Electromagnetic transient simulation method of photovoltaic multi-converter system

Abstract

The invention discloses an electromagnetic transient simulation method for a photovoltaic multi-converter system. The method comprises the following steps: establishing an electromagnetic transient model comprising a photovoltaic array, a direct current side capacitor, an inverter bridge arm, a filtering unit and a power grid connection module; an interpolation algorithm is dynamically selected according to a simulation scene, a Newton interpolation method is adopted for step length switching and control delay compensation, and a Lagrange interpolation method is adopted for multi-rate subsystem data interaction; implementing a three-level variable step size control strategy based on physical characteristics; a damping interpolation discretization method is adopted to suppress numerical oscillation; and integrating an extreme scene adaptive mechanism. According to the method, a scene-algorithm-precision adaptive matching framework is constructed, Newton interpolation and Lagrange interpolation algorithms are dynamically selected through quantitative criteria, and multi-time-scale co-simulation, damping adaptive suppression and an extreme scene adaptation mechanism are combined, so that the simulation performance is comprehensively improved; the problems that an existing simulation method is low in efficiency, poor in precision and weak in adaptability in a large-scale photovoltaic system are solved.

Description

Technical Field

[0001] This invention relates to the field of power system simulation technology, specifically to an electromagnetic transient simulation method for a photovoltaic multi-converter system. Background Technology

[0002] With the rapid development of photovoltaic power generation technology and the continuous expansion of photovoltaic power plant scale, the photovoltaic multi-converter system, as the core power generation unit, faces unprecedented challenges in its simulation analysis.

[0003] Currently, traditional electromagnetic transient simulation methods face an inherent contradiction between simulation efficiency and accuracy when dealing with large-scale systems containing dozens to hundreds of parallel inverters. To accurately capture the rapid switching processes of power electronic devices, fixed simulation step sizes at the microsecond level or even smaller must be used, resulting in single-condition simulations of large-scale systems taking up to several hours, severely restricting the efficiency of engineering design optimization. Secondly, there is a significant problem of numerical oscillation, as the non-physical oscillation amplitude caused by switching actions typically reaches over 15%, seriously affecting the accuracy of key indicators such as harmonic analysis and overvoltage assessment, and reducing the reliability of simulation results. Furthermore, the application of existing interpolation algorithms exhibits fragmentation, because while Newton interpolation is computationally efficient, its accuracy is limited, while Lagrange interpolation is highly accurate but computationally complex. This means that current technologies have failed to achieve complementary advantages between the two algorithms based on the characteristics of the simulation scenario. In addition, there is the issue of tight model coupling, making system decomposition difficult. There are strong electromagnetic interactions between inverters and between inverters and the grid. Traditional methods solve the system as a whole, failing to adopt differentiated simulation strategies for parts with different dynamic characteristics.

[0004] In summary, we propose an electromagnetic transient simulation method for photovoltaic multi-converter systems. Summary of the Invention

[0005] The purpose of this invention is to provide an electromagnetic transient simulation method for photovoltaic multi-converter systems, which has the advantages of balancing simulation efficiency, accuracy and adaptability, and solves the problems of low efficiency, poor accuracy and weak adaptability in existing electromagnetic transient simulation technologies for photovoltaic multi-converter systems.

[0006] To achieve the above objectives, the present invention provides the following technical solution: an electromagnetic transient simulation method for a photovoltaic multi-converter system, comprising: establishing an electromagnetic transient model including a photovoltaic array, DC-side capacitors, inverter bridge arms, filter units, and grid connection modules; dynamically selecting an interpolation algorithm according to the simulation scenario, wherein Newton interpolation is used for step size switching and control delay compensation, and Lagrange interpolation is used for multi-rate subsystem data interaction; implementing a three-level variable step size control strategy based on physical characteristics; using a damped interpolation discretization method to suppress numerical oscillations; and integrating an extreme scenario adaptive mechanism.

