A method for decoupling branches in electromagnetic transient simulation

Abstract

The application belongs to the technical field of electromagnetic transient simulation, and particularly relates to a kind of electromagnetic transient simulation branch decoupling methods, including LCL decoupling branch based on the boundary node voltage obtained, and the intermediate capacitor voltage across the intermediate capacitor C calculated in the state update formula under the state update formula of the system stored and using forward branch travel strategy;The application adopts pure implicit integration method to discretize the energy storage element of LCL decoupling branch, and constructs state quantity update formula combined with forward branch propulsion strategy, so that the algorithm can maintain numerical stability under any simulation step Δt condition, and completely solves the problem that traditional delay insertion method and other schemes only have conditional stability and step length is limited by the minimum inductance value of the circuit.

Description

Technical Field

[0001] This invention belongs to the field of electromagnetic transient simulation technology, specifically relating to a method for decoupling electromagnetic transient simulation branches. Background Technology

[0002] Electromagnetic transient simulation can comprehensively and accurately depict the high-frequency dynamic characteristics of power systems and has become a key means to master the operating characteristics of new power systems. However, with the continuous advancement of multi-regional AC / DC grid integration and high-proportion renewable energy grid connection, the scale of power systems is constantly expanding. In order to accurately simulate the high-frequency dynamic characteristics brought about by a large number of power electronic devices, the simulation step size required for power system simulation is gradually decreasing. The larger simulation scale and smaller simulation step size bring higher computing power requirements to electromagnetic transient simulation. The electromagnetic transient simulation technology under the traditional serial computing mode cannot cope with the simulation scenarios of large-scale renewable energy grid connection and complex AC / DC coupling in terms of simulation efficiency.

[0003] Decoupled parallel simulation technology, as an important means to improve computational efficiency, is being integrated with electromagnetic transient simulation methods and has become a key focus of research in new power system simulation technologies. Decoupled parallel simulation technology effectively improves simulation efficiency and computational speed by decomposing the power system model into multiple independent or dependent sub-models and then processing them in parallel on multiple computational units. Branch decoupling, as the key to decomposing the overall power system model into multiple sub-models, plays a crucial role in parallel simulation.

[0004] The natural decoupling method for long transmission lines is currently the most widely used decoupling method. When there are long-distance transmission lines in a power system that can be simulated using a distributed parameter line model, the power system can be divided into sub-networks based on the natural decoupling characteristics of the distributed parameter line model. The advantage of this method is that the communication volume between sub-networks is small, and it does not incur additional computational burden due to parallel network division. However, this parallel network division computation method requires that the network contain long-distance transmission lines with a transmission time delay greater than the simulation step size, which limits the application scenarios of this method and lacks sufficient flexibility. The connection variables of the node splitting method, branch cutting method, MATE method, and SSN algorithm must all be solved serially, and the computational complexity of the connection variables of these algorithms increases exponentially with the number of sub-networks. The rapid increase in the rate of computation has led to insufficient simulation efficiency when performing large-scale power system simulation analysis. While explicit integration methods such as forward Euler integration can be used to discretize inductors and capacitors, enabling one-step time-delay parallel methods for network partitioning, numerical oscillations are easily triggered when the system experiences switching actions or sudden changes in non-state variables. Therefore, implicit integration methods are often required to ensure stability, thus reducing or even eliminating the advantages of parallel computing. A discretization method combining explicit and implicit integration methods typically requires iterative calculations to ensure numerical accuracy and stability, but the iteration process significantly increases the computational load, weakening the advantages of network partitioning parallel methods in improving simulation efficiency. While ideal transformer methods, damped impedance methods, and partial component reuse methods can achieve acceptable simulation accuracy, their numerical stability is often difficult to guarantee. In practical applications, additional stability checks or compensation by increasing simulation delays are required, increasing the complexity of the methods. Summary of the Invention

[0005] To address the aforementioned shortcomings of existing technologies, this invention provides an electromagnetic transient simulation branch decoupling method, applicable to systems comprising at least one subsystem. Subsystem The power system sub-network model includes the LCL decoupling branch connecting the two, where the LCL decoupling branch includes a first inductor L1, a second inductor L2, and an intermediate capacitor C. The method includes the following steps: S1. At the end of the nth simulation step, obtain the subsystem... boundary node voltage and subsystems boundary node voltage The acquired boundary node voltage is then transmitted to the LCL decoupling branch. S2, the LCL decoupling branch is based on the boundary node voltage obtained in step S1. With the system stored The first inductor current is calculated using the state update formula under the forward branching strategy. ; S3, the LCL decoupling branch is based on the boundary node voltage obtained in step S1. The first inductor current in step S2 With the system stored The second inductor current is calculated using the state update formula under the forward branching strategy. ; S4, the LCL decoupling branch is based on the first inductor current in step S2. The second inductor current in step S3 With the system stored The intermediate capacitor voltage is calculated using the state update formula under the forward branching strategy. ; S5, adjust the voltage of the intermediate capacitor. Feedback is sent to the subsystems respectively. and subsystem For the subsystem and subsystem The respective network equations are solved in parallel and all state variables inside the subsystem are updated to complete the simulation of the (n+1)th simulation step. S6. Determine whether the simulation end time has been reached. If not, proceed to the next simulation time and return to step S1. If the time has been reached, end the simulation.

