameter verification. The network topology verification includes identifying and removing isolated buses, verifying the consistency of balancing machine configuration and circuit breaker status. The device parameter verification includes verifying the non-negativity and rationality of generator, transformer, and line parameters, as well as the rationality of node power and voltage setting values.

[0016] The initial total active power imbalance of the system is calculated, and it is determined whether the power to be borne by the balancing machine exceeds its preset upper and lower power limits. If the power of the balancing machine exceeds the limit, the excess power is distributed proportionally to other controllable generators in the system other than the balancing machine, so as to obtain the initial set of node active power settings that meet the basic physical feasibility.

[0017] As a further technical solution, the step of performing DC power flow calculation based on the initial value set of the active power of the nodes to obtain the initial phase angle of each node includes:

[0018] Substituting the active power injection power of all nodes in the initial set of active power settings into the DC power flow equation matrix, as shown below:

[0019]

[0020] in, This represents the active power injection amount for all unbalanced nodes. The phase angles to be solved for all non-equilibrium nodes; Here is the admittance matrix;

[0021] Solving the linear equation yields the initial phase angle of each non-equilibrium node. Setting the phase angle of the equilibrium node to 0 gives the initial phase angle of all nodes in the network.

[0022] As a further technical solution, the identification of heavily loaded lines with a power angle difference exceeding a preset threshold includes:

[0023] Based on the DC power flow calculation results, calculate the absolute value of the power angle difference between the two ends of all lines;

[0024] The absolute value of the power angle difference is compared with a preset static stability safety threshold to identify lines that exceed the limit;

[0025] Using the sending and receiving busbars of each over-limit line as the center, a local topology adjustment area is delineated using a breadth-first search algorithm;

[0026] According to the preset priority order, the distribution of active resources in the local topology area is adjusted, and the active power output of the opposite area is adjusted synchronously to maintain the overall active power balance of the system.

[0027] Recalculate the DC power flow until the absolute value of the power angle difference of all lines is less than the static stability safety threshold.

[0028] As a further technical solution, the phased decoupling of power flow calculation and adjustment includes:

[0029] Phase 1: Fixed voltage amplitude, only... P-θThe active power flow is solved iteratively. If the iteration fails to converge, the weak bus is located based on the maximum phase angle correction, and topology search and active power resource adjustment are performed on its associated region until... P-θ Iterative convergence;

[0030] Second stage: Based on the phase angle distribution obtained in the first stage, fix the voltage phase angle and only perform... QU The reactive power flow is solved iteratively. If the iteration fails to converge, the weak bus is located based on the maximum voltage correction, and a topology search and reactive power resource adjustment are performed on its associated region until... QU Iterative convergence.

[0031] As a further technical solution, the method involves locating the weak busbar based on the maximum phase angle correction, and then performing topology search and active power resource adjustment on its associated region until... P-θ Iterative convergence includes:

[0032] when P-θ If the iteration reaches the preset maximum number of iterations and still fails to converge, extract the maximum phase angle correction amount within the iteration cycle that caused the non-convergence and its corresponding bus node.

[0033] Starting from the aforementioned busbar, a breadth-first search with a finite number of steps is used to determine the key adjustment area;

[0034] Perform at least one of the following operations on the critical adjustment area: verify and correct line and transformer parameters, adjust the active power output of generators in the area according to the principle of sensitivity or proportionality, and reduce non-critical loads in the area.

[0035] As a further technical solution, the method involves locating weak buses based on the maximum voltage correction amount and performing topology search and reactive power resource adjustment on their associated areas until... QU Iterative convergence includes:

[0036] when QU If the iteration fails to converge after reaching the preset maximum number of iterations, extract the maximum voltage correction amount within the iteration period that caused the non-convergence and its corresponding bus node.

[0037] Starting from the aforementioned busbar, a breadth-first search with a finite number of steps is used to determine the key adjustment area;

[0038] The reactive power status of the region is determined based on the sign of the maximum voltage correction. If the maximum voltage correction is positive, capacitive reactive power equipment is put into operation or inductive reactive power equipment is cut off according to priority. If the maximum voltage correction is negative, inductive reactive power equipment is put into operation or capacitive reactive power equipment is cut off according to priority.

[0039] As a further technical solution, the step of performing fully coupled power flow calculations based on the obtained power flow parameters as initial values ​​to obtain a final power flow solution that meets the convergence accuracy requirements includes:

[0040] Using the power flow parameters obtained after phased decoupling and adjustment as the initial values ​​for iteration, the active power imbalance and reactive power imbalance of each node in the kth iteration are calculated.

[0041] Construct the Jacobian matrix or coefficient matrix at the current iteration point, solve the correction equation to obtain the phase angle correction and voltage magnitude correction for each node, and correct the voltage magnitude and phase angle of each node respectively;

[0042] Determine whether the active power imbalance and reactive power imbalance at each node satisfy the convergence condition. If both conditions are met, output the final power flow solution; otherwise, set... k = k +1, continue iterating until the convergence accuracy requirement is met.

[0043] The second aspect of the present invention provides a power flow convergence system based on DC power flow preprocessing and staged decoupling.

[0044] A power flow convergence system based on DC power flow preprocessing and staged decoupling includes:

[0045] The data verification module is configured to perform multi-level verification and parameter adjustment on the raw power flow data, and to perform preliminary verification of the system active power balance to generate a set of initial values ​​for node active power settings that meet basic physical feasibility.

[0046] The DC power flow preprocessing module is configured to: perform DC power flow calculation based on the set of initial values ​​of the active power of the nodes, obtain the initial phase angle of each node, identify the heavily loaded lines whose power angle difference exceeds a preset threshold, adjust the local active power resources to make the power angle difference of all lines meet the constraints, and obtain the initial phase angle distribution.