[0007] Preferably, the selection criterion for the adaptive interpolation algorithm is as follows: when the dynamic rate of change of the system exceeds a set threshold and control delay compensation or parameter drift compensation is required, the Newton interpolation method with a computational complexity of O(n) is selected; when performing multi-rate subsystem data interaction, frequency adaptive filtering or fractional delay approximation, the Lagrange interpolation method with an interpolation order of 2-3 is selected.

[0008] Preferably, the specific implementation of the three-level variable step size control strategy includes: in the steady-state stage, when the DC voltage change rate du / dt < 50V / ms and the AC current change rate di / dt < 100A / ms, a first step size of 50-100μs is used; in the transient stage, when 50V / ms ≤ du / dt < 100V / ms or 100A / ms ≤ di / dt < 200A / ms, a second step size of 10-20μs is used; in the fault or switching stage, when du / dt ≥ 100V / ms or di / dt ≥ 200A / ms, a third step size of 1-5μs is used; during the step size switching process, Newton interpolation is used to reconstruct historical data, and the interpolation error threshold is set to 0.5%.

[0009] Preferably, the specific implementation of the multi-rate subsystem division and interpolation interaction is as follows: the system is divided into a fast subsystem with a step size of 1-5 μs, a medium-speed subsystem with a step size of 50-100 μs, and a slow subsystem with a step size of 1-10 ms; the subsystem interface data interaction is realized by using an interpolation formula based on the Lagrange basis function, wherein the synchronization period between the fast subsystem and the medium-speed subsystem is 10 fast steps, and the synchronization period between the medium-speed subsystem and the slow subsystem is 20 medium-speed steps.

[0010] Preferably, the specific implementation of the damping interpolation discretization method includes: constructing a quadratic interpolation function to describe the trajectory of variable change using data from the first three simulation steps; enabling damping interpolation within three steps before and after the switching event, with the damping factor adaptively adjusted according to the oscillation amplitude, taking a value of 0.03-0.05 under normal operating conditions and 0.08-0.1 under fault conditions; and reconstructing a smooth waveform through interpolation in the post-simulation processing stage to ensure that the proportion of the oscillation component to the real signal does not exceed 2%.

[0011] Preferably, it also includes a control delay compensation strategy: the inverter control delay parameters are obtained in the 2.5-1000Hz frequency band by using an improved vector matching method based on the Loewner matrix, and the parameters range from 0.5 to 2μs; the control delay is predicted and compensated based on the Newton interpolation method, and the delay-free control signal is reconstructed, with a compensation error not exceeding 5‰.

[0012] Preferably, the extreme scenario adaptation mechanism includes: switching to a 1μs step size and enabling high-damping interpolation within 0.3-0.5ms in a three-phase short-circuit scenario; automatically increasing the Lagrange interpolation order to the 3rd order in a voltage drop scenario; and compensating for the parameter drift of the IGBT on-resistance in real time through Newton interpolation in a temperature fluctuation scenario.

[0013] Preferably, it also includes model equivalence verification based on interpolation: setting five verification nodes: DC bus voltage, AC output current, power factor, switching frequency and output power; comparing the output of the detailed model and the equivalent model through real-time interpolation, and automatically switching to the detailed model calculation when the relative error of any node exceeds 3%.

[0014] Preferably, it also includes interpolation-assisted parallel computing optimization: the fast subsystem is allocated to 4-8 computing cores, the medium-speed subsystem is allocated to 2-4 computing cores, and the slow subsystem runs on a single core; a shared memory data area is established, the data copying overhead between processes is reduced through interpolation algorithms, and an asynchronous interpolation strategy is adopted to ensure that the utilization rate of each core is maintained at 70%-90%.

[0015] Preferably, the method is applicable to centralized inverter, string inverter and micro inverter architectures; the core module provides standardized implementation code, including step size switching logic, interpolation data interaction and damping suppression algorithm, and is compatible with MATLAB / Simulink and PSCAD / EMTDC simulation platforms.

