Switching transient modeling method and power electronic converter simulation method applying the same

Abstract

The application discloses a switching transient modeling method and a power electronic converter simulation method using the same. The switching transient modeling method comprises the following steps: A1, establishing a switching transient modeling object; the switching transient modeling object adopts a half-bridge circuit; A2, offline measuring the switching transient voltage and current of the half-bridge circuit; A3, dividing the turn-on and turn-off processes of the half-bridge circuit into three sections respectively, and using the three sections to describe the switching transient voltage and current and the change relationship between the switching transient voltage and current and the working condition. The method is simple in steps, reasonable in design, and convenient to realize. Through the device-level real-time simulation modeling method of the power electronic converter based on adaptive curve fitting, the transient curve waveform fitted can be adaptively adjusted according to the change of the working condition, the real-time simulation step length can be realized to be as low as 25 ns, the real-time simulation result can be highly consistent with the offline simulation result, the effect is remarkable, and the method is convenient to popularize.

Description

Technical Field

[0001] This invention belongs to the field of power electronics simulation technology, specifically relating to a switching transient modeling method and a power electronic converter simulation method using the same method. Background Technology

[0002] Real-time simulation technology plays a crucial role in the R&D cycle of power electronic converters. By employing hardware-in-the-loop (HIL) testing based on real-time simulation to construct a semi-physical test environment, comprehensive verification of the controller can be performed even without a real power electronic converter, enabling a parallel development process for both the converter and the controller. Simultaneously, HIL allows for rapid iteration of controller design, safe and worry-free debugging of control code, exploration of operational boundaries, testing of extreme performance, and simulation of fault conditions. This provides a fast, safe, and efficient digital verification environment for converter R&D, effectively improving the overall development efficiency.

[0003] The form of the real-time simulation model directly determines the functionality that HIL testing can achieve. Considering the characteristics of power electronic converters, such as high switching frequency and strong nonlinearity, real-time simulation models of converters are mainly classified into the following categories based on the power switching model used:

[0004] The first type is the equivalent averaging model based on switching cycle states. This model averages the voltage and current changes caused within a switching cycle, treating the power electronic converter as a linear model to describe its static behavior. This type of model is mainly used for real-time simulation of large-scale power networks, where the converter is only a functional component in the modeling. Because it ignores switching events, it cannot be used for HIL testing of power electronic converters.

[0005] The second category is real-time simulation models of converter systems, constructed using ideal switches, binary resistor switch models, and ADC switch models. The ideal switch model treats turning on as a short circuit and turning off as a short circuit; the binary resistor model treats turning on as a small resistance (10... -5 ~10 -3 Ω), the turn-off is equivalent to a large resistance (10 Ω), 3 ~10 6 (Ω); The ADC model treats turn-on as a small inductor and turn-off as a capacitor. The system-level real-time simulation model retains the description of switching events but ignores the behavior of the power semiconductor devices themselves, such as switching transients, device stress, switching losses, EMI, etc. Therefore, HIL can only verify closed-loop control functions and cannot provide any information other than steady-state voltage and current.

[0006] Due to the strict limitations on model computation time in real-time simulation, equivalent average models and system-level real-time simulation models have long been the main technical approaches for achieving real-time simulation of power electronics due to their simple structure and ease of solution. For example, binary resistor models and ADC models are widely used in current commercial real-time simulation systems, such as those from companies like RTDS, Opal-RT, Typhoon HIL, and Yuankuan Energy. With the development of FPGA simulation technology, device-level real-time simulation of power electronics has become a research hotspot in this field in recent years. Device-level real-time simulation uses transient behavior models of power semiconductors to simulate switching transient current and voltage waveforms, making real-time simulation more realistic. It can simulate non-ideal switching processes of converters, switching losses, device stress, EMI, and electrothermal coupling characteristics. This further expands the testing capabilities of HIL, enabling the testing of optimization control strategies, device protection strategies, and thermal management strategies based on device stress, converter efficiency, and electromagnetic interference suppression. However, device-level real-time simulation faces stricter computation time constraints: on the one hand, switching transient processes are rapid, and to accurately describe the switching transient waveforms, the step size of real-time simulation needs to be as low as nanoseconds; on the other hand, transient models often have complex model structures and nonlinear characteristics, making them difficult to solve and time-consuming. Therefore, the main applications of device-level real-time simulation are currently high-power converters with long switching transient processes, such as power conversion based on MMC converters, high-voltage DC circuit breakers, and high-speed train traction systems.

