Electromechanical and electromagnetic hybrid simulation method and system for ultra-large-scale power system

Abstract

The application discloses a method and system for electromechanical and electromagnetic hybrid simulation of an ultra-large-scale power system, and relates to the field of power system simulation analysis, which comprises the following steps: collecting basic data of a power system to be simulated, and completing unified time reference and unified interface number processing to build a hybrid simulation database; then, based on the basic data, the power system is divided into an electromechanical transient side and an electromagnetic transient side, an interface bus is arranged between the two sides, and a Norton equivalence relationship at the interface is built through a node admittance matrix; further, frequency-related network equivalence fitting is performed in combination with electromechanical side frequency scanning data, and after state space modeling and discretization processing, interface equivalence parameters under the current communication period are output; consistency-related parameter calculation is performed, and a bidirectional consistency data interaction interface is established. The application realizes high-precision and high-consistency collaborative hybrid simulation between the electromechanical transient side and the electromagnetic transient side.

Description

Technical Field

[0001] This invention relates to the field of power system simulation analysis, and in particular to a hybrid electromechanical and electromagnetic simulation method and system for ultra-large-scale power systems. Background Technology

[0002] With the continuous integration of new energy power plants, flexible DC transmission systems, power electronic devices, and large-scale AC / DC hybrid systems into the power grid, modern power systems are gradually exhibiting characteristics of large structural scale, numerous control links, and complex dynamic processes. These systems contain both electromechanical transient processes reflecting unit speed, power angle, and power flow evolution, and electromagnetic transient processes reflecting converters, electromagnetic coupling, and fault impacts. In existing technologies, electromechanical transient simulation, electromagnetic transient simulation, and electromechanical-electromagnetic hybrid simulation have all been applied to power system operation analysis, fault research, control strategy evaluation, and grid connection performance verification. Electromechanical-electromagnetic hybrid simulation is also gradually becoming an important technical direction for the refined analysis of ultra-large-scale power systems.

[0003] However, existing technologies still have the following shortcomings: On the one hand, in the hybrid simulation of ultra-large-scale power systems, the equivalent representation of interface boundaries is still insufficient in reflecting the coupling relationship of multiple interfaces, frequency correlation characteristics, and dynamic memory characteristics; on the other hand, there are still insufficient consistency issues between the electromechanical transient side and the electromagnetic transient side in terms of interface quantity extraction, boundary mapping, and data interaction within the communication cycle. In particular, under fault disturbances, control switching, and three-phase unbalanced conditions, there is still room for further improvement in the bidirectional consistency transmission of interface quantities. Summary of the Invention

[0004] In view of the aforementioned existing problems, the present invention is proposed.

[0005] Therefore, this invention provides a hybrid electromechanical and electromagnetic simulation method for ultra-large-scale power systems to solve the problems of insufficient equivalent representation of interface boundaries and insufficient consistency of interface quantity interaction between the electromechanical transient side and the electromagnetic transient side in the existing technology.

[0006] To solve the above-mentioned technical problems, the present invention provides the following technical solution: In a first aspect, the present invention provides a hybrid electromechanical and electromagnetic simulation method for ultra-large-scale power systems, which includes: collecting basic data of the ultra-large-scale power system to be simulated, and unifying the time base and interface number of all data to form a hybrid simulation database.

[0007] Based on the basic data, the ultra-large-scale power system is divided into electromechanical transient side partition and electromagnetic transient side partition, and an interface bus is set between the two. The node admittance matrix of the electromechanical side network is established, and non-interface nodes are eliminated to obtain the equivalent admittance matrix and equivalent current at the interface. Norton equivalent relationship at the interface is constructed.

[0008] Based on the frequency scanning data from the electromechanical side, the Norton equivalent relationship is fitted by a frequency correlation network. Then, through state space modeling and discretization, the interface equivalent admittance matrix and interface equivalent injection current vector under the current communication cycle are output.

[0009] Based on the interface equivalent admittance matrix and the interface equivalent injected current column vector, calculate the consistent interface voltage phasor, consistent interface current phasor, three-sequence components, active power, and reactive power for the current communication cycle.

[0010] Establish a bidirectional consistent data exchange interface between the electromechanical transient side and the electromagnetic transient side, enabling the electromechanical transient side to exchange interface quantities with the electromagnetic transient side.

[0011] As a preferred embodiment of the electromechanical and electromagnetic hybrid simulation method for ultra-large-scale power systems described in this invention, the basic data includes: the topological connection relationship between the main grid and the substation, bus parameters, line parameters, transformer parameters, generator parameters, load parameters, protection and control parameters, interface bus location, interface branch impedance, system rated frequency, partition boundary information, three-phase instantaneous voltage and three-phase instantaneous current at the interface bus, frequency scan data of the electromechanical side admittance matrix, and interface message timestamps, message sequence numbers, transmission cycles, and packet loss flags on the communication side.

[0012] As a preferred embodiment of the electromechanical and electromagnetic hybrid simulation method for ultra-large-scale power systems described in this invention, the construction of the Norton equivalent relationship at the interface includes: in a single-port scenario with only one interface bus, injecting a current source at the interface location and setting the injected current of the remaining nodes to zero, obtaining the interface node voltage by solving the relationship between the network node voltage and the injected current corresponding to the node admittance matrix, and obtaining the equivalent admittance at the interface based on the interface node voltage.

[0013] As a preferred embodiment of the electromechanical and electromagnetic hybrid simulation method for ultra-large-scale power systems described in this invention, the Norton equivalent relationship at the interface further includes: in scenarios with multiple hybrid simulation interface buses, treating all interface nodes as a unified set of coupling ports, dividing the node admittance matrix into blocks according to non-interface nodes and interface nodes, and eliminating variables of non-interface nodes to obtain the interface equivalent admittance matrix and the equivalent current converted from non-interface nodes to interface nodes.

[0014] By combining the injected current and voltage at the interface node, a Norton-form relationship between the interface voltage and current is constructed.

[0015] As a preferred embodiment of the electromechanical and electromagnetic hybrid simulation method for ultra-large-scale power systems described in this invention, the step of performing frequency-related network equivalent fitting on the Norton equivalent relationship at the interface based on the frequency scanning data of the electromechanical side includes: performing frequency scanning on the electromechanical side network outside the interface based on the determined set of interface nodes and node order to obtain node admittance matrix samples at discrete frequency points.

[0016] The vector fitting method is used to parametrically fit each admittance element in the node admittance matrix sample, and each admittance element is expressed as a rational function to construct an equivalent model of the frequency correlation network.

[0017] As a preferred embodiment of the electromechanical and electromagnetic hybrid simulation method for ultra-large-scale power systems described in this invention, the frequency-dependent network equivalent model is converted into a state-space form, and boundary dynamic memory state variables are added to establish a dynamic equivalent model of the electromechanical side interface boundary.

[0018] The dynamic equivalent model is discretized using the implicit trapezoidal integral method, and the discretized interface boundary relationships are organized into Norton-type boundaries under the current communication cycle to output the interface equivalent admittance matrix and the interface equivalent injection current vector of the current communication cycle.

