Online Optimization Control Method and System for Soft Switching of Multi-Active Bridge Converters

Abstract

This application discloses a soft-switching online optimization control method and system for multiple active bridge converters. The method includes: calculating the high-frequency link pole current at two rising edges of the H-bridge output square wave voltage at each port based on a unified time-domain high-frequency link current analytical model and circuit parameters; determining the state of each port based on the high-frequency link pole current and selecting target ports whose DC voltage meets preset voltage conditions as time-domain optimization targets; determining the smaller of the high-frequency link pole currents at the two rising edges as the target pole current and adjusting the global common duty cycle based on the target pole current; dynamically adjusting the phase shift angle between ports using a decoupling matrix to compensate for power drops caused by changes in duty cycle; solving the problems of poor practicality and closed-loop instability caused by sudden changes in duty cycle, and possessing advantages such as low computational cost, high transmission efficiency, and strong scalability.

Description

Technical Field

[0001] This invention relates to the field of circuits, and more particularly to a soft-switching online optimization control method and system for multiple active bridge converters. Background Technology

[0002] With the widespread integration of DC power sources such as new energy sources and electric vehicles, multi-port isolated bidirectional DC power routers, as key equipment for efficient power transmission and distribution, are showing great development potential. Among them, multi-active bridge converters (MABs, such as quad active bridges (QABs)) replace traditional DC buses by expanding multiple H-bridge modules on high-frequency transformers (HFTs), achieving mutual isolation between different DC networks at multiple ports, bidirectional power transmission, and high energy density.

[0003] However, the efficient operation of the MAB is severely limited by the loss of zero-voltage soft switching. Especially under light load or port voltage mismatch, the junction capacitance of the switching device cannot fully discharge within the dead time. Under traditional phase-shift modulation (PSM), since there is only one set of control variables (such as phase shift angle), the system cannot actively control the high-frequency link current at the switching moment, which easily leads to the loss of ZVS, resulting in a sharp increase in switching losses and significantly reducing the operating efficiency of the MAB.

[0004] To extend the soft-switching range, existing technologies typically introduce duty cycles on top of phase-shift control, forming multiple degrees of freedom modulation (MDFM). However, with the increase in control degrees of freedom (e.g., QAB has 7 control degrees of freedom), the possible combinations of different control variables can reach thousands. Most existing soft-switching optimization strategies rely on offline optimization algorithms (such as particle swarm optimization) for global optimization and store the results in a lookup table of the digital controller for online retrieval. This "offline calculation + online lookup" approach has three significant bottlenecks:

[0005] 1) Time-consuming, labor-intensive, and narrow in coverage: The actual operating conditions are constantly changing, and discrete lookup tables cannot accurately cover all operating conditions.

[0006] 2) Poor closed-loop control stability: When the operating state crosses the lookup table interval, the duty cycle will change abruptly. This change will cause violent oscillations in port voltage and current, which will seriously affect the closed-loop control stability of MAB.

[0007] 3) Poor topology scalability: Due to the lack of a unified mathematical analytical model, once the number of ports in the MAB changes (for example, from four ports to five ports), it is necessary to re-model and pre-calculate in an extremely tedious way, which makes it very impractical for engineering. Summary of the Invention

[0008] The following is an overview of the topics described in detail in this article.

[0009] The purpose of this application is to at least partially solve one of the technical problems existing in the related technologies. The embodiments of this application provide a method and system for online optimization control of soft switching of multiple active bridge converters.

[0010] An embodiment of the first aspect of this application provides a soft-switching online optimization control method for a multi-active bridge converter, comprising:

[0011] Obtain the circuit parameters of each port of the multi-active bridge converter;

[0012] Based on the unified time-domain high-frequency link current analytical model, the high-frequency link pole current at the start and end rising edges of the H-bridge output square wave voltage at each port is calculated according to the circuit parameters.

[0013] The state of each port is determined based on the high-frequency chain pole current, and the target port whose DC voltage meets the preset voltage condition is selected as the time domain optimization target based on the state of each port.

[0014] The smaller of the high-frequency link pole currents at the start and end rising edges of the square wave voltage output by the target port H-bridge is determined as the target pole current, and the global common duty cycle is adjusted according to the target pole current.

[0015] The phase shift angle between ports is dynamically adjusted using a decoupling matrix to compensate for the power drop caused by changes in duty cycle.

[0016] According to certain embodiments of the first aspect of this application, the circuit parameters include DC voltage, inductance parameters, switching frequency, and target transmission power.