[0007] Currently, the main device-level real-time simulation models used both domestically and internationally include: 1) nonlinear equivalent circuit models; 2) piecewise linear transient models; 3) quasi-transient models; and 4) curve fitting models. Nonlinear equivalent circuit models refer to directly using nonlinear equivalent circuit models such as IGBTs, MOSFETs, and diodes in real-time simulation. While this method maximizes the consistency between real-time and offline models, the nonlinearity and discontinuities of the model lead to significant computational overhead, limiting simulation speed. It is generally only used for microsecond-level switching transient simulations. Piecewise linear transient models linearize the nonlinear equivalent circuit model piecewise and use a discrete event-based triggering mechanism to handle model discontinuities, reducing simulation latency to 50ns in a highly parallel manner. However, this approach consumes significant FPGA computing resources, limiting the topology size of the simulated object. Quasi-transient models consider the differences in time scale between electrical network behavior and switching transient behavior, decoupling system behavior simulation from switching behavior simulation using a two-level simulation structure. Datasheet-driven models, high-resolution models, and Hammerstein models are representative examples. This two-stage simulation method can simulate switching transients with a minimum accuracy of 5ns. However, the switching transient model is not involved in the simulation of the electrical network, meaning that the simulation accuracy of the converter is still at the system level.

[0008] Curve fitting models describe the switching transient process by fitting transient waveforms obtained from offline measurements and then proportionally adjusting them according to steady-state voltage and current in real-time simulations. Unlike nonlinear equivalent circuit models and piecewise linear models, curve fitting models do not involve numerical solutions to the switching equivalent circuit model, thus requiring less computation and allowing for smaller real-time simulation step sizes. Compared to quasi-transient models, curve fitting models can apply transient behavior to the simulation of electrical networks, thereby improving the accuracy of converter real-time simulation models. However, because curve fitting models use switching transient waveforms obtained from offline measurements at specific operating points, while the switching transient process varies with operating conditions, curve fitting models have poor versatility and cannot adapt to simulations with a wide range of operating conditions. A dynamic curve fitting model has been proposed, where the rise and fall curves of the transient process are described using RL and RC circuits. However, the switching transient process has high-order characteristics, and the accuracy is insufficient when using first-order circuits such as RL and RC. Furthermore, the simulation step size of this model is 500 ns, which cannot accurately describe the switching transient process when simulating most converters.

[0009] This shows that real-time simulation at the device level of power electronic converters faces a contradiction between simulation accuracy and simulation speed, while curve fitting models that can balance model accuracy and simulation speed have the problem of poor versatility. Summary of the Invention

[0010] The technical problem to be solved by this invention is to address the shortcomings of the prior art by providing a switching transient modeling method and a power electronic converter simulation method using the same method. The method is simple in steps, reasonable in design, and easy to implement. Through the device-level real-time simulation modeling method of power electronic converter based on adaptive curve fitting, the fitted transient curve waveform can be adaptively adjusted according to changes in operating conditions. It can achieve a real-time simulation step size of at least 25ns, and the real-time simulation results can maintain a high degree of consistency with the offline simulation results. The effect is significant and easy to promote.

[0011] To solve the above-mentioned technical problems, the technical solution adopted by the present invention is: a switching transient modeling method, comprising the following steps:

[0012] Step A1: Establish a switching transient modeling object; the switching transient modeling object adopts a half-bridge circuit;

[0013] Step A2: Measure the switching transient voltage and current of the half-bridge circuit offline;

[0014] Step A3: Divide the turn-on and turn-off processes of the half-bridge circuit into three segments to describe the transient voltage and current of the switch, as well as the relationship between the transient voltage and current of the switch and the operating conditions.

[0015] In the aforementioned switching transient modeling method, the half-bridge circuit in step A1 includes two IGBT diode commutation units, and commutation in a single switching event occurs only in one of the commutation units; the input voltage V of the half-bridge circuit... CC and input current I L These values ​​are constants in each simulation step.