[0006] Further improvements to this technical solution include step S1, which includes: S11. At the end of the nth simulation step, the control subsystem With subsystem The system performs the update calculation of all internal state variables in parallel to obtain the system simulation result of the nth simulation step, where the state variables include the voltage of each node and the current flowing through each branch. S12. Based on the system simulation results obtained in step S11 for the nth simulation step, determine the subsystem. With subsystem Electrical quantities at the connection nodes between them, specifically electrical quantities of the subsystem Output boundary node voltage and subsystems Output boundary node voltage , where n represents the simulation time to which the electrical quantity belongs; S13. The boundary node voltage determined in step S12 is... and As a known input signal, it is transmitted from the subsystem via a data communication interface or shared memory. With subsystem The solution unit is transferred to the independent solution unit responsible for calculating the LCL decoupling branch.

[0007] Further improvements to this technical solution include the following: In step S13, transmitting the boundary node voltage to the LCL decoupling branch specifically includes: S131, Set the boundary node voltage and Each voltage source is considered as an independent voltage source applied to both ends of the LCL decoupling branch. In the calculation model of the LCL decoupling branch, a system is established based on... The first controlled voltage source model for the output voltage and the model with The second controlled voltage source model for the output voltage; S132. In the equivalent circuit of the LCL decoupling branch, the positive terminal of the first controlled voltage source model is connected to the end of the first inductor L1 furthest from the intermediate capacitor C, and its negative terminal is connected to the subsystem. The reference ground potential is connected, and the positive terminal of the second controlled voltage source model is connected to the end of the second inductor L2 away from the intermediate capacitor C, and its negative terminal is connected to the reference ground potential of the subsystem β.

[0008] Further improvements to this technical solution include step S2, which includes: S21. In the independent solution unit of the LCL decoupling branch, the first inductor L1 is discretized using a pure implicit integration method to obtain the discretized first inductor current of the first inductor L1. : ; in, The first inductor current stored at the previous moment; The second inductor current stored in the previous moment; L1 is the voltage of the intermediate capacitor stored at the previous moment; C is the inductance value of the first inductor; Δt is the capacitance value of the intermediate capacitor; and Δt is the simulation step size. The resistance value is the value of the resistor connected in series with the first inductor L1; Let L1 be the inductance value of the first inductor. This is the conductance value connected in parallel with the intermediate capacitor C; S22, Discretize the first inductor current The forward branching strategy is used to update the calculation of the first inductor current. : ; in, M is the first equivalent admittance parameter associated with the first inductor branch, and M is the second equivalent admittance parameter associated with the capacitor branch. S23. Calculate the first inductor current obtained in step S22. The numerical values ​​are stored and output as known quantities.

[0009] Further improvements to this technical solution include the following: In step S22, the calculation is performed using a preset update formula, specifically including: S221. Calculate the first equivalent admittance parameter based on the circuit parameters of the LCL decoupling branch and the simulation step size Δt. The second equivalent admittance parameter M, where , ; S222. Read the known input quantities from the storage unit, including the first inductor current at time n. Second inductor current Intermediate capacitor voltage and the boundary node voltage from step S1 ; S223. The parameters calculated in step S221 are... M and the input values ​​read in step S222 Along with the fixed circuit parameters L1, C, and simulation step size Δt, these are substituted into the update formula for the first inductor current to obtain the first inductor current at time n+1. Numerical solution.

[0010] Further improvements to this technical solution include step S3, which includes: S31. In the independent solution unit of the LCL decoupling branch, the second inductor L2 is discretized using a pure implicit integration method to obtain the discretized second inductor current of the second inductor L2. : ; Where L2 is the inductance value of the second inductor; The resistance value is the value of the resistor connected in series with the second inductor L2; Let L2 be the inductance value of the second inductor. S32, Discretized second inductor current The forward branching strategy is used to update the calculation of the second inductor current. : ; in, This is the second equivalent admittance parameter associated with the second inductor branch; S33. The second inductor current calculated in step S32 is... The numerical values ​​are stored and output as known quantities.

[0011] Further improvements to this technical solution include the following: In step S32, the calculation is performed using a preset update formula, specifically including: S321. Based on the circuit parameters of the LCL decoupling branch and the simulation step size Δt, calculate the first equivalent admittance parameter respectively. The second equivalent admittance parameter M and the third equivalent admittance parameter Wherein, parameter M characterizes the equivalent discrete admittance of the intermediate capacitor C under the rules of the pure implicit integration method; , The resistance value is the value of the second inductor L2 connected in series; S322. Read all known input quantities from the storage unit, including the historical state quantity of the first inductor current at the nth simulation step. Second inductor current and intermediate capacitor voltage and the boundary node voltage obtained from step S1 and ; S323, The parameters calculated in step S321 are... The input quantity read in step S322 Along with fixed circuit parameters , , Substituting the simulation step size Δt into the second inductor current update formula, we obtain the second inductor current at the (n+1)th simulation step size. Numerical solution.

[0012] Further improvements to this technical solution include step S4, which includes: S41. In the independent solution unit of the LCL decoupling branch, the intermediate capacitor C is discretized using a pure implicit integration method to obtain the discretized intermediate capacitor voltage of the intermediate capacitor C. : ; S42, Discretized intermediate capacitor voltage The intermediate capacitor voltage is updated using a forward branching strategy. : ; S43. The intermediate capacitor voltage calculated in step S42 is... The numerical values ​​are stored and output as known quantities.