[0019] As a preferred embodiment of the electromechanical and electromagnetic hybrid simulation method for ultra-large-scale power systems described in this invention, the calculation of the consistency interface voltage phasor, consistency interface current phasor, three-sequence components, active power, and reactive power in the current communication cycle includes: recovering the three-phase voltage phasor and three-phase current phasor using the three-phase instantaneous voltage and current at the interface, and calculating the active power and reactive power in the current communication cycle.

[0020] As a preferred embodiment of the electromechanical and electromagnetic hybrid simulation method for ultra-large-scale power systems described in this invention, the calculation of the consistent interface voltage phasor, consistent interface current phasor, three-sequence components, active power, and reactive power for the current communication cycle further includes: reading the interface equivalent injection current vector for the current communication cycle, combining it with the recovered active power and reactive power, performing consistency recovery on the interface quantities to obtain the consistent interface voltage phasor for the current communication cycle; and calculating the consistent interface current phasor for the current communication cycle based on the consistent interface voltage phasor, the interface equivalent admittance matrix, and the interface equivalent injection current vector for the current communication cycle.

[0021] As a preferred embodiment of the electromechanical and electromagnetic hybrid simulation method for ultra-large-scale power systems described in this invention, the establishment of a bidirectional consistent data interaction interface between the electromechanical transient side and the electromagnetic transient side includes: sending the consistent interface voltage phasor, consistent interface current phasor, three-sequence components, active power and reactive power of the current communication cycle output by the electromagnetic transient side to the electromechanical transient side.

[0022] After reading the interface quantities from the electromechanical transient side, the interface boundary state is updated by combining the interface boundary dynamic equivalent model, generating the interface boundary voltage phasor and interface boundary current phasor fed back by the electromechanical side in the current communication cycle.

[0023] The electromagnetic transient side output and the electromechanical transient side feedback are exchanged correspondingly under the same interface number, the same time base, and the same communication cycle.

[0024] As a preferred embodiment of the electromechanical and electromagnetic hybrid simulation method for ultra-large-scale power systems described in this invention, the establishment of a bidirectional consistent data interaction interface between the electromechanical transient side and the electromagnetic transient side further includes: calculating voltage consistency deviation, current consistency deviation, and power consistency deviation based on the consistency interface quantity output by the electromagnetic transient side and the interface boundary quantity fed back by the electromechanical side, respectively, to determine the comprehensive consistency deviation.

[0025] When the overall consistency deviation meets the preset conditions, the joint simulation of the next communication cycle begins.

[0026] When the overall consistency deviation does not meet the preset conditions, interface consistency correction is performed.

[0027] In the event of fault disturbance, control switching, sudden change in three-phase imbalance, or continuous increase in overall consistency deviation, the consistency enhancement transmission mode is activated.

[0028] Secondly, the present invention provides a hybrid electromechanical and electromagnetic simulation system for ultra-large-scale power systems, comprising: a data construction module for collecting basic data of the ultra-large-scale power system to be simulated, and unifying the time base and interface number of all data to form a hybrid simulation database.

[0029] The network equivalent module is used to divide the ultra-large-scale power system into electromechanical transient side partitions and electromagnetic transient side partitions based on basic data, set up interface buses, establish the node admittance matrix of the electromechanical side network, eliminate non-interface nodes, obtain the equivalent admittance matrix and equivalent current at the interface, and construct the Norton equivalent relationship at the interface.

[0030] The dynamic fitting module is used to perform frequency correlation network equivalent fitting on Norton equivalent relationships based on frequency scanning data from the electromechanical side, and outputs the interface equivalent admittance matrix and interface equivalent injection current vector under the current communication cycle through state space modeling and discretization.

[0031] The parameter calculation module is used to calculate the consistent interface voltage phasor, consistent interface current phasor, three-sequence components, active power, and reactive power of the current communication cycle based on the interface equivalent admittance matrix and the interface equivalent injected current column vector.

[0032] The bidirectional interaction module is used to establish a bidirectional consistent data interaction interface between the electromechanical transient side and the electromagnetic transient side, enabling the electromechanical transient side and the electromagnetic transient side to exchange interface quantities.

[0033] The beneficial effects of this invention are as follows: By performing frequency-related network equivalent fitting, dynamic equivalent modeling, and communication cycle discretization on the electromechanical transient side interface boundary, the interface boundary parameters can be output and updated according to the communication cycle, thereby ensuring the continuity and temporal coordination of the interface boundary representation during hybrid simulation. By extracting the fundamental wave of the three-phase instantaneous quantities on the electromagnetic transient side, restoring interface quantity consistency, and performing bidirectional consistency verification and feedback correction, the bidirectional consistency transmission of interface quantities between the electromechanical transient side and the electromagnetic transient side is achieved, ensuring consistency and stability of interface interaction during the joint simulation of ultra-large-scale power systems. Attached Figure Description

[0034] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0035] Figure 1 This is a flowchart of a hybrid electromechanical and electromagnetic simulation method for ultra-large-scale power systems.

[0036] Figure 2 The schematic diagram for solving the electromechanical equivalent admittance for the Norton equivalent circuit.

[0037] Figure 3 This is a schematic diagram of a multi-port interface bus.

[0038] Figure 4 This is a time-series distribution characteristic diagram of the change in admittance matrix under different schemes during the fault disturbance process.

[0039] Figure 5 This diagram illustrates the peak differences and fallback process of the injected current variation in the initial stage of a fault under different schemes. Detailed Implementation

[0040] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0041] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0042] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a single or selective embodiment that is mutually exclusive with other embodiments.

[0043] Reference Figures 1-5 This is one embodiment of the present invention, which provides a hybrid electromechanical and electromagnetic simulation method for ultra-large-scale power systems, including the following steps: S1. Collect basic data of the ultra-large-scale power system to be simulated, and unify the time base and interface number of all data to form a hybrid simulation database.

[0044] Furthermore, the basic data includes: the topology connection relationship between the main network and the substation, bus parameters, line parameters, transformer parameters, generator parameters, load parameters, protection and control parameters, interface bus location, interface branch impedance, system rated frequency, partition boundary information, three-phase instantaneous voltage and three-phase instantaneous current at the interface bus, frequency scan data of electromechanical side admittance matrix, and interface message timestamps, message sequence numbers, transmission cycles, and packet loss flags on the communication side.

[0045] It should be noted that, firstly, basic data collection is performed on the ultra-large-scale power system to be simulated. The collected data includes: the topology of the main network and substations, bus parameters, line parameters, transformer parameters, generator parameters, load parameters, protection and control parameters, interface bus locations, interface branch impedances, system rated frequency, and partition boundary information. In this embodiment, structured network parameters are uniformly exported from the EMS (Energy Management System) model library; three-phase instantaneous voltage and three-phase instantaneous current at the interface bus are uniformly collected by a fault recording device; electromechanical side admittance matrix frequency scan data are uniformly obtained by applying frequency perturbations to the electromechanical side partitioned network in the PowerFactory power system simulation analysis software, used for subsequent frequency-related network equivalence fitting; and interface message timestamps, message sequence numbers, transmission cycles, and packet loss markers on the communication side are uniformly recorded using UDP (User Datagram Protocol) communication for subsequent synchronization interaction.