[0016] In the above-described switching transient modeling method, the offline measurement of the half-bridge circuit's switching transient voltage under rated voltage and current in step A2 is... and current Represented as:

[0017]

[0018] Among them, f i (t) represents the offline measured IGBT current as a function of time, f v (t) represents the IGBT voltage measured offline as a function of time.

[0019] The aforementioned switching transient modeling method, in step A3, divides the turn-on and turn-off processes of the half-bridge circuit into three segments, specifically including:

[0020] First turn-on stage: The gate voltage of the IGBT is lower than the turn-on threshold, and the IGBT remains in a stable off state. The IGBT current i c The voltage v is zero. ceEqual to the steady-state turn-off value;

[0021] Phase 2 activation: IGBT current i c The steady-state input current I L As the direction increases, the diode remains forward biased, and the voltage across the diode is clamped; during this stage, the IGBT current i c With offline measured switching transient current Increase at the same rate until exceeding the peak current I L +I rr I rr The maximum reverse recovery current of the diode, i is the current i of the IGBT. c Represented as:

[0022] i c =f i (t)

[0023] IGBT voltage v ce During this stage, the voltage of the IGBT will drop slightly due to stray inductance. ce Able to pass through the current i of the IGBT c The same method is used to obtain the measurement curve, and then the value is proportionally reduced and assigned. The voltage v of the IGBT is... ce Represented as:

[0024] v ce =K v *f v (t)

[0025] Among them, K v To reduce the scale, and V ce (off) represents the steady-state turn-off voltage of the IGBT before it is turned on. This represents the steady-state turn-off voltage of the IGBT obtained through offline measurement.

[0026] The third stage of turn-on: The reverse recovery current of the diode reaches its peak and begins to decay, the diode is reverse biased, and the voltage across the IGBT drops sharply until it reaches the steady-state turn-on voltage; then the curve amplitude is adjusted according to the coefficient. and K v The curve is scaled and shifted to the left to seamlessly connect with the curve of the second turn-on stage, in which the IGBT current i c Represented as:

[0027]

[0028] in, The peak reverse recovery current of the diode obtained from offline measurement, t rrThe time it takes for the diode's reverse recovery current to reach its peak value. This refers to the time it takes for the diode reverse recovery current to reach its peak value, measured offline. The steady-state input current of the half-bridge circuit is measured offline.

[0029] The voltage v of the IGBT at this stage ce Represented as:

[0030]

[0031] First stage of shutdown: IGBT voltage v ce As the voltage rises, the diode remains reverse biased, and the IGBT current i c It drops slightly, and then the curve adjusts according to the proportion of the turn-off delay time. Scaling the time axis, the IGBT current i c After scaling the curve on the time axis, it is then adjusted according to the ratio K. i Its amplitude is scaled, and the current i of the IGBT in this stage is... c Represented as:

[0032]

[0033] in, I c(on) This refers to the steady-state turn-on current of the IGBT before it is turned off. The steady-state turn-on current of the IGBT is obtained from offline measurement. t d (off) represents the IGBT turn-off delay time. This refers to the IGBT turn-off delay time obtained from offline measurements.

[0034] The voltage v of the IGBT at this stage ce Represented as:

[0035]

[0036] Second stage of shutdown: When the IGBT voltage v ce Exceeding the steady-state voltage value v ce(off) When the diode is forward biased, the IGBT current i c The voltage V of the IGBT will drop sharply during this stage. ce It continues to follow the growth trend of the first stage of shutdown until the voltage peak value (V) is reached. sp +v ce(off) The current i of the IGBT c The curve is first scaled M on the time axis. tf Multiply by 1, then scale its magnitude by K. iThe current i of the IGBT during this stage is times that of the IGBT. c Represented as:

[0037]

[0038] Among them, t f0 This represents the initial moment when the IGBT current begins to decrease. This represents the initial moment of the IGBT current decrease obtained from offline measurements. t f For IGBT current fall time, The IGBT current fall time is obtained from offline measurement;

[0039] Third stage of shutdown: During this stage, the voltage v of the IGBT is... ce It can be equivalent to the voltage spike and steady-state voltage value v. ce(off) The superposition of these elements is represented as:

[0040]

[0041] Among them, t sp The moment when the IGBT turn-off voltage reaches its peak. This refers to the moment when the IGBT turn-off voltage reaches its peak, as measured offline. This is the steady-state turn-off voltage of the IGBT obtained through offline measurement.