[0016] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0017] This invention fully leverages the advantages of different interpolation algorithms through a scene-adaptive interpolation algorithm selection mechanism, significantly improving simulation efficiency while ensuring computational accuracy.

[0018] This invention employs a three-level variable step size control strategy based on physical characteristics to achieve intelligent adjustment of the simulation step size, effectively balancing the allocation of computing resources under different operating conditions.

[0019] This invention suppresses numerical oscillations at the algorithm level by using a damped interpolation discretization method, thereby enhancing the accuracy and reliability of simulation results.

[0020] This invention achieves decoupled analysis of dynamic processes at different time scales through a multi-rate subsystem co-simulation architecture, thereby improving model flexibility and reusability.

[0021] This invention ensures the simulation stability of the method under complex working conditions such as faults and disturbances through an extreme scenario adaptive mechanism.

[0022] This invention lowers the technical implementation threshold by standardizing the implementation scheme, making it easier for engineers to quickly apply and further develop the technology.

[0023] This invention comprehensively covers different architectures such as centralized, string, and micro inverters to meet the simulation needs of diverse photovoltaic systems.

[0024] The model equivalence verification mechanism based on interpolation in this invention ensures the accuracy of simplified models and supports efficient analysis of large-scale systems. Attached Figure Description

[0025] Figure 1 This is a flowchart illustrating the overall implementation of the present invention;

[0026] Figure 2 This is a schematic diagram of the selection logic for the adaptive interpolation algorithm of the present invention;

[0027] Figure 3 This is the state transition diagram of the three-level variable step size control strategy of the present invention;

[0028] Figure 4 This is a schematic diagram illustrating the working principle of the extreme scenario adaptation mechanism of this invention. Detailed Implementation

[0029] An electromagnetic transient simulation method for a photovoltaic multi-converter system includes: establishing an electromagnetic transient model that includes a photovoltaic array, DC-side capacitors, inverter bridge arms, filter units, and grid connection modules;

[0030] The interpolation algorithm is dynamically selected based on the simulation scenario. Newton interpolation is used for step size switching and control delay compensation, while Lagrange interpolation is used for data interaction of multi-rate subsystems.

[0031] Implement a three-level variable step size control strategy based on physical characteristics;

[0032] A damped interpolation discretization method is used to suppress numerical oscillations;

[0033] Integrated extreme scenario adaptive mechanism.

[0034] This technical solution: System modeling method

[0035] A detailed electromagnetic transient model is established, including the photovoltaic array, DC-side capacitors, inverter arms, LCL filters, and grid connection units. Appropriate modeling strategies are adopted for photovoltaic systems of different scales.

[0036] For photovoltaic power plants of 100 MW and above, a hierarchical clustering equivalent modeling method is adopted. Inverters with similar operating states are grouped into equivalent groups, and their external characteristics are characterized by Thevenin equivalent circuits. Equivalent parameters are obtained by fitting measured data from multiple typical operating conditions using the least squares method to ensure that the equivalent error is controlled within a reasonable range.

[0037] Adaptive interpolation selection mechanism

[0038] Based on feature quantization criteria from the simulation scenario, the most suitable interpolation algorithm is dynamically selected:

[0039] For scenarios with high computational efficiency requirements, such as step size switching, control delay compensation, and parameter drift compensation, the Newton interpolation method with a computational complexity of O(n) is selected.

[0040] For scenarios requiring high accuracy, such as multi-rate subsystem data interaction, frequency adaptive filtering, and fractional delay approximation, the Lagrange interpolation method with an interpolation order of 2-3 is selected.

[0041] Three-level variable step size control strategy

[0042] Based on the quantitative evaluation of the system's dynamic characteristics, intelligent adjustment of the simulation step size is achieved:

[0043] Steady-state phase (system rate of change <5% / ms): Historical data are reconstructed using Newton interpolation with a step size of 100μs.