[0046] It should also be noted that, in order to support hybrid simulation, preferably, the data refresh cycle of the electromechanical side is set to 5~20ms, the waveform sampling cycle of the electromagnetic side is set to 20~100μs, and all data are given a unified time base and a unified interface number to form a hybrid simulation database that can be called by both the electromechanical transient solver and the electromagnetic transient solver.

[0047] It should be noted that interface nodes are preferentially selected as grid connection points of new energy power plants, AC buses of flexible DC converter stations, critical connection buses at the receiving end, or boundary buses of areas highly sensitive to local faults, in order to ensure that the physical partitions between the electromechanical side and the electromagnetic side have clear electrical boundaries.

[0048] The high-sensitivity areas of local faults are determined based on the voltage sensitivity index obtained from offline fault simulation. The voltage sensitivity index is derived from the average statistical results of the voltage change rate of the candidate bus at the interface under typical three-phase short circuit, two-phase short circuit, and tripping disturbance conditions.

[0049] The voltage sensitivity index is used to characterize the voltage response sensitivity of candidate buses under typical disturbances. Typical disturbances include at least three-phase short circuits, two-phase short circuits, and tripping disturbances, and preferably cover representative scenarios under different fault locations, different fault durations, and different operating modes. For each candidate bus, its voltage change rate under each disturbance condition is statistically analyzed, and the statistical results under each condition are averaged to obtain the voltage sensitivity index corresponding to the candidate bus.

[0050] To ensure that the interface partitioning results cover dynamically sensitive areas without overly expanding the electromagnetic transient range, it is preferable to sort all candidate buses by their voltage sensitivity indices from highest to lowest. Then, considering offline fault simulation results, partition size constraints, and interface quantity control requirements, the candidate buses with the highest voltage sensitivity indices are selected as the high-sensitivity interface boundary buses. Preferably, the proportion of high-sensitivity interface boundary buses among all candidate buses is controlled within the range of 20% to 30%; a lower proportion is used when fault-sensitive areas are concentrated, and a higher proportion is used when there are many new energy access points, flexible DC converter stations, or critical interconnection sections in the system.

[0051] When the voltage sensitivity indices of multiple candidate buses are similar and all are located near the boundary of the high-sensitivity range, the bus that meets at least one of the following conditions should be preferentially selected as the interface boundary bus: First, it is located at the grid connection point of the new energy power station or the AC bus of the flexible DC converter station; second, it is located at the critical connection section of the receiving end; third, it exhibits significant voltage drop, current surge, or power fluctuation after a fault disturbance. This avoids the problem of unstable interface boundary selection caused by relying on a single statistical value.

[0052] Interface nodes are preferentially selected as grid connection points of new energy power plants, AC busbars of flexible DC converter stations, critical connection busbars at the receiving end, or boundary busbars of areas highly sensitive to local faults, so as to ensure that the physical partitions between the electromechanical side and the electromagnetic side have clear electrical boundaries.

[0053] The consistency of the topology section and the measurement section is checked. If the interface node naming is inconsistent, the switch status is mismatched, or the parameters are missing, the error is corrected before the equivalent value is executed.

[0054] S2. Based on the basic data, the ultra-large-scale power system is divided into electromechanical transient side partition and electromagnetic transient side partition, and an interface bus is set between the two. The node admittance matrix of the electromechanical side network is established, and non-interface nodes are eliminated to obtain the equivalent admittance matrix and equivalent current at the interface. Norton equivalent relationship at the interface is constructed.

[0055] Furthermore, constructing the Norton equivalent relationship at the interface includes: in a single-port scenario with only one interface bus, injecting a current source at the interface location and injecting current to zero at the remaining nodes, obtaining the interface node voltage by solving the relationship between the network node voltage and the injected current corresponding to the node admittance matrix, and obtaining the equivalent admittance at the interface based on the interface node voltage.

[0056] The Norton equivalent relationship at the interface also includes: in scenarios with multiple hybrid simulation interface buses, all interface nodes are treated as a unified set of coupling ports, the node admittance matrix is ​​divided into blocks according to non-interface nodes and interface nodes, and the variables of non-interface nodes are eliminated to obtain the interface equivalent admittance matrix and the equivalent current from non-interface nodes to interface nodes.

[0057] By combining the injected current and voltage at the interface node, a Norton-form relationship between the interface voltage and current is constructed.

[0058] It should be noted that the ultra-large-scale power system is divided into electromechanical transient side partitions and electromagnetic transient side partitions, with an interface bus between them. At the electromechanical transient solution scale, the large-scale power grid outside the interface bus is converted into a Norton equivalent network that accurately reflects the voltage-current relationship at the interface. This reduces the network dimensionality and solution burden of the co-simulation of the ultra-large-scale power system without directly retaining the detailed component models of all distant networks. If the detailed model of the entire network is directly retained on the electromagnetic transient side, the dimensionality of the state variables would be too high, making real-time performance difficult to guarantee. If an oversimplified static boundary is used, the support characteristics and coupling effects of the distant main network on the interface bus would be lost. Therefore, the node admittance matrix of the electromechanical side network is first established based on the system topology, line impedance, transformer parameters, generator reactance, load admittance, and bus connection relationships.

[0059] It should also be noted that, assuming the electromechanical network is a linear network and has only one port, the following expansion relationship is assumed between the node admittance matrix, the network node voltage, and the current injected into the network: ; in, The number of nodes in the electromechanical side network is represented by... to The node admittance matrix constitutes the electromechanical side network, where diagonal elements represent the self-admittance of the corresponding node, and off-diagonal elements represent the mutual admittance between different nodes. to Forming the voltage vector of network nodes, to This forms the node injection current vector. The relationship between network node voltages and the current injected into the network is used to describe the matrix mapping relationship between the node voltages and the injected currents in the electromechanical side network.

[0060] For a single-port scenario with only one interface bus, injecting a unit current source at the interface location and setting the injected current to zero at the remaining nodes, the interface node voltage can be obtained by solving the relationship between the network node voltage and the injected current. Based on the interface node voltage, the equivalent admittance at the interface can be further calculated, expressed as: ; in, The equivalent admittance at the interface. It means equivalent, and is an abbreviation of "equivalent". Indicates the interface node voltage. Indicates the interface node. This represents the unit current. The complex network outside the interface is converted into parallel admittance at the interface, thus characterizing the external network's ability to support the interface node voltage.

[0061] For scenarios with multiple hybrid simulation interface buses, instead of calculating each interface node as an independent single port, all interface nodes are treated as a unified set of coupled ports. This is because in ultra-large-scale systems, different interfaces often exhibit significant correlations through main network channels, reactance coupling, and power flow distribution. Simply splitting them into multiple independent single ports easily overlooks the coupling effects between interfaces, leading to amplified boundary errors during hybrid simulation. Therefore, as... Figure 3 As shown, assuming the electromechanical side network has a total of There are 1 node, and the interface node is 1. There are 1, then there are 1 non-interface nodes. Rearrange the node positions of each node so that the interface node is at the end of the matrix, and establish the following block admittance relationship, expressed as: ; in, Indicates the voltage of the non-interface node. Indicates the interface node voltage. This indicates the injected current at non-interface nodes. This indicates the injected current at the interface node. This represents the admittance submatrix between non-interface nodes. This represents the admittance submatrix between non-interface nodes and interface nodes. This represents the admittance submatrix between interface nodes and non-interface nodes. This represents the admittance submatrix between interface nodes.