[0013] Further improvements to this technical solution include step S5, which includes: S51. After completing the calculation in step S4 in the independent solution unit of the LCL decoupling branch, calculate the intermediate capacitor voltage at the (n+1)th simulation step. The numerical values ​​are transmitted to the solution subsystem via data communication interface or shared memory. The computational unit and the computational unit responsible for solving the subsystem β; S52, in the subsystem In the calculation model, the received intermediate capacitor voltage Introducing it into its network equations as an independent voltage source excitation, specifically in the subsystem At the original connection node of the decoupling branch from LCL, establish an output voltage that is constant. A controlled voltage source to replace the original physical branch connection; S53. In the calculation model of subsystem β, perform the same operation as in step S52 simultaneously, and change the intermediate capacitor voltage. As an output voltage that is constant The controlled voltage source is introduced into its network equations, making the subsystem With subsystem β, given definite boundary conditions Then, each system independently and in parallel solves its own internal network algebraic equations and updates the state variables of all node voltages and branch currents, together completing the full system simulation of the (n+1)th simulation step.

[0014] A further improvement to this technical solution is that, in step S52, the intermediate capacitor voltage is introduced into the subsystem as a voltage source. The network equations specifically include: S521, in the subsystem In the equivalent circuit model, locate the boundary node that was originally connected to the first inductor L1 in the LCL decoupling branch, and mark this node as an internal node. And establish a new branch road between this node and the reference point; S522. Define the newly added branch as an ideal controlled voltage source, whose terminal voltage The value of t=(n+1)Δt at the simulation time is determined by the received intermediate capacitor voltage value, i.e., it is set as follows: The model parameters of the voltage source are then correlated with the dynamically updated values ​​of the intermediate capacitor voltage. S523, Subsystem modified based on steps S521 and S522 The circuit topology is reconstructed to form its nodal admittance matrix. and the right-hand current source vector Among them, the controlled voltage source The known potential difference is treated as a right-hand term in the network equation, and then the system of linear equations is solved. Obtain the voltage vectors of all internal nodes of the subsystem at time n+1. ,in It includes the voltage values ​​of all nodes except the reference ground.

[0015] The beneficial effects of this invention are as follows: This invention employs a pure implicit integration method to discretize the energy storage elements of the LCL decoupled branch throughout the process. Combined with a forward branching strategy, it constructs a state update formula, enabling the algorithm to maintain numerical stability under any simulation step size Δt. This completely solves the problems of traditional delayed insertion methods and other schemes that only have conditional stability and whose step size is limited by the minimum capacitance value of the circuit. It does not require sacrificing simulation step size and efficiency to ensure stability.

[0016] On the one hand, by using a forward branching strategy, the implicitly coupled state update relationships in the LCL decoupling branches are transformed into a sequential computation process. This eliminates the need to establish and solve complex linear equation systems, as well as iterative computation, thus removing the additional computational overhead caused by matrix solving and iteration. On the other hand, the subsystem... β can solve its own network equations independently and in parallel based on the boundary conditions of LCL decoupled branch feedback, avoiding the problem of exponential growth in computational complexity caused by serial solving of connection variables in schemes such as node splitting method, and significantly improving the simulation speed of large-scale AC / DC hybrid and high-proportion renewable energy grid-connected power systems.

[0017] This invention is based on a general LCL decoupling branch structure design, which does not depend on the existence of specific topologies such as long transmission lines in the power system. It breaks through the application scenario limitations of the natural decoupling method for long transmission lines and can be flexibly adapted to the sub-network simulation needs of various power systems such as new power systems with a large number of power electronic devices and multi-regional AC / DC power grids, and has good engineering versatility.

[0018] This invention, through pure implicit integration and ordered state update logic, accurately characterizes the electrical characteristics and state change patterns of the LCL decoupled branch while ensuring unconditional stability. Unlike the ideal transformer method or damped impedance method, it does not require additional stability verification or simulation delay compensation, thus simplifying the simulation operation process, reducing the complexity of the method, and ensuring the accuracy of the simulation results.

[0019] This invention transforms the subsystem boundary node voltage into a controlled voltage source connected to the LCL decoupling branch, and simultaneously feeds back the intermediate capacitor voltage of the decoupling branch as a controlled voltage source to the subsystem. This constructs a clear electrical connection and signal transmission logic for the sub-network model. Furthermore, the calculation model, parameter definition, and update formula of each step all have clear physical meaning and mathematical basis. The design of the solution unit and the data interaction method can be directly combined with the solution framework of existing electromagnetic transient simulation programs, making it easy to implement in engineering and integrate into the system.

[0020] This invention updates the first inductor current, the second inductor current, and the intermediate capacitor voltage of the LCL decoupling branch sequentially. Each calculation is based on previously obtained known quantities. At the same time, orderly data transmission between the subsystem and the decoupling branch is achieved through a data communication interface or shared memory, avoiding numerical oscillations caused by sudden changes in non-state variables, ensuring the continuity and reliability of the entire simulation process, and improving the accuracy of the simulation results. Attached Figure Description

[0021] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0022] Figure 1 This is a schematic flowchart illustrating a method according to an embodiment of the present invention.

[0023] Figure 2 This is a diagram of the LCL branch topology.