[0044] Transient phase (system change rate 5%-10% / ms): Use a 10μs step size to balance accuracy and efficiency;

[0045] Fault / Switching Phase (System Change Rate > 10% / ms): Employs a 1μs step size to accurately capture high-frequency dynamic processes;

[0046] The step size switching threshold is dynamically adjusted based on interpolation error estimation to ensure numerical stability during the switching process.

[0047] Damped interpolation discretization method

[0048] A three-stage suppression strategy is adopted, consisting of "quadratic interpolation trajectory fitting + multi-factor damping control + post-processing smoothing".

[0049] The damping factor is adaptively adjusted according to the oscillation amplitude, and the calculation formula is as follows:

[0050]

[0051] in, The damping factor, This represents the oscillation amplitude (the difference between the current sampled value and the steady-state value). The value is the nominal value of the variable.

[0052] Multi-platform adaptation

[0053] The core algorithm offers multiple standardized implementations, including S-functions in MATLAB / Simulink and custom modules in PSCAD / EMTDC, ensuring consistency and usability of the method in different simulation environments.

[0054] The present invention will now be described in detail with reference to the accompanying drawings and embodiments. These embodiments are intended to fully illustrate the technical details and implementation methods of the present invention and should not be construed as limiting the scope of protection of the present invention.

[0055] Example 1: Simulation Analysis of a 100MW Photovoltaic Power Plant

[0056] System configuration and parameter settings

[0057] Using a 100MW photovoltaic power plant as the simulation object, the system contains 200 500kW string inverters. Through operational state clustering analysis, the 200 inverters are aggregated into 10 equivalent groups. Thevenin parameters of the equivalent groups are obtained by least squares fitting:

[0058]

[0059] in, This represents the total equivalent resistance of the entire inverter group (e.g., 20 units). This represents the total equivalent inductance of the entire inverter group. Let be the active power of the i-th inverter. Equivalent resistance Reactive power It is the equivalent inductance.

[0060] The main simulation parameters are configured as shown in the table below:

[0061] Parameter categories Parameter name Parameter value Remark System parameters DC bus voltage 1500V - AC side voltage 35kV - Switching frequency 15kHz - Simulation parameters steady-state step size 100μs du / dt < 50 V / ms and di / dt < 100 A / ms Transient step size 10μs 50 V / ms ≤ du / dt < 100 V / ms or 100 A / ms ≤ di / dt < 200 A / ms Fault step 1μs du / dt≥100V / ms or di / dt≥200A / ms interpolation parameters Number of Newton interpolation nodes 3 - Lagrange interpolation order 2 - Interpolation error threshold 0.5% -

[0062] Adaptive interpolation selection implementation

[0063] The selection of the adaptive interpolation algorithm is based on the scene type identifier and the quantitative criteria of the system's dynamic rate of change, and is achieved through the following decision logic:

[0064] Step 1: Initial selection based on scenario type

[0065] When the scenario type is step size switching, control delay compensation or parameter drift compensation, the Newton interpolation method with a computational complexity of O(n) should be given priority, where n is the number of interpolation nodes.

[0066] Step 2: Precision-based secondary selection

[0067] When the scenario involves multi-rate subsystem data interaction, frequency adaptive filtering, or fractional delay approximation, the Lagrange interpolation method with an interpolation order of 2-3 is preferred.

[0068] Step 3: Final decision based on dynamic rate of change

[0069] For other scenarios or situations where the characteristics of the current scenario are unclear, a quantitative decision based on the system's dynamic rate of change is adopted:

[0070] When the dynamic rate of change of the system δ>0.1 (i.e., the rate of change exceeds 10% / ms), Newton interpolation method is selected to prioritize computational efficiency.

[0071] When the dynamic rate of change of the system δ≤0.1, the Lagrange interpolation method is selected to prioritize interpolation accuracy.