[0062] Furthermore, to retain only interface boundary variables, the non-interface node variables in the block admittance relation are eliminated, resulting in: ; in, express The inverse matrix, This represents the interface equivalent admittance matrix obtained after eliminating non-interface nodes. This represents the equivalent current term referred to the interface node after conversion from the non-interface node. To facilitate standardized interface exchange with the electromagnetic transient side, the equivalent admittance matrix of the interface is uniformly defined as: ; ; in, express The inverse matrix, This represents the interface equivalent admittance matrix obtained after eliminating non-interface nodes. This represents the equivalent current referred from the non-interface node to the interface. Further combining this with the equivalent injected current from the external electromechanical network, the voltage-current relationship at the interface can be reorganized into Norton form, expressed as: ; The fundamental frequency equivalent admittance relationship of the electromechanical side interface was obtained, but the voltage and current relationship at the interface mainly reflects the static support characteristics near the power frequency. For the electromechanical and electromagnetic hybrid simulation of ultra-large-scale power systems, in addition to power frequency power flow exchange, there may also be low-frequency oscillation components, subsynchronous coupling components, and transient components at the interface.

[0063] S3. Based on the frequency scanning data of the electromechanical side, the Norton equivalent relationship is fitted by the frequency correlation network, and the interface equivalent admittance matrix and interface equivalent injection current vector under the current communication cycle are output through state space modeling and discretization.

[0064] Furthermore, based on the frequency scanning data of the electromechanical side, frequency correlation network equivalent fitting of the Norton equivalent relationship at the interface includes: based on the determined set of interface nodes and node order, performing frequency scanning on the electromechanical side network outside the interface to obtain node admittance matrix samples at discrete frequency points.

[0065] The vector fitting method is used to parametrically fit each admittance element in the node admittance matrix sample, and each admittance element is expressed as a rational function to construct an equivalent model of the frequency correlation network.

[0066] The frequency scan data is derived from the node admittance response samples obtained after applying frequency perturbations to the electromechanical transient side network outside the interface. The scan frequency band preferably covers the steady-state support range, low-frequency oscillation range, subsynchronous coupling range, and transition response frequency bands that may be involved in fault transients and control switching near the power frequency, so as to ensure that the constructed frequency-dependent network equivalent model can simultaneously reflect the response characteristics of the interface boundary in steady-state, low-frequency dynamic, and transient processes.

[0067] In the vector fitting process, the fitting poles, residues, constant term coefficients, and first-order term coefficients are all obtained through iterative fitting from the nodal admittance matrix samples at discrete frequency points. Specifically, the fitting poles characterize the dominant dynamic response features of the electromechanical network outside the interface in different frequency ranges; the residues characterize the coupling strength between interface nodes in the corresponding dynamic modes; the constant term coefficients reflect the static support components of the admittance elements; and the first-order term coefficients reflect the linear change trend of the admittance elements during frequency variations. None of these parameters are directly given manually; rather, they are obtained driven by frequency scan sample data.

[0068] The selection of the total number of fitting poles is based on the principle of balancing fitting accuracy and computational efficiency. For interface admittance elements with relatively smooth response changes, fewer fitting poles are sufficient to meet the boundary equivalence requirements. For interface admittance elements with obvious resonance characteristics, subsynchronous coupling characteristics, or dynamic transition characteristics, the total number of fitting poles is increased to improve the approximation accuracy of the frequency-related response. Preferably, the convergence of the deviation in amplitude and phase between the fitted model and the original scanned sample is used as the criterion. If the fitting deviation still significantly affects the dynamic response of the interface boundary, the total number of fitting poles is increased and the fitting is re-executed.

[0069] To ensure that the fitted frequency-dependent network equivalent model can stably participate in time-domain co-simulation, stability and physical rationality checks are performed on the fitting results after vector fitting. If significant non-physical oscillations, response distortion, or numerical instability are found in the fitting results, the fitting is re-performed by redistributing initial poles, increasing sampling frequency density, or adjusting the fitting order until a frequency-dependent network equivalent model that can be used for subsequent state-space modeling is obtained.

[0070] Furthermore, boundary dynamic memory state variables are used to record the dynamic state information of the electromechanical transient side network outside the interface, which still has a continuing influence on the current interface boundary response within the preceding communication cycle. Since there is not only instantaneous power flow exchange at the interface but also historical influences of dynamic processes, boundary dynamic memory state variables are introduced when converting the frequency-dependent network equivalent model into state-space form. This ensures that the discretized expression of the interface boundary reflects not only the input of the current communication cycle but also the continuous effect of the preceding communication cycle state on the current cycle. The purpose of using the implicit trapezoidal integral method for discretization is to ensure that the dynamic equivalent model is consistent with the communication cycle of the co-simulation and to improve the numerical stability during the discretization solution process.

[0071] The frequency-dependent network equivalent model is converted into a state-space form, and boundary dynamic memory state variables are added to establish a dynamic equivalent model of the electromechanical side interface boundary.

[0072] The implicit trapezoidal integral method is used to discretize the dynamic equivalent model, and the discretized interface boundary relationship is organized into Norton-type boundary under the current communication cycle to output the interface equivalent admittance matrix and the interface equivalent injection current vector of the current communication cycle.

[0073] It should be noted that for the electromechanical and electromagnetic hybrid simulation of ultra-large-scale power systems, the interface not only involves power frequency power flow exchange, but may also exhibit low-frequency oscillation components, subsynchronous coupling components, and transient response components higher than the power frequency under fault transients, control switching, and unbalanced operating conditions. If a fixed power frequency admittance is still used for boundary substitution, although it can approximately reflect the steady-state support relationship, it is difficult to accurately characterize the phase lag, frequency dependence, and historical evolution of the dynamic current response at the interface, thus leading to distortion of energy exchange between the electromechanical and electromagnetic sides.

[0074] Therefore, based on the obtained Norton equivalent, the frequency correlation characteristics and dynamic memory characteristics of the network are further introduced, so that the electromechanical side boundary is transformed from a static admittance to a dynamic equivalent model that can be discretely solved.

[0075] It should also be noted that, firstly, the voltage-current Norton relationship network on the electromechanical transient side is scanned using frequency scanning data to obtain several discrete frequency point nodal admittance matrix samples. To represent these discrete frequency point admittance matrix samples as a network function that continuously varies with frequency, the FDNE (Frequency Dependent Network Equivalents) model is adopted. The essence of FDNE is a continuously varying frequency nodal admittance matrix, where each element is a function of frequency. FDNE can be represented as a nodal admittance matrix that is a function of frequency.

[0076] The total number of interface nodes is Then, at any scanning frequency, the node admittance matrix corresponding to the electromechanical network on the outer side of the interface is denoted as: ; in, This represents the interface node admittance matrix in the frequency domain. Indicates the first The first interface node and the first Frequency-dependent admittance elements between interface nodes Represents frequency domain variables, and Both represent the interface node number.