[0042] This invention also discloses a power electronic converter simulation method, which applies the above-mentioned switching transient modeling method. The simulation method is a real-time simulation method, and the specific steps include:

[0043] Step B1: Read the simulation results from the previous step;

[0044] Step B2: Obtain the input voltage V of the current simulation step and the half-bridge circuit. CC and input current I L This achieves decoupling of the half-bridge circuit;

[0045] Step B3: Determine the IGBT diode commutation unit currently operating in the half-bridge circuit based on the direction of the input current;

[0046] Step B4: Determine the switching states of the IGBT and diode in the IGBT-diode commutation unit; the switching states include the on-state steady state, the off-state steady state, the on-state transient state, and the off-state transient state;

[0047] Step B5: When the IGBT and diode are in the turn-on steady state and turn-off steady state, the steady-state model is used to solve the half-bridge circuit to obtain the lower diode voltage and the upper bridge arm current; when the IGBT and diode are in the turn-on transient state and turn-off transient state, the switching transient waveform is solved using the aforementioned switching transient modeling method to obtain the lower diode voltage and the upper bridge arm current.

[0048] Step B6: Using the lower transistor voltage and upper bridge arm current as inputs, solve the remaining network;

[0049] Step B7: End the current simulation step and wait for the next simulation to begin.

[0050] In the power electronic converter simulation method described above, step B2 involves obtaining the input voltage V of the current simulation step and the half-bridge circuit. CC and input current I L Decoupling of a half-bridge circuit can be achieved using transmission line modeling, voltage-current coupling, or explicit integration.

[0051] Compared with the prior art, the present invention has the following advantages:

[0052] 1. The method of the present invention has simple steps, reasonable design, and is easy to implement.

[0053] 2. The power semiconductor adaptive curve fitting transient model proposed in this invention can solve the problem of insufficient simulation accuracy of traditional curve fitting models under different voltage and current conditions, and realize the simulation of power semiconductor switching transient characteristics of curve fitting models in a wide range of operating conditions.

[0054] 3. The real-time simulation method based on the adaptive curve fitting model proposed in this invention can resolve the contradiction between the large computational load of device-level real-time simulation and the short required simulation step size, and realize the real-time transient simulation of converter devices with a step size as low as 25ns.

[0055] 4. This invention can be effectively applied in power electronics simulation. The real-time simulation results can maintain a high degree of consistency with the offline simulation results, with significant effects and easy promotion.

[0056] In summary, the method of this invention is simple in steps, reasonable in design, and easy to implement. Through the real-time simulation modeling method of power electronic converter device level based on adaptive curve fitting, the fitted transient curve waveform can be adaptively adjusted according to changes in operating conditions. It can achieve a real-time simulation step size of as low as 25ns, and the real-time simulation results can maintain a high degree of consistency with the offline simulation results. The effect is significant and easy to promote.

[0057] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description

[0058] Figure 1 This is a flowchart of the switching transient modeling method of the present invention;

[0059] Figure 2 This is a transient waveform diagram of the inductive switch of the IGBT diode commutation unit of the present invention;

[0060] Figure 3 This is a flowchart of the power electronic converter simulation method of the present invention;

[0061] Figure 4 This is a comparison chart of the transient simulation results of the present invention;

[0062] Figure 5 This is a comparison chart of switching time and energy loss during the transient process of this invention;

[0063] Figure 6 This is a model diagram of the two-phase interleaved parallel Boost converter of the present invention;

[0064] Figure 7 This is a comparison chart of FPGA simulation results for the two-phase interleaved parallel Boost converter of this invention;

[0065] Figure 8 This is a model diagram of the three-phase two-level inverter of the present invention;

[0066] Figure 9 This is a comparison chart of FPGA simulation results for the three-phase two-level inverter of this invention. Detailed Implementation

[0067] like Figure 1 and Figure 2 As shown, the switching transient modeling method of the present invention includes the following steps:

[0068] Step A1: Establish a switching transient modeling object; the switching transient modeling object adopts a half-bridge circuit;

[0069] Step A2: Measure the switching transient voltage and current of the half-bridge circuit offline;

[0070] Step A3: Divide the turn-on and turn-off processes of the half-bridge circuit into three segments to describe the transient voltage and current of the switch, as well as the relationship between the transient voltage and current of the switch and the operating conditions.