[0024] Figure 3 This is the topology diagram after decoupling the LCL branch. Detailed Implementation

[0025] To make the objectives, features, and advantages of this invention more apparent and understandable, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings of the specific embodiments. Obviously, the embodiments described below are only some embodiments of this invention, and not all embodiments. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0026] Unless otherwise defined, all technical and scientific terms used in this invention have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. The terminology used in this specification is for the purpose of describing particular embodiments only and is not intended to limit the invention.

[0027] Figure 1 This is a schematic flowchart illustrating an electromagnetic transient simulation branch decoupling method provided by the present invention. The method is applied to systems comprising at least one subsystem. Subsystem The flowchart describes a power system sub-network model that includes the LCL decoupling branch connecting the two inductors. The LCL decoupling branch consists of a first inductor L1, a second inductor L2, and an intermediate capacitor C. Depending on the requirements, the order of steps in this flowchart can be changed, and some steps can be omitted.

[0028] like Figure 1 As shown, the method includes: S1. At the end of the nth simulation step, obtain the subsystem... boundary node voltage and subsystems boundary node voltage The acquired boundary node voltage is then transmitted to the LCL decoupling branch. S2, LCL decoupling branch based on boundary node voltage Calculate the current flowing through the first inductor L1 at the (n+1)th simulation step. ; S3. Based on the boundary node voltage obtained in step S1 With the system stored The second inductor current is calculated using the state update formula under the forward branching strategy. S4. Based on the boundary node voltage obtained in step S1 and stored in the system The intermediate capacitor voltage is calculated using the state update formula under the forward branching strategy. ; S5, adjust the voltage of the intermediate capacitor. Feedback is sent to the subsystems respectively. and subsystem For the subsystem and subsystem The respective network equations are solved in parallel and all state variables inside the subsystem are updated to complete the simulation of the (n+1)th simulation step. S6. Determine whether the simulation end time has been reached. If not, proceed to the next simulation time and return to step S1. If the time has been reached, end the simulation.

[0029] To facilitate understanding of the present invention, the following description further illustrates the electromagnetic transient simulation branch decoupling method provided by the present invention, based on the principle of the electromagnetic transient simulation branch decoupling method and the process of decoupling the electromagnetic transient simulation branch in the embodiments.

[0030] Specifically, the decoupling of the LCL branch: In electromagnetic transient simulation, commonly used decoupled network branch structures include... Figure 2 As shown. This type of network decoupling method mainly utilizes the relatively slow change of state variables of energy storage elements (inductors L or intermediate capacitors C) in the network to achieve delayed information transmission between subsystems, thereby achieving network decoupling. Therefore, each branch of the network must contain at least one energy storage element (L or C) to utilize its state variables for time-delay decoupling. In contrast, resistors R and conductance G only describe the algebraic relationship between voltage and current and do not possess energy storage characteristics; therefore, they are not necessary components for achieving network decoupling. After the above decoupling process, the original system can be divided into three parts: subsystems Subsystem β and the LCL decoupling branch located between them.

[0031] from Figure 3 As can be seen from the decoupling structure shown, the LCL decoupling branch in the middle acts as a connecting subsystem. The interface between the branch and subsystem β plays a crucial role in transmitting voltage and current information between subsystems. Therefore, the numerical stability of this branch directly affects the stability and accuracy of the entire network simulation. To ensure the stable operation of the decoupled simulation, an unconditional stability solution method suitable for LCL decoupled branches is proposed, which can maintain the numerical stability of the system under arbitrary simulation step sizes, thereby improving the reliability and computational efficiency of the network electromagnetic transient simulation.

[0032] First, step S1 includes: S11. At the end of the nth simulation step, the control subsystem With subsystem The system performs the update calculation of all internal state variables in parallel to obtain the system simulation result of the nth simulation step, where the state variables include the voltage of each node and the current flowing through each branch. S12. Based on the system simulation results obtained in step S11 for the nth simulation step, determine the subsystem. With subsystem Electrical quantities at the connection nodes between them, specifically electrical quantities of the subsystem Output boundary node voltage and subsystems Output boundary node voltage , where n represents the simulation time to which the electrical quantity belongs; S13. The boundary node voltage determined in step S12 is... and As a known input signal, it is transmitted from the subsystem via a data communication interface or shared memory. With subsystem The solution unit is transferred to the independent solution unit responsible for calculating the LCL decoupling branch.

[0033] Furthermore, step S13, which involves transmitting the boundary node voltage to the LCL decoupling branch, specifically includes: S131, Set the boundary node voltage and Each voltage source is considered as an independent voltage source applied to both ends of the LCL decoupling branch. In the calculation model of the LCL decoupling branch, a system is established based on... The first controlled voltage source model for the output voltage and the model with The second controlled voltage source model for the output voltage; S132. In the equivalent circuit of the LCL decoupling branch, the positive terminal of the first controlled voltage source model is connected to the end of the first inductor L1 furthest from the intermediate capacitor C, and its negative terminal is connected to the subsystem. The reference ground potential is connected, and the positive terminal of the second controlled voltage source model is connected to the end of the second inductor L2 away from the intermediate capacitor C, and its negative terminal is connected to the reference ground potential of the subsystem β. S133. In the solution unit of the LCL decoupling branch, load the branch topology and parameters defined in steps S131 and S132, which include the first controlled voltage source model and the second controlled voltage source model, to calculate the first inductor current in step S2. Provide voltage boundary conditions .