[0077] After obtaining the frequency response in matrix form, each admittance element in the nodal admittance matrix is ​​parametrically fitted. A vector fitting method is preferred, where the first element is parametrically fitted to the second element. The first interface node and the first The frequency-dependent admittance element between interface nodes is represented as follows: ; in, This represents the total number of fitted poles. Indicates the index of the fitted pole. Indicates the first Each fitted pole, Indicates the first The first interface node and the first The interface node is at the The residues corresponding to the fitted poles Indicates the first The first interface node and the first The constant term coefficients among the interface nodes Indicates the first The first interface node and the first The coefficients of the linear terms among the interface nodes.

[0078] After combining all admittance elements, the equivalent model of the frequency-dependent network at the interface boundary is obtained. To enable the frequency-domain fitting results to participate in time-domain co-simulation, the equivalent model of the frequency-dependent network is further converted into a generalized state-space form, expressed as: ; ; in, Represents the state matrix, Represents the input matrix, Indicates the output matrix. Represents a through matrix. Represents the identity matrix. Represents the boundary dynamic memory state variable. This represents the derivative of the boundary dynamic memory state variable with respect to time. Represents the interface boundary voltage vector. This represents the interface boundary current vector. The boundary dynamic memory state variable is used to characterize the dynamic influence of the electromechanical network outside the interface on the interface boundary at the current and historical moments.

[0079] To ensure the communication cycle of the dynamic equivalent model remains consistent with the joint simulation of the electromechanical and electromagnetic transient sides, an implicit trapezoidal integral method is used to discretize the generalized state-space form. Let the communication step size be... Then the first Each communication moment satisfies: ; ; in, Indicates the first Boundary dynamic memory state variables at each communication moment Indicates the first Boundary dynamic memory state variables at each communication moment Indicates the first Interface boundary voltage vector at each communication moment Indicates the first Interface boundary voltage vector at each communication moment Indicates the first Interface boundary current vector at each communication moment Indicates the communication step size.

[0080] Rearranging the discrete relationships into Nortonian boundary expressions for the current communication cycle, we obtain: ; in, Represents the interface equivalent admittance matrix. Indicates the first The interface equivalent injected current vector at each communication moment.

[0081] Furthermore, the interface equivalent admittance matrix Represented as: ; No. Interface equivalent injection current vector at each communication moment Represented as: ; in, The equivalent admittance relation used to characterize the interface boundary of the current communication cycle. Used to characterize the The equal-value injection relationship at the interface boundary at each communication moment.

[0082] S4: Based on the interface equivalent admittance matrix and the interface equivalent injected current column vector, calculate the consistent interface voltage phasor, consistent interface current phasor, three-sequence components, active power and reactive power of the current communication cycle.

[0083] Furthermore, calculating the consistent interface voltage phasor, consistent interface current phasor, three-sequence components, active power, and reactive power for the current communication cycle includes: recovering the three-phase voltage phasor and three-phase current phasor using the three-phase instantaneous voltage and current at the interface, and calculating the active power and reactive power within the current communication cycle.

[0084] The calculation of the consistent interface voltage phasor, consistent interface current phasor, three-sequence components, active power, and reactive power for the current communication cycle also includes: reading the interface equivalent injected current vector for the current communication cycle, combining it with the active power and reactive power recovered from the three-phase instantaneous voltage and three-phase instantaneous current, performing consistency recovery on the interface quantities, and obtaining the consistent interface voltage phasor for the current communication cycle.

[0085] Furthermore, based on the consistent interface voltage phasor, the interface equivalent admittance matrix, and the interface equivalent injected current vector, the consistent interface current phasor for the current communication cycle is calculated.

[0086] Among them, the interface equivalent admittance matrix is ​​used to characterize the admittance constraint relationship of the interface boundary in the current communication cycle, and participates in the boundary recovery in the process of consistency interface current phasor calculation and subsequent interface consistency verification; the interface equivalent injected current vector is used to characterize the equivalent injection relationship of the interface boundary in the current communication cycle.

[0087] It should be noted that the fundamental frequency extraction, power calculation, and interface quantity recovery are performed on the three-phase instantaneous voltage and three-phase instantaneous current of the interface bus collected from the electromagnetic transient side to obtain the first... Consistent interface voltage phasor, consistent interface current phasor, three-sequence components, active power, and reactive power at each communication moment.

[0088] This embodiment uses a single interface node as an example for illustration. For a single interface node, the interface equivalent admittance matrix output in step S3 is... The corresponding equivalent admittance is denoted as In the The equivalent injected current vector at each communication moment interface The corresponding equivalent injection current is denoted as .

[0089] First, the three-phase instantaneous voltages and three-phase instantaneous currents acquired by the fault recording device are preprocessed. The three-phase instantaneous voltages are denoted as follows: and The three-phase instantaneous currents are denoted as follows: and .in, Representing time respectively The instantaneous values ​​of phase A, phase B, and phase C voltages. Representing time respectively The instantaneous values ​​of phase A, phase B, and phase C currents. Preprocessing includes time alignment, dimensional unification, and outlier handling to ensure that the electromagnetic side sampling time aligns with the... Each communication moment corresponds to a specific time.

[0090] The following uses the A-phase voltage as an example to illustrate the fundamental component extraction process. Assume the instantaneous value of the A-phase voltage satisfies: ; in, This represents the fundamental amplitude of phase A voltage. Indicates the fundamental angular frequency. This represents the fundamental phase angle of phase A voltage.

[0091] Projecting the phase A voltage onto the sine and cosine bases respectively, we get: ; ; in, This represents the average projected component of phase A voltage onto the sinusoidal base. This represents the average projected component of the phase A voltage onto the cosine basis. Indicates the fundamental frequency period. This represents the integral variable.

[0092] Then the fundamental amplitude and fundamental phase angle of phase A voltage are respectively: ; ; Using the same method, the fundamental amplitude of phase B voltage can be obtained separately. Phase angle of fundamental wave of phase B voltage C-phase voltage fundamental amplitude Phase angle of the fundamental wave of phase C voltage For three-phase currents, the fundamental amplitude of phase A current can be obtained using the same process. B-phase current fundamental amplitude C-phase current fundamental amplitude and corresponding phase angle .

[0093] After obtaining the fundamental amplitude values ​​of the three-phase voltage, the fundamental amplitude values ​​of the three-phase current, and the corresponding phase angles, the three-phase voltage phasors and the three-phase current phasors can be further recovered, and three-sequence transformation can be performed to obtain the positive-sequence voltage, negative-sequence voltage, zero-sequence voltage, as well as the positive-sequence current, negative-sequence current, and zero-sequence current.

[0094] Based on this, calculate the first... Active power at each communication interface and reactive power Preferably, the active power and reactive power are obtained based on the recovered three-phase voltage phasors and three-phase current phasors.

[0095] To ensure that the interface quantities recovered from the electromagnetic side maintain consistency with the equivalence relationship of the interface boundaries, further measures are taken. and Perform boundary consistency matching on the interface quantities. Consistent interface voltage phasor at each communication moment Represented as: ; in, Indicates the first The consistent interface voltage phasor at each communication moment, where j represents the imaginary unit. Indicates the first Equivalent injection current at each communication moment .

[0096] Furthermore, the first Consistent interface current phasor at each communication moment Represented as: ; in, Indicates the first Consistent interface current phasor at each communication moment.