[0047] The phased decoupled power flow calculation and adjustment module is configured to: combine the preset initial voltage amplitude with the initial phase angle distribution to perform phased decoupled power flow calculation and adjustment to obtain power flow parameters;

[0048] The fully coupled power flow verification module is configured to perform fully coupled power flow calculations based on the obtained power flow parameters as initial values, and obtain the final power flow solution that meets the convergence accuracy requirements.

[0049] A third aspect of the present invention provides a computer-readable storage medium having a program stored thereon, which, when executed by a processor, implements the steps in the power flow convergence method based on DC power flow preprocessing and phased decoupling as described in the first aspect of the present invention.

[0050] A fourth aspect of the present invention provides an electronic device, including a memory, a processor, and a program stored in the memory and executable on the processor, wherein the processor executes the program to implement the steps in the power flow convergence method based on DC power flow preprocessing and phased decoupling as described in the first aspect of the present invention.

[0051] The above one or more technical solutions have the following beneficial effects:

[0052] (1) This invention eliminates parameter configuration errors through data verification, then quickly locates and adjusts heavily loaded lines by utilizing the linearization characteristics of DC power flow, and finally precisely pinpoints specific weak points through the correction amount analysis in the phased decoupling iteration. This diagnostic path from macro to micro and from parameters to physics completely changes the traditional method's reliance on manual experience for trial and adjustment, and realizes intelligent diagnosis and precise location of the root cause of the problem.

[0053] (2) High-quality initial phase angle values ​​are obtained through DC power flow preprocessing, providing a good starting point for subsequent nonlinear iterations. A fully phased decoupling strategy of "global active power first, then local reactive power" is adopted, effectively cutting off the traditional PQ decomposition method. P-θ and QU The iterative process involves cross-coupling oscillation paths. When any iteration fails to converge, a regional adjustment mechanism based on the maximum correction amount can implement precise intervention within a minimal scope, avoiding blind disturbances to the entire system. Finally, the high-precision approximate solution obtained in the preceding stages is used as the initial value, enabling the fully coupled power flow calculation to converge rapidly within a very small number of iterations. The entire scheme forms a complete closed loop from preprocessing and decoupling adjustment to global verification, significantly improving the convergence and numerical robustness of power flow calculations under complex conditions such as renewable energy fluctuations and heavy load operation.

[0054] (3) In the DC preprocessing stage, this invention adjusts active power resources only in local areas of lines with abnormal power angle differences; in the phased decoupling stage, the adjustment area with the smallest impact range is precisely delineated through a finite number of topology searches based on the maximum phase angle or voltage correction. The adjustment methods are prioritized according to equipment response speed and adjustment cost, ensuring the economy and speed of adjustment. This mechanism can solve the convergence problem while maintaining the stability of the original system operating state to the greatest extent, providing reliable and efficient decision support for dispatchers.

[0055] Advantages of additional aspects of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. Attached Figure Description

[0056] The accompanying drawings, which form part of this invention, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an improper limitation of the invention.

[0057] Figure 1 This is a flowchart of the method in the first embodiment.

[0058] Figure 2 This is a schematic diagram of the DC power flow preprocessing flow of the first embodiment.

[0059] Figure 3 This is a schematic diagram of the secondary topology search in the first embodiment.

[0060] Figure 4 For the first embodiment P-θ A schematic diagram of the iterative active power flow adjustment process.

[0061] Figure 5 For the first embodiment QU A schematic diagram of the iterative reactive power flow adjustment process.

[0062] Figure 6 This is a schematic diagram of the power flow path and convergence region in the first embodiment.

[0063] Figure 7 This is a system structure diagram of the second embodiment. Detailed Implementation

[0064] It should be noted that the following detailed descriptions are exemplary and intended to provide further illustration of the invention. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0065] It should be noted that the terminology used herein is for the purpose of describing particular implementations only and is not intended to limit the exemplary implementations of the present invention.

[0066] Where there is no conflict, the embodiments and features in the embodiments of the present invention can be combined with each other.

[0067] The overall approach proposed in this invention is as follows: First, through multi-level data verification and preliminary active power balance check, initial value deviations caused by unreasonable original parameter configurations and active power supply-demand imbalances are eliminated; second, high-quality initial phase angle values ​​are quickly constructed using linearized solutions of the DC power flow model, and the identification and adjustment of overloaded sections are achieved based on the power angle difference threshold warning; subsequently, following the physical logic of "global active power first, then local reactive power," the following steps are performed respectively... P-θ and QUThe process involves independent iterations, and the system's weak points are located using the maximum phase angle correction and maximum voltage correction. A regional power flow convergence improvement strategy is implemented using breadth-first search to guide the power flow back to the convergence domain. Finally, the high-quality approximate solution obtained after phased decoupling and adjustment is used as the starting point for iteration, and the final verification is performed through fully coupled power flow calculation.

[0068] Example 1

[0069] This embodiment discloses a power flow convergence method based on DC power flow preprocessing and staged decoupling;

[0070] like Figure 1 As shown, the power flow convergence method based on DC power flow preprocessing and staged decoupling includes:

[0071] Step S1 involves performing multi-level verification and parameter adjustment on the original power flow data, and conducting a preliminary check of the system's active power balance to generate a set of initial node active power settings that meet basic physical feasibility.

[0072] First, network topology verification is carried out. A power grid topology traversal algorithm is used to identify and remove isolated buses without connected branches to ensure that all operating nodes are connected to the main grid. Verification and assurance are made that each regional sub-grid is equipped with at least one balancing machine. If there is a case where no balancing machine is configured, the synchronous generator with the largest capacity in the region is automatically designated as the balancing machine. The status of the circuit breakers at both ends of each line is compared. If there is an inconsistency where one end is closed and the other end is open, it is corrected to a unified status according to the power grid operation specifications.

[0073] Then, equipment parameter verification was performed, checking the non-negativity of the impedance parameters of all generators and transformers, and simultaneously verifying the lines and transformers. X / R Check whether the parameters are within the reasonable range of the project; check the sign of the reactive power compensation equipment capacity and correct capacitive and inductive configuration errors; verify the active and reactive power configurations of all PQ nodes to ensure that their power factors are within the reasonable range and avoid extreme configurations; verify the voltage settings of all PV nodes to ensure that the voltage settings of PV nodes are within the acceptable range allowed by the power grid operation standards.