[0072] The system's dynamic rate of change δ is defined as follows:

[0073]

[0074] In the formula, This is the change in voltage. This is the voltage reference value. The change in current This is the current reference value.

[0075] Adaptive adjustment of damping factor

[0076] The damping factor in the damping interpolation discretization method adopts an adaptive adjustment strategy based on the oscillation amplitude and is determined through the following mathematical relationship:

[0077] Damping factor calculation formula:

[0078]

[0079] in, is the damping factor, dimensionless, with a value range of [0.01, 0.1]. The base damping coefficient is set to 0.02 to ensure minimal damping effect. The amplitude sensitivity coefficient, with a value of 0.08, controls the rate of change of damping with the oscillation amplitude. This represents the oscillation amplitude (the difference between the current sampled value and the steady-state value). These are the rated values ​​for variables, determined based on the specific physical quantities (such as rated voltage and rated current). It is a minimum value function to ensure that the ratio does not exceed 1.0 and to avoid excessive damping.

[0080] Constraints during implementation:

[0081] The damping factor is ultimately limited to a value between 0.01 and 0.1.

[0082] Damping interpolation is forcibly enabled within three simulation steps before and after the switching event occurs.

[0083] Under normal operating conditions, the damping factor ranges from 0.03 to 0.05.

[0084] The damping factor ranges from 0.08 to 0.1 under fault or transient conditions.

[0085] Simulation Result Analysis

[0086] The key performance indicators of a 100MW photovoltaic power plant were simulated and analyzed using the method of this invention:

[0087] Single-condition simulation time: 6.8 minutes.

[0088] DC bus voltage steady-state error: 1.2%.

[0089] AC output current THD: 3.3%.

[0090] Oscillation amplitude during switch operation: 1.1%.

[0091] Oscillation decay time: 1.2ms.

[0092] Simulation results show that the method of the present invention significantly improves computational efficiency and effectively suppresses numerical oscillations while ensuring simulation accuracy.

[0093] Example 2: Simulation of Three-Phase Short Circuit Fault Scenarios

[0094] Fault settings and simulation parameters

[0095] Based on Example 1, a three-phase short-circuit fault is set to occur on the 35kV side at t=2s, with a peak short-circuit current of 12kA, and the fault is cleared after 200ms.

[0096] To address the specific requirements of the fault scenario, the simulation parameters were adjusted as follows:

[0097] Fault detection threshold: di / dt ≥ 120 A / ms

[0098] Step size switching delay: <0.3ms

[0099] Fault step size: 1μs

[0100] Damping factor: 0.09

[0101] Data synchronization cycle: 5 fast steps

[0102] Fault Scenario Adaptation Mechanism

[0103] Simulation adaptation after a fault occurs is achieved through the following process:

[0104] Real-time monitoring of the rate of change of system state variables;

[0105] When the rate of change is detected to exceed the fault threshold, step size switching is immediately triggered;

[0106] High-damping interpolation is enabled to suppress numerical oscillations during fault transients;

[0107] Adjust the subsystem synchronization frequency to ensure real-time data interaction;

[0108] After the fault is cleared, the simulation parameters are gradually restored to normal.

[0109] Simulation effect evaluation

[0110] In a three-phase short-circuit fault scenario, the method of the present invention exhibits the following advantages:

[0111] Simulation error of peak short-circuit current: 1.9%.

[0112] Voltage recovery time after fault clearance: 28ms.

[0113] Oscillation amplitude during the fault: 1.8%.

[0114] Numerical stability during simulation: good, with no divergence.

[0115] Example 3: Simulation of a microinverter system

[0116] System characteristics and modeling considerations

[0117] Micro-inverter systems differ significantly from large centralized systems and require special consideration in modeling and simulation:

[0118] High-frequency switching characteristics: The switching frequency is typically 50-100kHz, which is much higher than that of centralized inverters;

[0119] Distributed control: Communication delay has a significant impact on system dynamics;

[0120] Topology: Complex topologies such as Buck-Boost + full bridge are often used.