[0097] S5: Establish a bidirectional consistent data exchange interface between the electromechanical transient side and the electromagnetic transient side, enabling the electromechanical transient side and the electromagnetic transient side to exchange interface quantities.

[0098] Furthermore, establishing a bidirectional consistent data interaction interface between the electromechanical transient side and the electromagnetic transient side includes sending the consistent interface voltage phasor, consistent interface current phasor, three-sequence components, active power, and reactive power of the current communication cycle output by the electromagnetic transient side to the electromechanical transient side.

[0099] After reading the interface quantities from the electromechanical transient side, the interface boundary state is updated by combining the interface boundary dynamic equivalent model, generating the interface boundary voltage phasor and interface boundary current phasor fed back by the electromechanical side in the current communication cycle.

[0100] The electromagnetic transient side output and the electromechanical transient side feedback are exchanged correspondingly under the same interface number, the same time base, and the same communication cycle.

[0101] Establishing a bidirectional consistency data interaction interface between the electromechanical transient side and the electromagnetic transient side also includes: calculating voltage consistency deviation, current consistency deviation, and power consistency deviation based on the consistency interface quantity output by the electromagnetic transient side and the interface boundary quantity fed back by the electromechanical side, and determining the comprehensive consistency deviation.

[0102] When the overall consistency deviation meets the preset conditions, the joint simulation of the next communication cycle begins.

[0103] When the overall consistency deviation does not meet the preset conditions, interface consistency correction is performed.

[0104] In the event of fault disturbance, control switching, sudden change in three-phase imbalance, or continuous increase in overall consistency deviation, the consistency enhancement transmission mode is activated.

[0105] Preset conditions refer to the first The overall consistency deviation at any given communication time is no greater than the allowable threshold. Allowable threshold The threshold values ​​are derived from offline co-simulation calibration results, specifically obtained by tuning the interface deviation statistics under typical steady-state operating conditions, fault disturbance conditions, control switching conditions, and three-phase unbalance conditions. During threshold tuning, priority is given to ensuring the stability of bidirectional interface quantity transmission and the convergence of the co-simulation, while also considering interface interaction accuracy and computational efficiency. Preferably, a threshold value of [missing information] is allowed. The threshold can be set in stages according to scale, number of interfaces, disturbance intensity, and communication cycle length. Under normal operating conditions, a smaller threshold should be used, while under strong disturbance conditions, the threshold can be appropriately relaxed, but it should be kept within a range that can ensure the consistency of interface transmission.

[0106] When the overall consistency deviation exceeds the allowable threshold When an inconsistency is detected in the interface transmission between the electromechanical transient side and the electromagnetic transient side during the current communication cycle, interface consistency correction is triggered. Interface consistency correction preferably includes at least one of the following methods: First, based on the interface boundary voltage phasor fed back from the electromechanical side, the consistent interface voltage phasor recovered during the current communication cycle is corrected; second, based on the corrected interface voltage phasor, the consistent interface current phasor is recalculated; third, keeping the interface number, time base, and boundary state variables of the current communication cycle unchanged, only the interface boundary quantities are remapped and recalculated. Through these methods, the interface boundary quantities return to the dynamic boundary constraints of the electromechanical side without changing the overall partition structure.

[0107] The enhanced consistency transmission mode is used to handle more dynamic scenarios such as fault disturbances, control switching, sudden changes in three-phase imbalance, or continuous expansion of deviations. Specifically, when the overall consistency deviation continues to increase over multiple consecutive communication cycles, or when a sudden deviation significantly exceeds the allowable range occurs within a single communication cycle, the current interface boundary quantity exchange is considered to have entered a highly dynamic and sensitive state, and the enhanced consistency transmission mode is activated. The enhanced consistency transmission mode includes at least one of the following: increasing the frequency of consistency deviation verification, shortening the interface quantity exchange cycle, increasing the number of boundary state updates within the same communication cycle, and prioritizing the transmission of interface quantities that have undergone significant changes. Through these measures, the electromechanical transient side and the electromagnetic transient side can maintain higher frequency and higher precision bidirectional consistency transmission even in highly dynamic scenarios.

[0108] Furthermore, once the consistency enhancement transmission mode is activated, if the overall consistency deviation falls back and stabilizes at the allowable threshold within a certain number of consecutive communication cycles... If the value is less than or equal to the specified value, the consistency enhancement transmission mode will be exited, and the interface quantity exchange method under the normal communication cycle will be restored.

[0109] It should be noted that the electromagnetic transient side will be the first At each communication moment, the electromagnetic side consistency interface quantity is sent to the electromechanical transient side. The electromagnetic side consistency interface quantity includes at least the consistency interface voltage phasor. Consistency Interface Current Phasor Positive sequence voltage, negative sequence voltage, zero sequence voltage, positive sequence current, negative sequence current, zero sequence current, active power and reactive power After the electromechanical transient side reads the consistency interface quantity from the electromagnetic side, it updates the interface boundary state by combining the interface boundary dynamic equivalent model and the boundary dynamic memory state variables at the current communication moment, and generates the first... The interface boundary voltage phasor of the electromechanical side feedback at each communication moment and interface boundary current phasor .

[0110] To ensure consistency in the interface transmission between the electromechanical transient side and the electromagnetic transient side, the consistency of the electromagnetic side's interface quantities and the electromechanical side's feedback boundary quantities is further verified. Voltage consistency deviation at each communication moment Current consistency deviation and power consistency deviation They are represented as follows: ; ; ; in, Indicates the first Voltage consistency deviation at each communication moment Indicates the first Current consistency deviation at each communication moment Indicates the first Power consistency deviation at each communication moment The interface boundary current phasor representing the electromechanical side feedback .

[0111] Furthermore, the first Overall consistency deviation at each communication moment Represented as: ; in, Indicates the first The maximum consistency deviation in the three dimensions of voltage, current, and power at a given communication moment.

[0112] when At that time, the judgment of the first At any given moment, the interface transmission between the electromechanical transient side and the electromagnetic transient side satisfies the consistency condition. The electromechanical transient side transmits the interface boundary voltage phasor fed back from the electromechanical side. The signal is sent to the electromagnetic transient side, which uses the interface boundary voltage phasor fed back from the electromechanical side. Continue performing instantaneous electromagnetic calculations for the next communication cycle.

[0113] when At that time, the judgment of the first Inconsistencies exist in the interface transmission between the electromechanical transient side and the electromagnetic transient side at certain communication moments, triggering interface consistency correction. Interface consistency correction includes: based on the interface boundary voltage phasor fed back from the electromechanical side. The consistent interface voltage phasor obtained in step S4 Make corrections, or based on the interface boundary current phasor of the electromechanical side feedback. For consistent interface current phasors The corrected interface values ​​are then used for consistency checks in the current communication cycle or for joint simulation in the next communication cycle.

[0114] In the event of fault disturbances, control switching, sudden changes in three-phase imbalance, or overall consistency deviations When the consistency deviation increases continuously, the enhanced consistency delivery mode is activated. In this mode, the frequency of interface consistency checks is increased, or the interface quantity exchange cycle is shortened, and the number of boundary state updates within the same communication cycle is increased until the overall consistency deviation is accounted for. Restore to the allowed threshold Within.