[0074] Furthermore, a preliminary verification of the system's active power balance is conducted, including: active power balance check, power flow problem localization, and strategies to improve power flow convergence. Specifically:

[0075] When dealing with power flow problems in large power grids, it is necessary to first verify the active power balance constraints of the system to ensure that the total active power supply and demand of the entire system are roughly balanced. If there is a huge difference between the total generation and the total load in the initial stage, the balancing machine will not be able to handle this excessively large unbalanced power, which will directly lead to the failure of subsequent power flow calculations.

[0076] First, the initial unbalanced power of the system is calculated, as shown in the following formula:

[0077] (1)

[0078] In the formula, This refers to the power imbalance of the system. The total load power of the system. For system line loss, Given the total power generation of the system, the system line loss formula is as follows:

[0079] (2)

[0080] in, m This is a proportionality coefficient, typically around 1%-5%. Then, calculate the total power that the system balancing machine should handle:

[0081] (3)

[0082] In the formula, Set the power for the balancing machine. The total power required for the balancing machine.

[0083] Secondly, in the process of locating the trend problem, all balancing mechanisms are traversed to determine the responsibilities they need to bear. Are the values ​​within the preset maximum and minimum upper and lower limits of the balancing machine? If the power limit of the balancing machine is outside the range, it means that the initial active power configuration for the current operating condition is physically infeasible and there is a risk of exceeding the limit. Therefore, the correction process must be initiated.

[0084] Once the balancing machine's power is determined to be exceeding the limit, a power flow convergence improvement strategy is adopted, initiating an adaptive adjustment mechanism based on minimum intervention and proportional allocation to reduce the balancing machine's power exceeding the limit. The active power is distributed reasonably to other controllable generators within the system to achieve a new active power balance.

[0085] (4)

[0086] By selecting generators within the region, excluding balancing machines, whose current active power output has not reached its power limit, an adjustable generator set is formed. The power exceeding the limit will be located. The power is allocated to the adjustable generator set according to the original generator node power ratio. Each generator in the system. The allocation formula is as follows:

[0087] (5)

[0088] in, This represents the adjusted active power of the k-th generator. Let be the original active power of the k-th generator; This represents the active power adjustment of the k-th generator.

[0089] The final output is a set of initial values ​​for the nodal active power settings that satisfy the basic active power balance. It is used for subsequent DC power flow preprocessing.

[0090] Step S2: Based on the initial set of active power values ​​for the nodes, perform DC power flow calculations to obtain the initial phase angle of each node. Identify heavily loaded lines whose power angle difference exceeds a preset threshold. Adjust local active power resources to ensure that the power angle difference of all lines meets the constraints, thus obtaining the initial phase angle distribution, as shown below. Figure 2 As shown.

[0091] Step S21: After completing the preliminary balance check of the system's active power and obtaining the set of active power setpoints that satisfy the constraints. Then, the rapid computational advantage of the DC power flow model is used to determine the initial phase angle distribution of the system. This model is based on simplifying assumptions and linearizes the complex nonlinear power flow equations.

[0092] The following assumptions are made:

[0093] (1) The voltage at each node of the power system is usually near the rated voltage, and is approximately considered to be .

[0094] (2) The voltage phase angle difference between the two ends of the line is very small, so it is assumed that... Therefore, there is , .

[0095] (3) In ultra-high voltage power grids, the line resistance is much smaller than the reactance, therefore the resistance is negligible. .

[0096] Based on the above assumptions, nodes in the traditional AC power flow equation i To the node j active power transmitted It is linearized and depends only on the phase angle difference between the two ends and the line reactance. Its calculation formula is as follows:

[0097] (6)

[0098] In the formula, For nodes i To the node j The active power transmitted, The difference in phase angle between the two ends, For line reactance, This is the line susceptance.

[0099] Will Active power injection of all nodes Substituting the values, we obtain the matrix form of the DC power flow equations for the entire system:

[0100] (7)

[0101] In the formula, This represents the active power injection amount for all unbalanced nodes. The phase angles to be solved for all non-equilibrium nodes are given, and the phase angles for the equilibrium nodes are given. Set the reference value to 0. Let be the admittance matrix, with its diagonal and off-diagonal elements as follows:

[0102] (8)

[0103] in, The self-susceptance between nodes.

[0104] By solving this simplified linear equation, the initial phase angle of each node can be calculated very quickly.

[0105] Step S22: Locate the power flow problem and identify heavily loaded lines whose power angle difference exceeds a preset threshold.

[0106] Calculate the absolute value of the phase angle difference between the two busbars at both ends of all transmission lines in the DC power flow solution. :

[0107] (9)

[0108] In DC power flow diagnosis, the power angle difference between the two busbars at both ends of the line is considered. Compared with the preset static stability threshold A comparison is made. This threshold is usually set in the range of 30° to 45°, which is significantly less than the theoretical static stability limit, and serves as a safety warning line for power flow operation.

[0109] If the absolute value of the phase angle difference of a certain transmission line Exceeding the preset threshold The system immediately determines that the line is under severe overload or that there is a fundamental active power imbalance in its region. This large power angle difference means that its static stability margin is insufficient, and the power flow equations in this region are highly nonlinear, which may lead to convergence failure in subsequent staged decoupling calculations. The system will immediately initiate a local active power adjustment strategy.

[0110] Step S23: Taking the two busbars at both ends of the AC transmission line with abnormal power angle as the center, a breadth-first search (BFS) algorithm with a finite number of steps (two or three levels of neighborhood) is used, such as... Figure 3As shown. Quickly define a local topology region that includes all adjacent power supply and load nodes of the sending node. and the local topology region containing all adjacent nodes of the receiving node. .