[0034] Secondly, step S2 includes: S21. In the independent solution unit of the LCL decoupling branch, the first inductor L1 is discretized using a pure implicit integration method to obtain the discretized first inductor current of the first inductor L1. : ; in, The first inductor current stored at the previous moment; The second inductor current stored in the previous moment; L1 is the voltage of the intermediate capacitor stored at the previous moment; C is the inductance value of the first inductor; Δt is the capacitance value of the intermediate capacitor; and Δt is the simulation step size. The resistance value is the value of the resistor connected in series with the first inductor L1; Let L1 be the inductance value of the first inductor. This is the conductance value connected in parallel with the intermediate capacitor C; S22, Discretize the first inductor current The forward branching strategy is used to update the calculation of the first inductor current. : ; in, M is the first equivalent admittance parameter associated with the first inductor branch, and M is the second equivalent admittance parameter associated with the capacitor branch. S23. Calculate the first inductor current obtained in step S22. The numerical values ​​are stored and output as known quantities.

[0035] Furthermore, the calculation in step S22 using the preset update formula specifically includes: S221. Calculate the first equivalent admittance parameter based on the circuit parameters of the LCL decoupling branch and the simulation step size Δt. The second equivalent admittance parameter M, where , ; S222. Read the known input quantities from the storage unit, including the first inductor current at time n. Second inductor current Intermediate capacitor voltage and the boundary node voltage from step S1 ; S223. The parameters calculated in step S221 are... M and the input values ​​read in step S222 Along with the fixed circuit parameters L1, C, and simulation step size Δt, these are substituted into the update formula for the first inductor current to obtain the first inductor current at time n+1. Numerical solution.

[0036] In addition, step S3 includes: S31. In the independent solution unit of the LCL decoupling branch, the second inductor L2 is discretized using a pure implicit integration method to obtain the discretized second inductor current of the second inductor L2. : ; Where L2 is the inductance value of the second inductor; The resistance value is the value of the resistor connected in series with the second inductor L2; Let L2 be the inductance value of the second inductor. S32, Discretized second inductor current The forward branching strategy is used to update the calculation of the second inductor current. : ; in, This is the second equivalent admittance parameter associated with the second inductor branch; S33. The second inductor current calculated in step S32 is... The numerical values ​​are stored and output as known quantities.

[0037] Furthermore, the calculation in step S32 using the preset update formula specifically includes: S321. Based on the circuit parameters of the LCL decoupling branch and the simulation step size Δt, calculate the first equivalent admittance parameter respectively. The second equivalent admittance parameter M and the third equivalent admittance parameter Wherein, parameter M characterizes the equivalent discrete admittance of the intermediate capacitor C under the rules of the pure implicit integration method; , The resistance value is the value of the second inductor L2 connected in series; S322. Read all known input quantities from the storage unit, including the historical state quantity of the first inductor current at the nth simulation step. Second inductor current and intermediate capacitor voltage and the boundary node voltage obtained from step S1 and ; S323, The parameters calculated in step S321 are... The input quantity read in step S322 Along with fixed circuit parameters , , Substituting the simulation step size Δt into the second inductor current update formula, we obtain the second inductor current at the (n+1)th simulation step size. Numerical solution.

[0038] Next, step S4 includes: S41. In the independent solution unit of the LCL decoupling branch, the intermediate capacitor C is discretized using a pure implicit integration method to obtain the discretized intermediate capacitor voltage of the intermediate capacitor C. : ; S42, Discretized intermediate capacitor voltage The intermediate capacitor voltage is updated using a forward branching strategy. : ; S43. The intermediate capacitor voltage calculated in step S42 is... The numerical values ​​are stored and output as known quantities.

[0039] Finally, step S5 includes: S51. After completing the calculation in step S4 in the independent solution unit of the LCL decoupling branch, calculate the intermediate capacitor voltage at the (n+1)th simulation step. The numerical values ​​are transmitted to the solution subsystem via data communication interface or shared memory. The computational unit and the computational unit responsible for solving the subsystem β; S52, in the subsystem In the calculation model, the received intermediate capacitor voltage Introducing it into its network equations as an independent voltage source excitation, specifically in the subsystem At the original connection node of the decoupling branch from LCL, establish an output voltage that is constant. A controlled voltage source to replace the original physical branch connection; S53. In the calculation model of subsystem β, perform the same operation as in step S52 simultaneously, and change the intermediate capacitor voltage. As an output voltage that is constant The controlled voltage source is introduced into its network equations, making the subsystem With subsystem β, given definite boundary conditions Then, each system independently and in parallel solves its own internal network algebraic equations and updates the state variables of all node voltages and branch currents, together completing the full system simulation of the (n+1)th simulation step.

[0040] Furthermore, in step S52, the intermediate capacitor voltage is introduced into the subsystem as a voltage source. The network equations specifically include: S521, in the subsystem In the equivalent circuit model, locate the boundary node that was originally connected to the first inductor L1 in the LCL decoupling branch, and mark this node as an internal node. And establish a new branch road between this node and the reference point; S522. Define the newly added branch as an ideal controlled voltage source, whose terminal voltage The value of t=(n+1)Δt at the simulation time is determined by the received intermediate capacitor voltage value, i.e., it is set as follows: The model parameters of the voltage source are then correlated with the dynamically updated values ​​of the intermediate capacitor voltage. S523, Subsystem modified based on steps S521 and S522 The circuit topology is reconstructed to form its nodal admittance matrix. and the right-hand current source vector Among them, the controlled voltage source The known potential difference is treated as a right-hand term in the network equation, and then the system of linear equations is solved. Obtain the voltage vectors of all internal nodes of the subsystem at time n+1. ,in It includes the voltage values ​​of all nodes except the reference ground.