[0121] Dedicated parameter configuration

[0122] Specific simulation parameters are set to suit the characteristics of microinverters:

[0123] Parameter categories Parameter name Parameter value Applicable Scenarios Topology parameters DC side capacitor 470μF Power buffer LCL filter L1 400μH High frequency filtering LCL filter C 1.5μF Resonance suppression LCL filter L2 150μH Grid-side filtering Control parameters Current loop proportional gain 0.8 Dynamic response Current loop integral gain 50 steady-state accuracy Voltage loop proportional gain 0.2 stability Voltage loop integral gain 10 Adjustment accuracy Simulation parameters Switch delay compensation 150ns Newton interpolation Communication delay modeling 3ms Lagrange interpolation

[0124] Simulation performance

[0125] In a simulation of a system of 100 3kW microinverters, the method of this invention exhibits the following performance characteristics:

[0126] AC output current THD: 3.0%.

[0127] MPPT tracking accuracy: 99.2%.

[0128] Switch oscillation amplitude: 1.3%.

[0129] Simulation time for a single inverter: 1.5 seconds.

[0130] Technological advantages are reflected

[0131] Through the implementation and verification of the above embodiments, the present invention demonstrates multiple technical advantages:

[0132] Simulation accuracy assurance

[0133] The selection of adaptive interpolation algorithms based on scene characteristics ensures that the most suitable numerical method is used in different simulation stages. The multi-rate subsystem co-simulation architecture effectively decouples dynamic processes at different time scales, avoiding mutual interference. The damped interpolation discretization method suppresses numerical oscillations at the algorithmic level, guaranteeing the physical realism of the simulation results.

[0134] Computational efficiency optimization

[0135] A three-level variable step size control strategy enables intelligent allocation of simulation resources, maximizing computational efficiency while ensuring accuracy. Interpolation-assisted parallel computing optimization fully leverages the computing power of multi-core processors. A model equivalence verification mechanism significantly reduces computational complexity while ensuring accuracy.

[0136] Enhanced engineering applicability

[0137] The standardized implementation scheme supports mainstream simulation platforms, lowering the barrier to technology implementation. An adaptive mechanism for extreme scenarios ensures the reliability of the method under complex operating conditions. It comprehensively covers different inverter architectures, meeting diverse engineering needs.

[0138] Promotion and application value

[0139] This invention provides an efficient and reliable simulation tool for the design optimization, fault analysis, and control strategy verification of large-capacity photovoltaic power plants, which helps to promote the further development and application of photovoltaic power generation technology. The modular design and standardized interface of the method also facilitate subsequent functional expansion and performance upgrades.

[0140] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit the scope of protection of the present invention. 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 essence and scope of the technical solutions of the present invention.

Claims

1. A method of electromagnetic transient simulation of a photovoltaic multiconverter system, characterized in that, The application relates to a self-adaptive interpolation algorithm for electromagnetic transient simulation of photovoltaic grid-connected systems. The selection criterion of the self-adaptive interpolation algorithm is as follows: when the dynamic change rate of the system exceeds a set threshold value, control delay compensation or parameter drift compensation is needed, the Newton interpolation method with a calculation complexity of O(n) is selected; when multi-rate subsystem data interaction, frequency adaptive filtering or fractional order delay approximation are performed, the Lagrange interpolation method with an interpolation order of 2-3 orders is selected.

2. The method of electromagnetic transient simulation of a photovoltaic multivariable converter system according to claim 1, wherein, The specific implementation of the three-level variable step length control strategy includes the following: in a steady state stage, when the DC voltage change rate du / dt is less than 50 V / ms and the AC current change rate di / dt is less than 100 A / ms, a first step length of 50-100 mu s is adopted; in a transient stage, when 50 V / ms<=du / dt<100 V / ms or 100 A / ms<=di / dt<200 A / ms, a second step length of 10-20 mu s is adopted; in a fault or switching stage, when du / dt is greater than or equal to 100 V / ms or di / dt is greater than or equal to 200 A / ms, a third step length of 1-5 mu s is adopted; in the step length switching process, the Newton interpolation method is adopted to reconstruct the historical data, and the interpolation error threshold value is set to 0.5%.