[0111] In local topological regions as well as Internally, the system systematically identifies all controllable active power devices, including synchronous generators, energy storage, wind power, and photovoltaics, and obtains their current active power output in real time. and its effective upper and lower limits The adjustability margins of various types of equipment are calculated.

[0112] After completing resource identification and margin assessment of all adjustable active power equipment within the local topology area, power flow convergence improvement measures are initiated. The core objective of these measures is to adjust the active power balance within the area through the principles of minimal intervention and fastest response speed, thereby eliminating overload on lines with abnormal phase angles and... Reduced to a safe threshold c To avoid concentrating the effects of adjustments on the balancing bus, if active power injection is increased at the receiving end of an abnormal line, the system should correspondingly reduce generator output at the sending end of the transmission line to maintain a balance between total system power generation and total load demand.

[0113] To adjust the receiving end Taking the region's merits as an example, the focus of the adjustment will first be on the region. , The power supply within the system. Given the rapid and bidirectional regulation capabilities of energy storage, the first priority is to dispatch energy storage devices in local areas to fully utilize the system's energy storage resources.

[0114] When energy storage regulation capacity is insufficient, the second stage begins, adjusting the output of the synchronous generator, which serves as the primary power source. The adjustment amount is constrained within its adjustable margin range and allocated according to the current generator output ratio.

[0115] If, after the above two-stage power flow convergence improvement measures, the power angle difference of the abnormal line still has not been reduced to below the safety threshold, the output of wind power and photovoltaic units will be adjusted according to the current output ratio.

[0116] If the line is still overloaded, then identify the local area at the receiving end. Standby generator sets that are out of service but not under maintenance should be set to start status.

[0117] If all power regulation measures have been implemented, but the abnormal line is still overloaded, then the final safeguard measure should be activated, and the receiving end should be disconnected. Partial load on the region.

[0118] Adjusting the receiving end After the region's active power output is completed, the sending end should be adjusted accordingly based on the above priority order. The active power output of the region. After implementing any stage of power flow convergence improvement measures, this invention will recalculate the DC power flow and check the power angle difference of all branches. This adjustment-calculation cycle will continue until the power flow converges and the power angle difference of all lines meets the threshold constraint. This concludes the preprocessing based on DC power flow.

[0119] The final phase angles of each node obtained from the DC power flow preprocessing Vector and preset voltage amplitude This combination provides a high-quality initial power flow value with a reasonable phase angle distribution for subsequent phased decoupling iterations. .

[0120] Step S3: Combine the preset initial voltage amplitude with the initial phase angle distribution to perform phased decoupled power flow calculation and adjustment to obtain power flow parameters.

[0121] Unlike the standard PQ decomposition method P-θ and QU The mechanism of alternating correction within a single large cycle employs a decoupling strategy with absolute temporal independence in this embodiment. The system blocks the numerical interaction between the two by controlling logic parameters, only when the voltage amplitude is locked. P-θ Phase locking is only enabled after the iteration converges globally. QU Iteration. This mechanism prevents unconverged residuals from repeatedly coupling and oscillating between the two iterative dimensions due to alternating corrections.

[0122] The phased decoupling power flow convergence improvement described in this embodiment is achieved through logical reconstruction of the standard PQ decomposition method, realizing a three-mode switching operation mechanism based on variable instructions. By introducing two core parameters, 'Decoupled P Solution Logic' and 'Decoupled Q Solution Logic', the system can flexibly switch algorithm modes according to the calculation stage. When 'Decoupled P Solution Logic' is set to True and 'Decoupled Q Solution Logic' is set to False, the solver is a single... P-θ Iterative active power flow calculation; when 'Decoupled P Solution Logic' is False and 'Decoupled QSolution Logic' is True, the solver is a single... QUIterative reactive power flow calculation; when both are set to True, the algorithm logic is the traditional PQ decomposition method.

[0123] To clarify the computational principles of each mode, the power correction equations for the nodes are given below:

[0124] (10)

[0125] In the formula, This refers to the imbalance of active power. This refers to the reactive power imbalance. This is the sensitivity matrix of active power to voltage phase angle; This is the sensitivity matrix of active power to voltage amplitude; This is the sensitivity matrix of reactive power to voltage phase angle; This is the sensitivity matrix of reactive power to voltage amplitude; This is the phase angle correction amount; This is the voltage correction amount; This represents the voltage amplitude.

[0126] The elements of the Jacobian matrix are expressed as follows:

[0127] (11)

[0128] (12)

[0129] (13)

[0130] (14)

[0131] (15)

[0132] (16)

[0133] (17)

[0134] (18)

[0135] in, for H The off-diagonal elements of the matrix; for N The off-diagonal elements of the matrix; for J The off-diagonal elements of the matrix; for L The off-diagonal elements of the matrix; for H The diagonal elements of a matrix; These are the diagonal elements of matrix N; for J The diagonal elements of a matrix; for L The diagonal elements of a matrix; For connecting nodes i With nodes j The negative values ​​of the conductance of the branches between them; for i Nodes and j The phase angle difference between nodes.

[0136] Because high-voltage transmission networks typically have high... X / R Due to the characteristics of the ratio, the branch susceptance is much greater than the conductance. Based on the expressions for each element in the Jacobian matrix in Equation 11-18, the correction matrix for active power on voltage amplitude can be obtained. N The correction matrix for voltage amplitude is much lower than that for reactive power. L Similarly, the reactive power correction matrix for voltage phase angle... J The ones are also much smaller than H Matrix. Therefore PU and Q-θ The nodes are weakly coupled, and the PQ decomposition method simplifies the node voltage correction equation to the following formula.

[0137] (19)

[0138] The traditional PQ decomposition method will 2( n The system of linear equations of order -1 becomes two n The system of linear equations with coefficients of -1 is simplified to a symmetric matrix that remains unchanged during iteration. and .

[0139] (20)

[0140] (twenty one)

[0141] The nodal power imbalance expressed in polar coordinates is:

[0142] (twenty two)

[0143] (twenty three)

[0144] In the formula, For the first i Active power imbalance at nodes; For the first i The given active power of the node; For the first iThe reactive power imbalance at the node; For the first i The given reactive power of the node.