[0071] In this embodiment, the half-bridge circuit in step A1 includes two IGBT diode commutation units, and commutation in a single switching event occurs only in one of the commutation units; the input voltage V of the half-bridge circuit... CC and input current I L These values ​​are constants in each simulation step.

[0072] In this embodiment, the offline measurement of the half-bridge circuit in step A2 measures the switching transient voltage under rated voltage and current. and current Represented as:

[0073]

[0074] Among them, f i (t) represents the offline measured IGBT current as a function of time, f v (t) represents the IGBT voltage measured offline as a function of time.

[0075] In this embodiment, as Figure 2 As shown, step A3, which divides the turn-on and turn-off processes of the half-bridge circuit into three segments, specifically includes:

[0076] First turn-on stage: The gate voltage of the IGBT is lower than the turn-on threshold, and the IGBT remains in a stable off state. The IGBT current i c The voltage v is zero. ce Equal to the steady-state turn-off value;

[0077] Phase 2 activation: IGBT current i c The steady-state input current I L As the direction increases, the diode remains forward biased, and the voltage across the diode is clamped; during this stage, the IGBT current i c With offline measured switching transient current Increase at the same rate until exceeding the peak current I L +I rr I rr The maximum reverse recovery current of the diode, i is the current i of the IGBT. c Represented as:

[0078] i c =f i (t)

[0079] IGBT voltage v ce During this stage, the voltage of the IGBT will drop slightly due to stray inductance. ce Able to pass through the current i of the IGBT c The same method is used to obtain the measurement curve, and then the value is proportionally reduced and assigned. The voltage v of the IGBT is... ce Represented as:

[0080] v ce =K v *f v (t)

[0081] Among them, K v To reduce the scale, and V ce (off) represents the steady-state turn-off voltage of the IGBT before it is turned on. This represents the steady-state turn-off voltage of the IGBT obtained through offline measurement.

[0082] The third stage of turn-on: The reverse recovery current of the diode reaches its peak and begins to decay, the diode is reverse biased, and the voltage across the IGBT drops sharply until it reaches the steady-state turn-on voltage; then the curve amplitude is adjusted according to the coefficient. and K v The curve is scaled and shifted to the left to seamlessly connect with the curve of the second turn-on stage, in which the IGBT current i c Represented as:

[0083]

[0084] in, The peak reverse recovery current of the diode obtained from offline measurement, t rr The time it takes for the diode's reverse recovery current to reach its peak value. This refers to the time it takes for the diode reverse recovery current to reach its peak value, measured offline. The steady-state input current of the half-bridge circuit is measured offline.

[0085] The voltage v of the IGBT at this stage ce Represented as:

[0086]

[0087] First stage of shutdown: IGBT voltage v ce As the voltage rises, the diode remains reverse biased, and the IGBT current i c It drops slightly, and then the curve adjusts according to the proportion of the turn-off delay time. Scaling the time axis, the IGBT current i c After scaling the curve on the time axis, it is then adjusted according to the ratio K. i Its amplitude is scaled, and the current i of the IGBT in this stage is... c Represented as:

[0088]

[0089] in, I c(on) This refers to the steady-state turn-on current of the IGBT before it is turned off. The steady-state turn-on current of the IGBT is obtained from offline measurement. t d (off) represents the IGBT turn-off delay time. This refers to the IGBT turn-off delay time obtained from offline measurements.