3. The method of claim 1, wherein, The specific implementation of the multi-rate subsystem division and interpolation interaction is as follows: the system is divided into a fast subsystem with a step length of 1-5 mu s, a medium-speed subsystem with a step length of 50-100 mu s and a slow subsystem with a step length of 1-10 ms; an interpolation formula based on a Lagrange basis function is adopted to realize the data interaction of the subsystem interface, wherein the synchronization period of the fast subsystem and the medium-speed subsystem is 10 fast step lengths, and the synchronization period of the medium-speed subsystem and the slow subsystem is 20 medium-speed step lengths.

4. The method of claim 1, wherein, The specific implementation of the damping interpolation discretization method includes the following: the data of the first three simulation step lengths are adopted to construct a quadratic interpolation function to describe the variable change trajectory; damping interpolation is enabled within 3 step lengths before and after a switching event, and the damping factor is adaptively adjusted according to the oscillation amplitude, and the value is 0.03-0.05 under normal working conditions and 0.08-0.1 under fault working conditions; in a simulation post-processing stage, the waveform is reconstructed through interpolation to ensure that the proportion of the oscillation component in the real signal is not more than 2%.

5. The method of electromagnetic transient simulation of a photovoltaic multivariable converter system of claim 1, wherein, The control delay compensation strategy includes the following: an improved vector matching method based on a Loewner matrix is adopted to obtain the inverter control delay parameters in a frequency band of 2.5-1000 Hz, and the parameter range is 0.5-2 mu s; the control delay is predicted and compensated based on the Newton interpolation method, the non-delay control signal is reconstructed, and the compensation error is not more than 5 ‰.

6. The method of electromagnetic transient simulation of a photovoltaic multivariable converter system of claim 1, wherein, ​ 7. The method of electromagnetic transient simulation of a photovoltaic multivariable converter system of claim 1, wherein, The extreme scenario adaptation mechanism includes: switching to 1us step size and enabling high damping interpolation within 0.3-0.5ms under three-phase short-circuit scenario; automatically increasing the Lagrange interpolation order to 3 under voltage drop scenario; compensating for the parameter drift of IGBT on-resistance in real time through Newton interpolation under temperature fluctuation scenario.

8. The method of electromagnetic transient simulation of a photovoltaic multivariable converter system of claim 1, wherein, It also includes interpolation-based model equivalent verification: five verification nodes are set, including DC bus voltage, AC output current, power factor, switching frequency and output power; the outputs of the detailed model and the equivalent model are compared through real-time interpolation, and when the relative error of any node exceeds 3%, the detailed model calculation is automatically switched to.

9. The method of electromagnetic transient simulation of a photovoltaic multivariable converter system of claim 1, wherein, It also includes interpolation-assisted parallel computing optimization: fast subsystem is allocated to 4-8 computing cores, medium-speed subsystem is allocated to 2-4 computing cores, and slow-speed subsystem runs on a single core; a shared memory data area is established, data copying overhead between processes is reduced through interpolation algorithm, and asynchronous interpolation strategy is adopted to ensure that the utilization rate of each core is maintained at 70%-90%.

10. The method of electromagnetic transient simulation of a photovoltaic multivariable converter system of claim 1, wherein, The method is applicable to centralized inverters, string inverters and micro-inverter architectures; the core module provides standardized implementation code, including step size switching logic, interpolation data interaction and damping suppression algorithm, and is suitable for MATLAB / Simulink, PSCAD / EMTDC simulation platforms.

Images