[0145] In the iterative process of the traditional PQ decomposition method, the phase angle correction amount and voltage correction amount The updates continue to alternate within the same outer loop. While this alternating update mechanism is faster than the fully coupled Newton-Raphson method, it may fail under heavy load conditions if... P-θ It has not yet subsided, and what followed... QU Iteration generates new voltage correction values The newly revised voltage is obtained through Equation 22. This can have an impact, causing the convergence path to become circuitous or to oscillate at the convergence edge.

[0146] To avoid the disturbance to convergence caused by this alternating update mechanism, in this embodiment, the power flow calculation is decomposed into two stages, wherein the first stage is... P-θ Iterative phase, the second phase is QU The iterative phase consists of two completely independent yet continuous phases.

[0147] Since active power balance involves all generators and loads in the system, its imbalance... Adjustments need to be made through the active power sources across the entire network, exhibiting a global nature. Reactive power, on the other hand, is primarily used to support voltage, and its imbalance... Typically, regulation is achieved through locally configured reactive power sources, exhibiting significant locality. Based on these attributes, a strategy is adopted to address global issues first, followed by local issues, starting with... P-θ Iterative improvement of active power flow convergence, followed by further improvement based on... QU The convergence of reactive power flow is improved through iteration.

[0148] Step S31, based on P-θ Iterative active power flow adjustment, such as Figure 4 As shown.

[0149] Step S311: In the phased decoupling power flow convergence improvement, in the first stage, the solver parameters 'Decoupled P Solution Logic' are set to True and 'Decoupled Q Solution Logic' to False. The voltage magnitude remains constant during the iteration process, and only the active power equation is considered for iterative calculation of the active power flow. The calculation steps are as follows:

[0150] (1) Substitute the initial power flow values ​​obtained from DC power flow preprocessing Set, keeping the voltage amplitude constant during the iteration process, let the number of iterations be... n = 0;

[0151] (2) Substitute into Equation 22 to calculate the active power imbalance at each node. and find ;

[0152] (3) Solve Formula 20 to obtain the voltage phase angle correction for each node. ;

[0153] (4) Correct the phase angle of the voltage at each node. :

[0154] (twenty four)

[0155] (5) Judgment n If the maximum number of iterations has been reached, proceed to step (7); otherwise, proceed to step (6).

[0156] (6) Determine the active power imbalance at each node. Check if the convergence condition is met. If it is met, proceed to step (7); if not, set the iteration count to zero. n = n +1, and return to step (2) for iteration;

[0157] (7) Output the final power flow data set.

[0158] Step S312: Locating the power flow problem where the active power does not converge.

[0159] when P-θ The maximum active power imbalance within a preset maximum number of iterations. Still unable to meet the preset convergence accuracy When this occurs, it indicates that the system operating conditions corresponding to the current scheduling settings may have approached or exceeded the static stability limit, and the calculation model cannot find a solution under the current settings.

[0160] In this embodiment, if the power flow calculation fails to converge after reaching the maximum number of iterations, a correction based on the maximum phase angle is performed. Problem identification and regional adjustment strategies. The essence of active power transmission in a network is driven by the phase angle difference between nodes. During the iteration process, the active power imbalance... The power residual only reflects the current node and is a numerical indicator characterizing the balance gap; while the phase angle correction... This reflects the degree of phase angle correction the system attempts to eliminate the difference, and is an indicator of its balancing ability. Phase angle correction amount The larger the value, the greater the active power balancing capacity required by this node, and the easier it is to reach its node balancing capacity limit. Compared to , This reflects the fundamental reason why the system's active power does not converge.

[0161] locking The bus in question is actually located at the physical weak point in the power grid where active power dispatching resistance is greatest. Using this bus as the starting point, the adjustment area can be delineated using BFS, which can directly locate the cause of iterative divergence, thereby guiding the power flow calculation from the edge of the physical limit back to the effective convergence region.

[0162] In each iteration of the calculation, the system records the active power imbalance of all nodes in the network in real time. With phase angle correction Simultaneously extract and output the maximum phase angle correction within the current iteration cycle. and its corresponding bus nodes. Focus on observing the period approaching the maximum iteration count. Bus node. If the maximum phase angle correction is always located at a specific bus in multiple consecutive iterations, it indicates that the active power balance capacity at that node has reached the physical limit, and the bus can be determined as the physical weak point that causes the overall power flow calculation to diverge.

[0163] Step S313: Strategies to improve the convergence of power flow that does not converge with active power.

[0164] First, a finite-level BFS topology search program (typically a second or third-level neighborhood search) is run based on the magnitude of the correction to locate the adjacent regions of the bus node. Second, the model parameters of this region are validated, with a focus on checking the impedance parameters of the lines and transformers within that region. If a branch reactance is identified... X If the value far exceeds the physical norm, a standard parameter library will be called for comparison and correction to eliminate active power obstruction caused by model parameter errors. Finally, it will be determined whether there is a line overload or near transmission limit; if so, power flow will be adjusted by increasing the power output at the sending end. Generator units with high sensitivity to bottleneck sections within the region will be identified, and their active power output at the sending end will be increased proportionally to the current output, ensuring that the unit output does not exceed the limit. If adjusting the generator output at the sending end still cannot improve the power flow convergence, load reduction control will be implemented in the receiving end area of ​​the line. Priority will be given to cutting off the non-critical loads that have the most significant impact on the bottleneck section within this area, reducing the power demand at the receiving end.

[0165] After all adjustments are completed, the system will synchronously transmit the corrected generator output, load level, and verified branch parameters back to the initial values ​​for power flow calculation. The system will then restart. P-θ Iteratively, power flow calculations are performed on the adjusted data baseline. If the calculations meet the convergence accuracy requirements... e p Then proceed sequentially to the next step. QUThe process involves iterative steps and reactive power adjustment. If convergence is still not achieved, the number of steps in the first-level BFS search is increased, and the above diagnostic logic is iterated until a set of physically feasible power flow data for the entire network is obtained. Finally, the converged power flow data with improved active power flow convergence is output.