[0090] The voltage v of the IGBT at this stage ce Represented as:

[0091]

[0092] Second stage of shutdown: When the IGBT voltage v ce Exceeding the steady-state voltage value v ce(off) When the diode is forward biased, the IGBT current i c The voltage V of the IGBT will drop sharply during this stage. ce It continues to follow the growth trend of the first stage of shutdown until the voltage peak value (V) is reached. sp +v ce(off) The current i of the IGBT c The curve is first scaled M on the time axis. tf Multiply by 1, then scale its magnitude by K. i The current i of the IGBT during this stage is times that of the IGBT. c Represented as:

[0093]

[0094] Among them, t f0 This represents the initial moment when the IGBT current begins to decrease. This represents the initial moment of the IGBT current decrease obtained from offline measurements. t f For IGBT current fall time, The IGBT current fall time is obtained from offline measurement;

[0095] Third stage of shutdown: During this stage, the voltage v of the IGBT is... ce It can be equivalent to the voltage spike and steady-state voltage value v. ce(off) The superposition of these elements is represented as:

[0096]

[0097] Among them, t sp The moment when the IGBT turn-off voltage reaches its peak. This refers to the moment when the IGBT turn-off voltage reaches its peak, as measured offline. This is the steady-state turn-off voltage of the IGBT obtained through offline measurement.

[0098] In practical implementation, the parameters involved in the switching transient modeling method of this invention are all relatively easy to obtain. The required parameters include a set of switching voltage and current waveforms measured at rated values, curves of switching time and steady-state current, and curves of diode reverse recovery peak current and steady-state current; these parameters can all be obtained from device datasheets. Therefore, the switching transient modeling method proposed in this invention has strong versatility and is easy to promote.

[0099] like Figure 3 As shown, the power electronic converter simulation method of the present invention is a real-time simulation method, and the specific steps include:

[0100] Step B1: Read the simulation results from the previous step;

[0101] Step B2: Obtain the input voltage V of the current simulation step and the half-bridge circuit. CC and input current I L This achieves decoupling of the half-bridge circuit;

[0102] Step B3: Determine the IGBT diode commutation unit currently operating in the half-bridge circuit based on the direction of the input current;

[0103] Step B4: Determine the switching states of the IGBT and diode in the IGBT-diode commutation unit; the switching states include the on-state steady state, the off-state steady state, the on-state transient state, and the off-state transient state;

[0104] Step B5: When the IGBT and diode are in the turn-on steady state and turn-off steady state, the steady-state model is used to solve the half-bridge circuit to obtain the lower diode voltage and the upper bridge arm current; when the IGBT and diode are in the turn-on transient state and turn-off transient state, the switching transient waveform is solved using the aforementioned switching transient modeling method to obtain the lower diode voltage and the upper bridge arm current.

[0105] Step B6: Using the lower transistor voltage and upper bridge arm current as inputs, solve the remaining network;

[0106] Step B7: End the current simulation step and wait for the next simulation to begin.

[0107] In this embodiment, step B2 involves obtaining the input voltage V of the current simulation step and the half-bridge circuit. CC and input current I L Decoupling of a half-bridge circuit can be achieved using transmission line modeling, voltage-current coupling, or explicit integration.

[0108] To verify the technical effectiveness of the power electronic converter simulation method of the present invention, a verification experiment was conducted.

[0109] Taking the FF450R12ME4 module as an example, an adaptive curve fitting model was established. The control model was an IGBT diode model built using the Model Architect tool in the Saber offline simulation software. The offline measured switching waveforms and switching times were obtained from the simulation results of this control model.

[0110] The transient simulation accuracy of the adaptive curve fitting model was tested using different voltages and currents. The current range was 100A to 400A, and the switching waveforms were tested at 490V and 584V respectively. Figure 4 This is a set of FPGA-based transient simulation results for a 300A / 500V switch. The upper part shows the Saber simulation results, and the lower part shows the FPGA-based simulation results. Figure 4 As can be seen, the simulation results of the two methods are consistent.

[0111] Figure 5 (a) and (b) are the switching times of the transient waveforms of the adaptive curve fitting model tested at 490V and 584V, respectively, including the current rise time and voltage fall time of the turn-on transient, and the current fall time and voltage fall time of the turn-off transient. Figure 5 (c) and (d) show the switching losses of the adaptive curve fitting model tested at 490V and 584V, respectively, including turn-on energy loss and turn-off energy loss under different load currents. Figure 5 It can be seen that the adaptive curve fitting model is accurate over a wide operating range. Although errors can be observed at some operating points, considering that the adaptive curve fitting model uses a fixed time step of 25ns for simulation, while Saber uses a variable time step of 1ns, such errors are difficult to avoid in fixed-step real-time simulation. Based on the adaptive curve fitting model, while ensuring the efficiency of device-level real-time simulation, the model can maintain high accuracy over a wide operating range.