[0041] Specifically, the state update equations are obtained as follows: Using a purely implicit integration method, the updated equations for the inductor current and capacitor voltage in the branch containing the energy storage element can be obtained: (1) (2) voltage Substituting the update equation into the inductor current update equation (2), we get: (3) Rearranging equation (3), we get: (4) Because the state update formula obtained after discretization of pure implicit integration has implicit coupling relationships between state variables, it is difficult to solve directly. Therefore, it is necessary to solve linear equations or perform iterative calculations. To avoid the computational overhead of matrix solving and iteration, this technical solution introduces a forward branching strategy to sequentially process the state update process, enabling each state variable to be updated sequentially according to the branch direction. This significantly improves computational efficiency while ensuring unconditional stability.

[0042] By adopting a forward branching strategy and applying the above update relationship to the LCL decoupling branch, we can obtain: (5) make , , Then, the above equation can be rearranged to obtain: (6) (7) Substituting formula (5) into formula (7) yields: (8) (9) Substituting formulas (6) and (8) into (9), we get: (10) Formulas (6), (8) and (10) together constitute the state variable update equation for the LCL decoupled branch.

[0043] By employing a forward branching strategy, state variables can be updated sequentially according to the physical connection order of the branches, transforming the originally coupled implicit update relationship into a sequential computation process. In this computation, each update step is based on the latest obtained state variables, thus avoiding the problems of establishing and solving linear equation systems or performing multiple iterations required in traditional implicit solution methods. Simultaneously, since this method still establishes the state update relationship based on a pure implicit integral scheme, it maintains the good numerical stability inherent in implicit methods, enabling the decoupled branches to operate stably under different simulation step sizes, thereby ensuring the stability and reliability of the sub-network electromagnetic transient simulation.

[0044] The specific simulation process after decoupling is as follows: Taking a complete simulation time step [nΔT, (n+1)ΔT] as an example, the simulation process of the decoupled system includes 5 steps.

[0045] S1: At time nΔT, the system has completed the simulation for the nth time step, and all state variables in the network at time nΔT are known. At this point, the subsystem... Boundary node voltage of subsystem β and The data is transmitted to the LCL decoupling branch in the middle.

[0046] S2: The LCL decoupling branch obtains the boundary node voltage. Then, the inductor current flowing through inductor L1 at time (n+1)ΔT will be calculated according to formula (6). .

[0047] S3: Based on the forward branching scheme, obtain the inductor current in step S2. After obtaining the value, combine it with the boundary node voltages known in step S1. The inductor current flowing through inductor L2 at time (n+1)ΔT will be obtained according to formula (7). .

[0048] S4: The LCL branch further calculates the intermediate capacitor voltage at time (n+1)ΔT according to formula (9). After the calculation is completed, this voltage will be fed back to the subsystem as a controlled voltage source. And subsystem β.

[0049] S5: Subsystem The intermediate capacitor voltage of subsystem β at time (n+1)ΔT Then, the network equations of each subsystem are solved in parallel, and all state variables within the subsystem are updated. This process continues until all state variables in the network at time (n+1)ΔT are obtained. At this point, the system completes the simulation for the (n+1)th time step and returns to step S1 to proceed to the calculation for the next time step.

[0050] Establish a pure implicit integration rule base: Discretize the differential relationships of energy storage elements (inductors, capacitors) using the backward pure implicit integration method. Let the simulation step size be Δt, and the time corresponding to the nth step be t=nΔt.

[0051] Discretization of inductor components: Continuous model: The branch of inductor L and series resistor R satisfies .

[0052] Discretization: in the interval Applying the backward Euler method: ; Substitute into the continuous model and take the value at time (n+1)Δt: ; Reconstructing the Norton equivalent circuit: The above equation can be rewritten as: ; Define parameters Then the equivalent model simplifies to: ; This equation is the one provided by the inductor branch when solving the network at time (n+1).

[0053] Discretization of capacitor components: Continuous model: The nodes of capacitor C and parallel conductance G satisfy... .

[0054] Discretization: Applying the backward Euler method: ; Reconstructed into Norton equivalent circuit: ; Define parameters Then the equivalent model simplifies to: ; This formula represents the relationship between the current in each branch connected to the capacitor node and the node voltage.

[0055] In software implementation, it is necessary to pre-set according to the formula. and Calculate the equivalent parameters (K, M) and the function of the history terms. Given circuit parameters. After setting the step size Δt, the program automatically calculates: ; And store historical data These form the basis for constructing the entire branch update equation.

[0056] Establish the complete discrete equation set for the LCL branch: Will Figure 3 LCL topology and formula and Combining these, we can list three core equations: First inductor branch (using formula) The voltage is ): ; Second inductor branch: ; Capacitor nodes (using formulas) The inflow current is ): ; (here) Using the value at time n satisfies the "known boundary conditions" setting in the forward branch strategy.