[0166] Step S32, based on QU Iterative reactive power flow adjustment, such as Figure 5 As shown.

[0167] Step S321: After completing the first stage of active power flow convergence and locking the global phase angle distribution, the second stage of phased decoupling to improve power flow convergence is performed. The solver parameters 'Decoupled P Solution Logic' are set to False and 'Decoupled Q Solution Logic' to True. In this stage, the voltage phase angle remains constant during the iteration process; only the voltage amplitude is iteratively solved for the reactive power balance equation. The calculation steps are as follows:

[0168] a) Substitute into the first stage P-θ The system power flow parameter set obtained after iterative convergence, keeping the voltage phase angle constant during the iteration process, is set to the number of iterations. m = 0.

[0169] b) Substitute into Equation 23 to calculate the reactive power imbalance at each node. and find .

[0170] c) Solve Equation 21 to obtain the voltage amplitude correction for each node. .

[0171] d) Correct the magnitude of the voltage at each node. :

[0172] (27)

[0173] e) Determine the number of iterations m Has the maximum number of iterations been reached? If so, proceed to step g; otherwise, proceed to step f.

[0174] f) Determine the reactive power imbalance at each node. Does it meet the preset convergence accuracy? e q If the condition is met, proceed to step g; otherwise, set the iteration count to zero. m = m +1, and return to step b for iteration.

[0175] g) Output the final power flow data set

[0176] Step S322: Locating the power flow problem of non-converging reactive power.

[0177] when QU Iterations within a preset maximum number of iterations Within, its maximum reactive power imbalance Δ Q Still unable to meet the preset convergence accuracy This indicates a severe imbalance between the current reactive power resource allocation and the voltage control target, causing the computational model to be unable to obtain a feasible solution under the current physical constraints.

[0178] When the reactive power flow calculation fails to converge, apply a correction based on the maximum voltage amplitude. Problem localization and regional adjustment strategies. Reactive power balance is highly coupled with node voltage amplitude. During the iteration process, reactive power imbalance... It only reflects the reactive power difference at the node, while the voltage amplitude correction is not included. It is the response of the reactive power difference after being mapped by the Jacobian matrix, reflecting the voltage adjustment range required by the system to eliminate the reactive power difference. The larger the value, the higher the reactive power regulation capability required for the node to maintain the target voltage, and the closer it is to its reactive power regulation limit. Compared to , This approach more fundamentally reflects the root cause of the system's reactive voltage non-convergence. A breadth-first search is then performed starting from this bus to delineate the critical regions associated with it.

[0179] In each QU During iterative calculations, the system records the reactive power imbalance of all nodes in the network in real time. With voltage amplitude correction Simultaneously extract the maximum voltage correction amount within the current iteration cycle. and its corresponding bus nodes. Focus on observing the period approaching the maximum iteration count. Bus node. If in multiple consecutive iterations, If a node is always located on a specific bus, it indicates that the voltage balance capability at that node has reached its physical limit, and the bus can be identified as a weak point in the power grid that causes reactive power flow divergence.

[0180] Step S323: Strategy to improve the convergence of power flow when reactive power fails to converge.

[0181] When reactive power flow fails to converge, a finite-level BFS topology search program is first run based on the magnitude of the correction, typically searching a second or third-level neighborhood. This bus voltage correction is then used to determine the final bus voltage level. The sign of the reactive power is used to determine whether the busbar is in a state of insufficient or excessive reactive power. For the type of insufficient reactive power, first increase the reactive power output of the Static Var Compensator (SVC) and Static Var Generator (SVG) with the fastest response speed in the area. If the reactive power deficit exceeds its dynamic response capability, further adjust the discrete reactive power equipment, connect parallel capacitors, and disconnect the parallel reactors in the area. If it is still not balanced, further disconnect the inductive load. For the type of excessive reactive power, first adjust the Static Var Compensator (SVC) and Static Var Generator (SVG) with the fastest response speed to reduce their reactive power output. If the reactive power deficit exceeds its dynamic response capability, connect parallel reactors and disconnect the parallel capacitors in the area. If the reactive power non-convergence is caused by the reciprocating jump of the DC converter transformer tap changer during the iteration process, the tap changer step size will be appropriately increased to avoid the algorithm triggering continuous step adjustment of discrete parameters under small reactive power fluctuations, thereby cutting off the numerical oscillation source and stabilizing the reactive power in the area.

[0182] After the reactive power flow convergence improvement is completed, the system will synchronously transmit the corrected grid parameters back to the system. QU The initial values ​​for the iterative calculation. The system restarts. QU Iteratively, power flow calculations are performed on the adjusted data baseline. If the calculations meet the convergence accuracy requirements... If convergence is still not achieved, the subsequent fully coupled power flow calculation will proceed sequentially. If convergence is still not achieved, the number of steps in the first-level BFS search will be increased and the above diagnostic logic will be iterated until a set of physically feasible power flow data for the entire network is obtained, and finally, the complete power flow data after the reactive power flow convergence is improved will be output.

[0183] Step S4: Based on the obtained power flow parameters as initial values, perform fully coupled power flow calculations to obtain the final power flow solution that meets the convergence accuracy requirements.

[0184] After completion P-θ and QU After two phases of decoupling and adjustment, the system has obtained high-precision power flow parameters. While this decoupling is efficient, it deliberately ignores... PU and Q-θ The cross-coupling between them means that the result can only be considered as a high-precision approximate solution.

[0185] Therefore, a fully coupled power flow verification is introduced in the final stage. Since the preceding decoupling adjustments have pushed the system power flow parameters into the minimum neighborhood of the final solution, forming extremely high-quality initial values ​​for iteration, the verification procedure can flexibly use the Newton-Raphson method (NR method) or the PQ decomposition method for global closed-loop solution. If the PQ decomposition method is used, both the 'Decoupled Power Logic' and 'Decoupled Q Solution Logic' parameters are set to True.