[0115] It should be noted that a hybrid simulation data interaction interface oriented towards high frequency and high load is established to achieve stable data exchange and joint simulation between the electromechanical transient side and the electromagnetic transient side. This step mainly includes three parts. First, with high accuracy and stability as the goal, multiple simulation models are combined based on FMI (Functional Mock-up Interface) to realize model exchange and electromechanical-electromagnetic joint simulation. Data output from simulation software such as PowerFactory is realized based on OPC (Industrial Communication Protocol), enabling the transmission of network equivalent results on the electromechanical side, interface quantities on the electromagnetic side, and related boundary variables within a unified interface framework. Second, with real-time performance as the goal, a data communication protocol is studied. UDP (User Datagram Protocol) is used to realize information exchange between the hardware platform and the simulation software. Start bits and stop bits are added as markers at the start and end of each character sent to ensure that the receiving end can recognize character boundaries and maintain synchronous simulation between the sender and receiver. Third, for hybrid simulation scenarios involving single-port and multi-port interfaces, the simulation errors under different port locations are studied and analyzed, and stability analysis is performed on the corresponding situations to ensure that the joint simulation still has good accuracy and robustness when the interface deployment location changes.

[0116] The electromechanical transient side uses the dynamic equivalent boundary as the interface input, while the electromagnetic transient side uses the interface node voltage, interface node current, active power, reactive power, and three-sequence components as the interface output. The electromechanical and electromagnetic sides exchange data cyclically according to a preset communication cycle. The electromechanical side updates the boundary state based on the received interface quantities, while the electromagnetic side continues to perform instantaneous electromagnetic calculations based on the interface node voltage fed back from the electromechanical side, thus forming a continuous closed-loop electromechanical-electromagnetic joint simulation process. To adapt to the real-time transmission requirements under high-frequency and high-load conditions, a fixed-format message is preferred to encapsulate the interface variables. The message includes at least the interface number, timestamp, step number, and key electrical quantities at the interface, so that the sending and receiving ends can complete data alignment and state updates on the same time base.

[0117] During the co-simulation process, the results of three-phase voltage, three-phase current, positive-sequence voltage, negative-sequence voltage, zero-sequence voltage, active power, reactive power, and interface errors at the interface bus are continuously recorded to evaluate the accuracy and stability of the hybrid simulation. For single-port scenarios, the focus is on analyzing the consistency of voltage and current responses at the interface boundaries. For multi-port scenarios, the focus is on analyzing the impact of the coupling relationships between interfaces on the simulation results.

[0118] like Figure 4The control group A shown in the figure only uses the fundamental frequency static Norton equivalent scheme, without frequency-dependent network equivalent fitting, dynamic equivalent modeling, and bidirectional consistency verification and feedback correction. Control group B uses frequency-dependent network equivalent fitting, dynamic equivalent modeling, and communication cycle discretization, but without bidirectional consistency verification and feedback correction. Experimental group C represents the complete scheme of this invention, simultaneously employing frequency-dependent network equivalent fitting, dynamic equivalent modeling, communication cycle discretization, fundamental frequency extraction, interface quantity consistency recovery, and bidirectional consistency verification and feedback correction, and initiating a consistency enhancement transmission mode when necessary. In the figure, lighter colors indicate smaller changes in the admittance matrix, suggesting smoother and more continuous updates of interface boundary parameters according to the communication cycle. As can be seen from the figure, experimental group C has a lighter overall color, indicating that this invention can effectively improve the continuity and temporal coordination of interface boundary representation.

[0119] Figure 5 This invention was used to compare the peak differences and decay processes of injected current changes during the initial stage of a fault under different schemes. Control group A used only the fundamental frequency static Norton equivalent, without frequency-dependent network equivalent fitting, dynamic equivalent modeling, or bidirectional consistency verification and feedback correction. Control group B used frequency-dependent network equivalent fitting, dynamic equivalent modeling, and communication cycle discretization, but without bidirectional consistency verification and feedback correction. Experimental group C represented the complete scheme of this invention, simultaneously employing frequency-dependent network equivalent fitting, dynamic equivalent modeling, communication cycle discretization, fundamental frequency extraction, interface quantity consistency recovery, and bidirectional consistency verification and feedback correction, and activating a consistency enhancement transmission mode when necessary. As shown in the figure, experimental group C exhibited a lower peak value and faster decay of injected current changes, indicating that this invention not only reduces the abrupt change in interface injected current under fault disturbances but also improves the stability of bidirectional consistency transmission of interface quantities between the electromechanical transient side and the electromagnetic transient side.

[0120] This embodiment also provides a hybrid electromechanical and electromagnetic simulation system for ultra-large-scale power systems, including: a data construction module, used to collect basic data of the ultra-large-scale power system to be simulated, and to unify the time base and interface number of all data to form a hybrid simulation database.

[0121] The network equivalent module is used to divide the ultra-large-scale power system into electromechanical transient side partitions and electromagnetic transient side partitions based on basic data, set up interface buses, establish the node admittance matrix of the electromechanical side network, eliminate non-interface nodes, obtain the equivalent admittance matrix and equivalent current at the interface, and construct the Norton equivalent relationship at the interface.

[0122] The dynamic fitting module is used to perform frequency correlation network equivalent fitting on Norton equivalent relationships based on frequency scanning data from the electromechanical side, and outputs the interface equivalent admittance matrix and interface equivalent injection current vector under the current communication cycle through state space modeling and discretization.

[0123] The parameter calculation module is used to calculate the consistent interface voltage phasor, consistent interface current phasor, three-sequence components, active power, and reactive power of the current communication cycle based on the interface equivalent admittance matrix and the interface equivalent injected current column vector.

[0124] The bidirectional interaction module is used to establish a bidirectional consistent data interaction interface between the electromechanical transient side and the electromagnetic transient side, enabling the electromechanical transient side and the electromagnetic transient side to exchange interface quantities.

[0125] In summary, this invention achieves the effect of outputting and updating interface boundary parameters according to the communication cycle by performing frequency-related network equivalent fitting, dynamic equivalent modeling, and communication cycle discretization on the electromechanical transient side interface boundary, thereby ensuring the continuity and temporal coordination of the interface boundary representation during hybrid simulation. Furthermore, by extracting the fundamental frequency of the three-phase instantaneous quantities on the electromagnetic transient side, restoring interface quantity consistency, and performing bidirectional consistency verification and feedback correction, the invention achieves bidirectional consistency transmission of interface quantities between the electromechanical transient side and the electromagnetic transient side, ensuring consistency and stability of interface interactions during the co-simulation of ultra-large-scale power systems.