[0057] Algebraic derivation for performing forward branching: The strategy is sequential elimination, rather than solving simultaneous equations (Eq1, Eq2, Eq3).

[0058] Step S2: First solve : It was observed that both (Eq1) and (Eq3) contain unknowns. Transform (Eq1) into: ; Substitute (Eq1') into (Eq3) to eliminate After sorting, we get Explicit update formula: ; At this point, It can be calculated independently because its right side consists of known historical quantities or inputs.

[0059] Step S3: Then solve : The obtained Substitute (Eq1') into (Eq2). Specifically: Substituting (Eq1') into (Eq2), we get only... Equations with known quantities.

[0060] Solving the equation, we get Explicit update formula: ; At this point, It can also be calculated.

[0061] Step S4: Final Solution : The obtained and Substituting directly into (Eq3), we get Explicit update formula: ; can also be and Substituting the expression, we obtain formula (10) which is entirely represented by the historical quantities and the input.

[0062] Although the present invention has been described in detail with reference to the accompanying drawings and preferred embodiments, the present invention is not limited thereto. Various equivalent modifications or substitutions can be made to the embodiments of the present invention by those skilled in the art without departing from the spirit and essence of the invention, and such modifications or substitutions should all be within the scope of the present invention. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should also be covered within the protection scope of the present invention.

Claims

1. A method for decoupling electromagnetic transient simulation branches, characterized in that, Applicable to at least a subsystem Subsystem The power system sub-network model includes the LCL decoupling branch connecting the two, where the LCL decoupling branch includes a first inductor L1, a second inductor L2, and an intermediate capacitor C. The method includes the following steps: S1. At the end of the nth simulation step, obtain the subsystem... boundary node voltage and subsystems boundary node voltage The acquired boundary node voltage is then transmitted to the LCL decoupling branch. S2, the LCL decoupling branch is based on the boundary node voltage obtained in step S1. With the system stored The first inductor current is calculated using the state update formula under the forward branching strategy. ; S3, the LCL decoupling branch is based on the boundary node voltage obtained in step S1. The first inductor current in step S2 With the system stored The second inductor current is calculated using the state update formula under the forward branching strategy. ; S4, the LCL decoupling branch is based on the first inductor current in step S2. The second inductor current in step S3 With the system stored The intermediate capacitor voltage is calculated using the state update formula under the forward branching strategy. ; S5, adjust the voltage of the intermediate capacitor. Feedback is sent to the subsystems respectively. and subsystem For the subsystem and subsystem The respective network equations are solved in parallel and all state variables inside the subsystem are updated to complete the simulation of the (n+1)th simulation step. S6. Determine whether the simulation end time has been reached. If not, proceed to the next simulation time and return to step S1. If the time has been reached, end the simulation.

2. The electromagnetic transient simulation branch decoupling method according to claim 1, characterized in that, Step S1 includes: S11. At the end of the nth simulation step, the control subsystem With subsystem The system performs the update calculation of all internal state variables in parallel to obtain the system simulation result of the nth simulation step, where the state variables include the voltage of each node and the current flowing through each branch. S12. Based on the system simulation results obtained in step S11 for the nth simulation step, determine the subsystem. With subsystem Electrical quantities at the connection nodes between them, specifically electrical quantities of the subsystem Output boundary node voltage and subsystems Output boundary node voltage , where n represents the simulation time to which the electrical quantity belongs; S13. The boundary node voltage determined in step S12 is... and As a known input signal, it is transmitted from the subsystem via a data communication interface or shared memory. With subsystem The solution unit is transferred to the independent solution unit responsible for calculating the LCL decoupling branch.

3. The electromagnetic transient simulation branch decoupling method according to claim 2, characterized in that, Step S13, which involves transmitting the boundary node voltage to the LCL decoupling branch, specifically includes: S131, Set the boundary node voltage and Each voltage source is considered as an independent voltage source applied to both ends of the LCL decoupling branch. In the calculation model of the LCL decoupling branch, a system is established based on... The first controlled voltage source model for the output voltage and the model with The second controlled voltage source model for the output voltage; S132. In the equivalent circuit of the LCL decoupling branch, the positive terminal of the first controlled voltage source model is connected to the end of the first inductor L1 furthest from the intermediate capacitor C, and its negative terminal is connected to the subsystem. The reference ground potential is connected, and the positive terminal of the second controlled voltage source model is connected to the end of the second inductor L2 away from the intermediate capacitor C, and its negative terminal is connected to the reference ground potential of the subsystem β.

4. The electromagnetic transient simulation branch decoupling method according to claim 2, characterized in that, Step S2 includes: S21. In the independent solution unit of the LCL decoupling branch, the first inductor L1 is discretized using a pure implicit integration method to obtain the discretized first inductor current of the first inductor L1. : ； in, The first inductor current stored at the previous moment; The second inductor current stored in the previous moment; L1 is the voltage of the intermediate capacitor stored at the previous moment; C is the inductance value of the first inductor; Δt is the capacitance value of the intermediate capacitor; and Δt is the simulation step size. The resistance value is the value of the resistor connected in series with the first inductor L1; Let L1 be the inductance value of the first inductor. This is the conductance value connected in parallel with the intermediate capacitor C; S22, Discretize the first inductor current The forward branching strategy is used to update the calculation of the first inductor current. : ； in, M is the first equivalent admittance parameter associated with the first inductor branch, and M is the second equivalent admittance parameter associated with the capacitor branch. S23. Calculate the first inductor current obtained in step S22. The numerical values ​​are stored and output as known quantities.