[0186] The calculation steps are as follows:

[0187] 1) Load the system power flow parameter set obtained after phased decoupling power flow calculation, and set the iteration number. k =0.

[0188] 2) Calculate the first according to equation 22-23. k Active power imbalance at each node in the next iteration and reactive power imbalance .

[0189] 3) If the NR method is used, the complete Jacobian matrix is ​​constructed according to Equations 11-18 to solve Equation 10 and obtain the phase angle correction for each node. and voltage amplitude correction amount If the PQ decomposition method is used, a coefficient matrix is ​​constructed, and the coefficients of each node are solved using formulas 20-21. as well as .

[0190] 4) Correct the amplitude of the voltage at each node. :

[0191] (28)

[0192] 5) Correct the phase angle of the voltage at each node. :

[0193] (29)

[0194] 6) Determine the active power imbalance at each node. and reactive power imbalance Check if the convergence condition is met. If it is, proceed to step 7); if not, set the iteration count to zero. k = k +1, return to step 2) and iterate.

[0195] 7) Calculate the power of the slack node and the total power of all lines.

[0196] Because the aforementioned phased adjustments provide high-quality approximate initial values, the fully coupled power flow calculation stage requires only a very small number of iterations to quickly and accurately locate a feasible power flow solution that meets the system's operational constraints and convergence accuracy standards. For example... Figure 6 As shown, through the power flow convergence improvement scheme proposed in this invention, non-convergent power flow conditions are gradually guided back to the effective convergence domain.

[0197] Example 2

[0198] This embodiment discloses a power flow convergence system based on DC power flow preprocessing and staged decoupling;

[0199] like Figure 7 As shown, the power flow convergence system based on DC power flow preprocessing and staged decoupling includes:

[0200] The data verification module is configured to perform multi-level verification and parameter adjustment on the raw power flow data, and to perform preliminary verification of the system active power balance to generate a set of initial values ​​for node active power settings that meet basic physical feasibility.

[0201] The DC power flow preprocessing module is configured to: perform DC power flow calculation based on the set of initial values ​​of the active power of the nodes, obtain the initial phase angle of each node, identify the heavily loaded lines whose power angle difference exceeds a preset threshold, adjust the local active power resources to make the power angle difference of all lines meet the constraints, and obtain the initial phase angle distribution.

[0202] The phased decoupled power flow calculation and adjustment module is configured to: combine the preset initial voltage amplitude with the initial phase angle distribution to perform phased decoupled power flow calculation and adjustment to obtain power flow parameters;

[0203] The fully coupled power flow verification module is configured to perform fully coupled power flow calculations based on the obtained power flow parameters as initial values, and obtain the final power flow solution that meets the convergence accuracy requirements.

[0204] Example 3

[0205] The purpose of this embodiment is to provide a computer-readable storage medium.

[0206] A computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps in the power flow convergence method based on DC power flow preprocessing and phased decoupling as described in Embodiment 1.

[0207] Example 4

[0208] The purpose of this embodiment is to provide an electronic device.

[0209] An electronic device includes a memory, a processor, and a program stored in the memory and executable on the processor. When the processor executes the program, it implements the steps in the power flow convergence method based on DC power flow preprocessing and phased decoupling as described in Embodiment 1.

[0210] The steps and methods involved in the apparatuses of Embodiments 2, 3, and 4 above correspond to those in Embodiment 1. For specific implementation details, please refer to the relevant description section of Embodiment 1. The term "computer-readable storage medium" should be understood as a single medium or multiple media including one or more instruction sets; it should also be understood as including any medium capable of storing, encoding, or carrying an instruction set for execution by a processor and enabling the processor to perform any of the methods in this invention.

[0211] Those skilled in the art will understand that the modules or steps of the present invention described above can be implemented using general-purpose computer devices. Optionally, they can be implemented using computer-executable program code, thereby allowing them to be stored in a storage device for execution by a computer device, or they can be fabricated as separate integrated circuit modules, or multiple modules or steps can be fabricated as a single integrated circuit module. The present invention is not limited to any particular combination of hardware and software.

[0212] While the specific embodiments of the present invention have been described above in conjunction with the accompanying drawings, this is not intended to limit the scope of protection of the present invention. Those skilled in the art should understand that various modifications or variations that can be made by those skilled in the art without creative effort based on the technical solutions of the present invention are still within the scope of protection of the present invention.

Claims

1. A power flow convergence method based on DC power flow preprocessing and staged decoupling, characterized in that, include: The raw power flow data is subjected to multi-level verification and parameter adjustment, and the initial verification of the system active power balance is performed to generate a set of initial values ​​for node active power settings. DC power flow calculation is performed based on the set of initial values ​​of active power of the nodes, the initial phase angle of each node is obtained, the heavily loaded lines whose power angle difference exceeds the preset threshold are identified, and the power angle difference of all lines is satisfied by adjusting the distribution of active resources in the local topology area to obtain the initial phase angle distribution. Combining the preset initial voltage amplitude with the initial phase angle distribution, staged decoupled power flow calculation and adjustment are performed to obtain power flow parameters; specifically, the staged decoupled power flow calculation and adjustment includes: Phase 1: Fixed voltage amplitude, only... P-θ The active power flow is solved iteratively. If the iteration fails to converge, the weak bus is located based on the maximum phase angle correction, and topology search and active power resource adjustment are performed on its associated region until... P-θ Iterative convergence; Second stage: Based on the phase angle distribution obtained in the first stage, fix the voltage phase angle and only perform... QU The reactive power flow is solved iteratively. If the iteration fails to converge, the weak bus is located based on the maximum voltage correction, and a topology search and reactive power resource adjustment are performed on its associated region until... QU Iterative convergence; Using the obtained power flow parameters as initial values, a fully coupled power flow calculation is performed to obtain the final power flow solution that meets the convergence accuracy requirements.