[0112] A real-time simulation modeling method based on adaptive curve fitting was used to perform real-time simulations of a two-phase interleaved parallel Boost converter and a three-phase inverter, respectively. The two-phase interleaved parallel Boost converter is shown below. Figure 6 As shown in Table 1, the parameters are as follows. It operates in open-loop control mode, with the two PWM channels 180° out of phase and both having a duty cycle of 0.4. The FPGA model uses 40-bit fixed-point numbers, and the final FPGA simulation step size is 25ns.

[0113] Table 1 Boost simulation parameters

[0114] enter 300V capacitance 1000μF inductance 400μH load 2.5Ω Switching frequency 20kHz Duty cycle 0.4

[0115] Figure 7This is a comparison chart of FPGA-based simulation results and Saber simulation results. Figure 7 It can be seen that the results of the two methods are highly consistent. By calculating the average error of each variable, the average error of the real-time simulation modeling method based on adaptive curve fitting is less than 0.5%.

[0116] Real-time FPGA simulation of a three-phase two-level inverter, with the three-phase inverter topology as follows: Figure 8 As shown in Table 2, the inverter uses SPWM technology for modulation and operates in open loop with an amplitude modulation ratio of 0.8. The FPGA model uses 40-bit fixed-point numbers, and the final FPGA simulation step size is 25ns.

[0117] Table 2 Inverter Simulation Parameters

[0118] enter 600V capacitance 1μF inductance 1μH load 1Ω carrier frequency 1kHz Dead Time 2μs

[0119] Figure 9 (a) Comparison of FPGA simulation results and Saber simulation results for three-phase current, from... Figure 9 It can be seen that the two are highly consistent, with an average error of less than 1%. By using R... L By changing the resistor value from 0.5Ω to 2Ω and testing the inverter's efficiency at different operating points with 0.25Ω intervals, output power ranging from 36kW to 100kW was obtained. Figure 9 (b) It can be seen that, based on steady-state characteristics, the error introduced by the real-time simulation modeling method based on adaptive curve fitting is still relatively small.

[0120] In summary, the real-time simulation modeling method based on adaptive curve fitting can achieve real-time transient simulation of power electronic converters with a step size as low as 25 ns.

[0121] The above description is merely a preferred embodiment of the present invention and does not constitute any limitation on the present invention. Any simple modifications, alterations, or equivalent structural changes made to the above embodiments based on the technical essence of the present invention shall still fall within the protection scope of the present invention.