5. The electromagnetic transient simulation branch decoupling method according to claim 4, characterized in that, Step S22, which involves applying a preset update formula for calculation, specifically includes: S221. Calculate the first equivalent admittance parameter based on the circuit parameters of the LCL decoupling branch and the simulation step size Δt. The second equivalent admittance parameter M, where , ; S222. Read the known input quantities from the storage unit, including the first inductor current at time n. Second inductor current Intermediate capacitor voltage and the boundary node voltage from step S1 ; S223. The parameters calculated in step S221 are... M and the input values ​​read in step S222 Along with the fixed circuit parameters L1, C, and simulation step size Δt, these are substituted into the update formula for the first inductor current to obtain the first inductor current at time n+1. Numerical solution.

6. The electromagnetic transient simulation branch decoupling method according to claim 4, characterized in that, Step S3 includes: S31. In the independent solution unit of the LCL decoupling branch, the second inductor L2 is discretized using a pure implicit integration method to obtain the discretized second inductor current of the second inductor L2. : ； Where L2 is the inductance value of the second inductor; The resistance value is the value of the resistor connected in series with the second inductor L2; Let L2 be the inductance value of the second inductor. S32, Discretized second inductor current The forward branching strategy is used to update the calculation of the second inductor current. : ； in, This is the second equivalent admittance parameter associated with the second inductor branch; S33. The second inductor current calculated in step S32 is... The numerical values ​​are stored and output as known quantities.

7. The electromagnetic transient simulation branch decoupling method according to claim 6, characterized in that, Step S32, which involves applying a preset update formula for calculation, specifically includes: S321. Based on the circuit parameters of the LCL decoupling branch and the simulation step size Δt, calculate the first equivalent admittance parameter respectively. The second equivalent admittance parameter M and the third equivalent admittance parameter Wherein, parameter M characterizes the equivalent discrete admittance of the intermediate capacitor C under the rules of the pure implicit integration method; , The resistance value is the value of the second inductor L2 connected in series; S322. Read all known input quantities from the storage unit, including the historical state quantity of the first inductor current at the nth simulation step. Second inductor current and intermediate capacitor voltage and the boundary node voltage obtained from step S1 and ; S323, The parameters calculated in step S321 are... The input quantity read in step S322 Along with fixed circuit parameters , , Substituting the simulation step size Δt into the second inductor current update formula, we obtain the second inductor current at the (n+1)th simulation step size. Numerical solution.

8. The electromagnetic transient simulation branch decoupling method according to claim 6, characterized in that, Step S4 includes: S41. In the independent solution unit of the LCL decoupling branch, the intermediate capacitor C is discretized using a pure implicit integration method to obtain the discretized intermediate capacitor voltage of the intermediate capacitor C. : ； S42, Discretized intermediate capacitor voltage The intermediate capacitor voltage is updated using a forward branching strategy. : ； S43. The intermediate capacitor voltage calculated in step S42 is... The numerical values ​​are stored and output as known quantities.

9. The electromagnetic transient simulation branch decoupling method according to claim 8, characterized in that, Step S5 includes: S51. After completing the calculation in step S4 in the independent solution unit of the LCL decoupling branch, calculate the intermediate capacitor voltage at the (n+1)th simulation step. The numerical values ​​are transmitted to the solution subsystem via data communication interface or shared memory. The computational unit and the computational unit responsible for solving the subsystem β; S52, in the subsystem In the calculation model, the received intermediate capacitor voltage Introducing it into its network equations as an independent voltage source excitation, specifically in the subsystem At the original connection node of the decoupling branch from LCL, establish an output voltage that is constant. A controlled voltage source to replace the original physical branch connection; S53. In the calculation model of subsystem β, perform the same operation as in step S52 simultaneously, and change the intermediate capacitor voltage. As an output voltage that is constant The controlled voltage source is introduced into its network equations, making the subsystem With subsystem β, given definite boundary conditions Then, each system independently and in parallel solves its own internal network algebraic equations and updates the state variables of all node voltages and branch currents, together completing the full system simulation of the (n+1)th simulation step.

10. The electromagnetic transient simulation branch decoupling method according to claim 9, characterized in that, In step S52, the intermediate capacitor voltage is introduced into the subsystem as a voltage source. The network equations specifically include: S521, in the subsystem In the equivalent circuit model, locate the boundary node that was originally connected to the first inductor L1 in the LCL decoupling branch, and mark this node as an internal node. And establish a new branch road between this node and the reference point; S522. Define the newly added branch as an ideal controlled voltage source, whose terminal voltage The value of t=(n+1)Δt at the simulation time is determined by the received intermediate capacitor voltage value, i.e., it is set as follows: The model parameters of the voltage source are then correlated with the dynamically updated values ​​of the intermediate capacitor voltage. S523, Subsystem modified based on steps S521 and S522 The circuit topology is reconstructed to form its nodal admittance matrix. and the right-hand current source vector Among them, the controlled voltage source The known potential difference is treated as a right-hand term in the network equation, and then the system of linear equations is solved. Obtain the voltage vectors of all internal nodes of the subsystem at time n+1. ,in It includes the voltage values ​​of all nodes except the reference ground.

Images