2. The power flow convergence method based on DC power flow preprocessing and staged decoupling as described in claim 1, characterized in that, The process involves multi-level verification and parameter adjustment of the original power flow data, and preliminary verification of the system's active power balance, generating a set of initial values ​​for node active power settings, including: The original power flow data is subjected to network topology verification and device parameter verification. The network topology verification includes identifying and removing isolated buses, verifying the consistency of balancing machine configuration and circuit breaker status. The device parameter verification includes verifying the non-negativity and rationality of generator, transformer, and line parameters, as well as the rationality of node power and voltage setting values. The initial total active power imbalance of the system is calculated, and it is determined whether the power to be borne by the balancing machine exceeds its preset upper and lower power limits. If the power of the balancing machine exceeds the limit, the excess power is proportionally distributed to other controllable generators in the system other than the balancing machine, thus obtaining the initial set of node active power settings.

3. The power flow convergence method based on DC power flow preprocessing and staged decoupling as described in claim 1, characterized in that, The process of calculating DC power flow based on the initial set of active power values ​​at the nodes to obtain the initial phase angle of each node includes: Substituting the active power injection power of all nodes in the initial set of active power settings into the DC power flow equation matrix, as shown below: in, This represents the active power injection amount for all unbalanced nodes. The phase angles to be solved for all non-equilibrium nodes; Here is the admittance matrix; Solving this equation yields the initial phase angles of each non-equilibrium node. Setting the phase angles of the equilibrium nodes to 0 gives the initial phase angles of all nodes in the network.

4. The power flow convergence method based on DC power flow preprocessing and staged decoupling as described in claim 1, characterized in that, The identification of overloaded lines with a power angle difference exceeding a preset threshold includes: Based on the DC power flow calculation results, calculate the absolute value of the power angle difference between the two ends of all lines; The absolute value of the power angle difference is compared with a preset static stability safety threshold to identify lines that exceed the limit; Using the sending-end bus and receiving-end bus of each over-limit line as the center, a local topology region is delineated using a breadth-first search algorithm; According to the preset priority order, the distribution of active resources in the local topology area is adjusted, and the active power output of the opposite area is adjusted synchronously to maintain the overall active power balance of the system. Recalculate the DC power flow until the absolute value of the power angle difference of all lines is less than the static stability safety threshold.

5. The power flow convergence method based on DC power flow preprocessing and staged decoupling as described in claim 1, characterized in that, The method involves locating weak buses based on the maximum phase angle correction, and then performing topology searches and active power resource adjustments in their associated regions until... P-θ Iterative convergence includes: when P-θ If the iteration reaches the preset maximum number of iterations and still fails to converge, extract the maximum phase angle correction amount within the iteration cycle that caused the non-convergence and its corresponding bus node. Starting from the aforementioned busbar, a breadth-first search with a finite number of steps is used to determine the key adjustment area; Perform at least one of the following operations on the critical adjustment area: verify and correct line and transformer parameters, adjust the active power output of generators in the area according to the principle of sensitivity or proportionality, and reduce non-critical loads in the area.

6. The power flow convergence method based on DC power flow preprocessing and staged decoupling as described in claim 1, characterized in that, The method involves locating weak buses based on the maximum voltage correction, and then performing topology searches and reactive power resource adjustments in their associated areas until... QU Iterative convergence includes: when QU If the iteration fails to converge after reaching the preset maximum number of iterations, extract the maximum voltage correction amount within the iteration period that caused the non-convergence and its corresponding bus node. Starting from the aforementioned busbar, a breadth-first search with a finite number of steps is used to determine the key adjustment area; The reactive power status of the region is determined based on the sign of the maximum voltage correction. If the maximum voltage correction is positive, capacitive reactive power equipment is put into operation or inductive reactive power equipment is cut off according to priority. If the maximum voltage correction is negative, inductive reactive power equipment is put into operation or capacitive reactive power equipment is cut off according to priority.

7. A power flow convergence system based on DC power flow preprocessing and staged decoupling, characterized in that, include: The data verification module is configured to perform multi-level verification and parameter adjustment on the raw power flow data, perform preliminary verification of the system active power balance, and generate a set of initial values ​​for node active power settings. The DC power flow preprocessing module is configured to: perform DC power flow calculation based on the set of initial values ​​of the active power of the nodes, obtain the initial phase angle of each node, identify the heavily loaded lines whose power angle difference exceeds a preset threshold, adjust the distribution of active resources in the local topology area to make the power angle difference of all lines meet the constraints, and obtain the initial phase angle distribution. The phased decoupling power flow calculation and adjustment module is configured to: combine a preset initial voltage amplitude with the initial phase angle distribution to perform phased decoupling power flow calculation and adjustment to obtain power flow parameters; specifically, the phased decoupling power flow calculation and adjustment includes: Phase 1: Fixed voltage amplitude, only... P-θ The active power flow is solved iteratively. If the iteration fails to converge, the weak bus is located based on the maximum phase angle correction, and topology search and active power resource adjustment are performed on its associated region until... P-θ Iterative convergence; Second stage: Based on the phase angle distribution obtained in the first stage, fix the voltage phase angle and only perform... QU The reactive power flow is solved iteratively. If the iteration fails to converge, the weak bus is located based on the maximum voltage correction, and a topology search and reactive power resource adjustment are performed on its associated region until... QU Iterative convergence; The fully coupled power flow verification module is configured to perform fully coupled power flow calculations based on the obtained power flow parameters as initial values, and obtain the final power flow solution that meets the convergence accuracy requirements.

8. A computer-readable storage medium having a program stored thereon, characterized in that, When the program is executed by the processor, it implements the steps in the power flow convergence method based on DC power flow preprocessing and phased decoupling as described in any one of claims 1-6.

9. An electronic device comprising a memory, a processor, and a program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the steps in the power flow convergence method based on DC power flow preprocessing and phased decoupling as described in any one of claims 1-6.

Images