Claims

1. A method for modeling switching transients, characterized in that, Includes the following steps: Step A1: Establish a switching transient modeling object; the switching transient modeling object adopts a half-bridge circuit; Step A2: Measure the switching transient voltage and current of the half-bridge circuit offline; Step A3: Divide the turn-on and turn-off processes of the half-bridge circuit into three segments to describe the transient voltage and current of the switch, as well as the relationship between the transient voltage and current of the switch and the operating conditions. The half-bridge circuit described in step A1 includes two IGBT diode commutation units, and commutation in a single switching event occurs only in one of the commutation units; the input voltage of the half-bridge circuit... and input current It is a constant value in each simulation step; The offline measurement of the half-bridge circuit in step A2 shows the switching transient voltage under rated voltage and current. and current Represented as: in, This represents the function of offline measured IGBT current versus time. A function representing the IGBT voltage versus time measured offline; Step A3 describes dividing the turn-on and turn-off processes of the half-bridge circuit into three segments, specifically including: First turn-on stage: The IGBT gate voltage is lower than the turn-on threshold, the IGBT remains in a stable off state, and the IGBT current... Zero, voltage Equal to the steady-state turn-off value; Phase Two Activation: IGBT Current Input current to steady state As the direction increases, the diode remains forward biased, and the voltage across the diode is clamped; this stage With offline measured switching transient current Increase at the same rate until the peak current is exceeded. , The maximum reverse recovery current of the diode, the Represented as: IGBT During this stage, the inductance will decrease slightly due to stray inductance, and the IGBT's... Able to communicate with The same method is used to obtain the measurement curve, and then the value is scaled down proportionally. This is how the IGBT's... Represented as: in, To reduce the scale, and , This represents the steady-state turn-off voltage of the IGBT before it is turned on. This represents the steady-state turn-off voltage of the IGBT obtained through offline measurement. The third stage of turn-on: The reverse recovery current of the diode reaches its peak and begins to decay, the diode is reverse biased, and the voltage across the IGBT drops sharply until it reaches the steady-state turn-on voltage; then the curve amplitude is adjusted according to the coefficient. and The curve is scaled and then shifted to the left to seamlessly connect with the curve in the second phase of opening. Represented as: in, , The peak reverse recovery current of the diode was obtained through offline measurement. The time it takes for the diode's reverse recovery current to reach its peak value. This refers to the time it takes for the diode reverse recovery current to reach its peak value, measured offline. The steady-state input current of the half-bridge circuit is measured offline. The IGBT described in this stage Represented as: ； Phase 1 shutdown: IGBT's As the voltage rises, the diode remains reverse biased. It drops slightly, and then the curve adjusts according to the proportion of the turn-off delay time. Scaling on the timeline After scaling the curve on the time axis, it is then adjusted according to the ratio. Its amplitude is scaled, as described in this stage Represented as: in, , This refers to the steady-state turn-on current of the IGBT before it is turned off. The steady-state turn-on current of the IGBT is obtained from offline measurement. , For IGBT turn-off delay time, This refers to the IGBT turn-off delay time obtained from offline measurements. The IGBT described in this stage Represented as: ； Second stage of shutdown: When the IGBT's Exceeding steady-state voltage value When the diode is in a forward bias state, the voltage across the diode will be in a forward bias state. The IGBT will decline sharply at this stage. It continues to follow the growth trend of the first stage of shutdown until the voltage peak value. , The curve is first scaled on the time axis. Multiply by 1, then scale its magnitude. Times, as described in this stage Represented as: in, This represents the initial moment when the IGBT current begins to decrease. This represents the initial moment of the IGBT current decrease obtained from offline measurements. , For IGBT current fall time, The IGBT current fall time is obtained from offline measurement; Third stage of shutdown: In this stage, the IGBT's It can be equivalent to voltage spikes and steady-state voltage values. The superposition of these elements is represented as: in, The moment when the IGBT turn-off voltage reaches its peak. This refers to the moment when the IGBT turn-off voltage reaches its peak, as measured offline. This is the steady-state turn-off voltage of the IGBT obtained through offline measurement.

2. A simulation method for a power electronic converter, characterized in that, The switching transient modeling method described in claim 1 is used, wherein the simulation method is a real-time simulation method, and the specific steps include: Step B1: Read the simulation results from the previous step; Step B2: Obtain the input voltage of the current simulation step and the half-bridge circuit. and input current This achieves decoupling of the half-bridge circuit; Step B3: Determine the IGBT diode commutation unit currently operating in the half-bridge circuit based on the direction of the input current; Step B4: Determine the switching states of the IGBT and diode in the IGBT-diode commutation unit; the switching states include the on-state steady state, the off-state steady state, the on-state transient state, and the off-state transient state; Step B5: When the IGBT and diode are in the turn-on steady state and turn-off steady state, the steady-state model is used to solve the half-bridge circuit to obtain the lower diode voltage and the upper bridge arm current; when the IGBT and diode are in the turn-on transient state and turn-off transient state, the switching transient waveform is solved using the aforementioned switching transient modeling method to obtain the lower diode voltage and the upper bridge arm current. Step B6: Using the lower transistor voltage and upper bridge arm current as inputs, solve the remaining network; Step B7: End the current simulation step and wait for the next simulation to begin.

3. The power electronic converter simulation method according to claim 2, characterized in that, Step B2 describes obtaining the input voltage of the current simulation step and the half-bridge circuit. and input current Decoupling of a half-bridge circuit can be achieved using transmission line modeling, voltage-current coupling, or explicit integration.

Images