[0126] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A hybrid electromechanical and electromagnetic simulation method for ultra-large-scale power systems, characterized in that, include: Collect basic data of the ultra-large-scale power system to be simulated, and unify the time base and interface number of all data to form a hybrid simulation database; Based on the basic data, the ultra-large-scale power system is divided into electromechanical transient side partition and electromagnetic transient side partition, and an interface bus is set between the two. The node admittance matrix of the electromechanical side network is established, and non-interface nodes are eliminated to obtain the equivalent admittance matrix and equivalent current at the interface. Norton equivalent relationship at the interface is constructed. Based on the frequency scanning data of the electromechanical side, the Norton equivalent relationship is fitted by the frequency correlation network, and the interface equivalent admittance matrix and interface equivalent injection current vector under the current communication cycle are output through state space modeling and discretization. Based on the interface equivalent admittance matrix and the interface equivalent injected current column vector, calculate the consistent interface voltage phasor, consistent interface current phasor, three-sequence components, active power and reactive power for the current communication cycle; Establish a bidirectional consistent data exchange interface between the electromechanical transient side and the electromagnetic transient side, enabling the electromechanical transient side to exchange interface quantities with the electromagnetic transient side.

2. The electromechanical and electromagnetic hybrid simulation method for ultra-large-scale power systems as described in claim 1, characterized in that, The basic data includes: the topology connection relationship between the main network and the substation, bus parameters, line parameters, transformer parameters, generator parameters, load parameters, protection and control parameters, interface bus location, interface branch impedance, system rated frequency, partition boundary information, three-phase instantaneous voltage and three-phase instantaneous current at the interface bus, frequency scan data of electromechanical side admittance matrix, and interface message timestamp, message sequence number, transmission period and packet loss flag on the communication side.

3. The electromechanical and electromagnetic hybrid simulation method for ultra-large-scale power systems as described in claim 1, characterized in that, The Norton equivalent relationship at the interface is constructed as follows: in a single-port scenario with only one interface bus, a current source is injected at the interface location, and the injected current at the remaining nodes is set to zero. The interface node voltage is obtained by solving the relationship between the network node voltage and the injected current corresponding to the node admittance matrix, and the equivalent admittance at the interface is obtained based on the interface node voltage.

4. The electromechanical and electromagnetic hybrid simulation method for ultra-large-scale power systems as described in claim 1, characterized in that, The Norton equivalent relationship at the interface construction also includes: in the scenario where there are multiple hybrid simulation interface buses, all interface nodes are treated as a set of coupling ports and the node admittance matrix is ​​divided into blocks according to non-interface nodes and interface nodes, and the variables of non-interface nodes are eliminated to obtain the interface equivalent admittance matrix and the equivalent current converted from non-interface nodes to interface nodes. By combining the injected current and voltage at the interface node, a Norton-form relationship between the interface voltage and current is constructed.

5. The electromechanical and electromagnetic hybrid simulation method for ultra-large-scale power systems as described in claim 1 or 4, characterized in that, The frequency correlation network equivalent fitting of the Norton equivalent relationship at the interface based on the frequency scanning data of the electromechanical side includes: performing frequency scanning on the electromechanical side network outside the interface based on the determined set of interface nodes and node order to obtain node admittance matrix samples at discrete frequency points; The vector fitting method is used to parametrically fit each admittance element in the node admittance matrix sample, and each admittance element is expressed as a rational function to construct an equivalent model of the frequency correlation network.

6. The electromechanical and electromagnetic hybrid simulation method for ultra-large-scale power systems as described in claim 5, characterized in that, The frequency-related network equivalent model is converted into a state-space form, and boundary dynamic memory state variables are added to establish a dynamic equivalent model of the electromechanical side interface boundary. The dynamic equivalent model is discretized using the implicit trapezoidal integral method, and the discretized interface boundary relationships are organized into Norton-type boundaries under the current communication cycle to output the interface equivalent admittance matrix and the interface equivalent injection current vector of the current communication cycle.

7. The electromechanical and electromagnetic hybrid simulation method for ultra-large-scale power systems as described in claim 1, characterized in that, The calculation of the consistent interface voltage phasor, consistent interface current phasor, three-sequence components, active power and reactive power in the current communication cycle includes: recovering the three-phase voltage phasor and three-phase current phasor using the three-phase instantaneous voltage and three-phase instantaneous current at the interface, and calculating the active power and reactive power in the current communication cycle; Read the interface equivalent injected current vector of the current communication cycle, combine it with the recovered active power and reactive power, perform consistency recovery on the interface quantities, and obtain the consistent interface voltage phasor of the current communication cycle; The consistent interface current phasor for the current communication cycle is calculated based on the consistent interface voltage phasor, the interface equivalent admittance matrix, and the interface equivalent injected current vector.

8. The electromechanical and electromagnetic hybrid simulation method for ultra-large-scale power systems as described in claim 7, characterized in that, The establishment of a bidirectional consistency data interaction interface between the electromechanical transient side and the electromagnetic transient side includes: sending the consistency interface voltage phasor, consistency interface current phasor, three-sequence components, active power and reactive power of the current communication cycle output by the electromagnetic transient side to the electromechanical transient side. After reading the interface quantities from the electromechanical transient side, the interface boundary state is updated by combining the dynamic equivalent model of the interface boundary, and the interface boundary voltage phasor and interface boundary current phasor fed back by the electromechanical side in the current communication cycle are generated. The electromagnetic transient side output and the electromechanical transient side feedback are exchanged correspondingly under the same interface number, the same time base, and the same communication cycle.

9. The electromechanical and electromagnetic hybrid simulation method for ultra-large-scale power systems as described in claim 8, characterized in that, The establishment of a bidirectional consistency data interaction interface between the electromechanical transient side and the electromagnetic transient side also includes: calculating voltage consistency deviation, current consistency deviation, and power consistency deviation based on the consistency interface quantity output by the electromagnetic transient side and the interface boundary quantity fed back by the electromechanical side, and determining the comprehensive consistency deviation. When the overall consistency deviation meets the preset conditions, the joint simulation will proceed to the next communication cycle. When the overall consistency deviation does not meet the preset conditions, interface consistency correction is performed; In the event of fault disturbance, control switching, sudden change in three-phase imbalance, or continuous increase in overall consistency deviation, the consistency enhancement transmission mode is activated.

10. A hybrid electromechanical and electromagnetic simulation system for ultra-large-scale power systems, based on the hybrid electromechanical and electromagnetic simulation method for ultra-large-scale power systems according to any one of claims 1 to 9, characterized in that, include: The data construction module is used to collect basic data of the ultra-large-scale power system to be simulated, and to unify the time base and interface number of all data to form a hybrid simulation database. The network equivalence module is used to divide the ultra-large-scale power system into electromechanical transient side partitions and electromagnetic transient side partitions based on basic data, set up interface buses, establish the node admittance matrix of the electromechanical side network, eliminate non-interface nodes, obtain the equivalent admittance matrix and equivalent current at the interface, and construct the Norton equivalence relation at the interface. The dynamic fitting module is used to perform frequency correlation network equivalent fitting on Norton equivalent relations based on frequency scanning data from the electromechanical side, and output the interface equivalent admittance matrix and interface equivalent injection current vector under the current communication cycle through state space modeling and discretization. The parameter calculation module is used to calculate the consistent interface voltage phasor, consistent interface current phasor, three-sequence components, active power and reactive power of the current communication cycle based on the interface equivalent admittance matrix and the interface equivalent injected current column vector. The bidirectional interaction module is used to establish a bidirectional consistent data interaction interface between the electromechanical transient side and the electromagnetic transient side, enabling the electromechanical transient side and the electromagnetic transient side to exchange interface quantities.

Images