[0017] According to certain embodiments of the first aspect of this application, the unified time-domain high-frequency link current analytical model decomposes a multi-degree-of-freedom modulated square wave voltage with a duty cycle into two sets of sub-square waves with halved amplitude, different phases, and containing only phase-shift modulation, so as to decompose the converter under multi-degree-of-freedom modulation into two sets of sub-converters containing only phase-shift modulation. Then, the high-frequency link current pole values ​​at the corresponding time are obtained by solving the two sets of sub-square waves and superimposing them using a linear combination interpolation method.

[0018] According to certain embodiments of the first aspect of this application, the unified time-domain high-frequency link current analytical model is configured with a port number parameter, which can be changed to be applicable to multiple active bridge converters with any number of ports.

[0019] According to certain embodiments of the first aspect of this application, the step of determining the state of each port based on the high-frequency chain pole current and selecting target ports whose DC voltage meets preset voltage conditions as time-domain optimization targets based on the state of each port includes:

[0020] Determine whether the state of each port meets the zero-voltage soft-switching condition based on the high-frequency chain pole current.

[0021] Select the target port with the smallest DC voltage from the ports that have lost the zero-voltage soft-switching condition as the time-domain optimization target.

[0022] According to certain embodiments of the first aspect of this application, adjusting the global common duty cycle based on the target pole current includes:

[0023] When the target pole current is greater than 0, the global common duty cycle is smoothly increased.

[0024] When the target pole current is less than 0, the global common duty cycle is reduced according to the set safety margin.

[0025] When the voltages at each port are restored to matching and soft switching can be achieved under pure phase-shift modulation, the common duty cycle is smoothly adjusted to 0.

[0026] According to certain embodiments of the first aspect of this application, an upper limit constraint is imposed on the global common duty cycle during the process of adjusting the global common duty cycle based on the target pole current;

[0027] The upper limit of the global common duty cycle is expressed as: ;

[0028] In the formula, f is the upper limit of the global duty cycle. s L is the switching frequency. ij，Max For the maximum equivalent series inductance, I ref，Max For the maximum target average current value, V' Min The minimum port voltage after being reduced to the reference port, where n is the number of ports.

[0029] According to certain embodiments of the first aspect of this application, dynamically adjusting the phase shift angle between ports using a decoupling matrix includes:

[0030] The current error is obtained by comparing the average current setpoint at each port with the actual sampled value.

[0031] The phase shift compensation amount is obtained by multiplying the output of the proportional-integral regulator of the current error by the decoupling matrix;

[0032] The phase shift angle between ports is dynamically adjusted based on the phase shift angle compensation amount.

[0033] An embodiment of the second aspect of this application provides a soft-switching online optimization control system for a multi-active bridge converter, comprising: a main circuit of a multi-active bridge converter and a controller connected to the main circuit;

[0034] The main circuit of the multi-active bridge converter includes a high-frequency transformer and multiple fully controlled H-bridge arms connected to the winding ports of the high-frequency transformer.

[0035] The controller is used to generate a pulse width modulation drive signal according to the online optimization control method for soft switching of the multi-active bridge converter according to any one of claims 1 to 8, and to control the on and off of each switching device in the fully controlled H-bridge arm, so as to realize soft switching of the switching devices under light load or voltage mismatch conditions.

[0036] According to an embodiment of the second aspect of this application, the controller includes a soft-switching online optimization control module and a power decoupling module;

[0037] The soft-switching online optimization control module is used to calculate the high-frequency link pole current at the start and end rising edges of the square wave output voltage of the H-bridge at each port based on the unified time-domain high-frequency link current analytical model and the circuit parameters; determine the state of each port based on the high-frequency link pole current; select the target port whose DC voltage meets the preset voltage condition as the time-domain optimization target based on the state of each port; determine the smaller value between the high-frequency link pole current at the start and end rising edges of the square wave output voltage of the target port H-bridge as the target pole current; and adjust the global common duty cycle based on the target pole current.

[0038] The power decoupling module is used to dynamically adjust the phase shift angle between ports using a decoupling matrix to compensate for power drops caused by changes in duty cycle.

[0039] The above-mentioned solution has at least the following beneficial effects:

[0040] (1) The unified time-domain high-frequency link current analytical model of this application has a concise form, completely abandoning the huge offline lookup table, and only requires simple multiplication operations to calculate the pole current value. In conventional digital signal processors (such as DSP_F28335), the calculation time is greatly shortened and almost no additional computational burden is added;

[0041] (2) By using a unified time-domain high-frequency link current analytical model, the traditional model is freed from the limitations of specific topologies. There is no need to re-derive complex state equations. Only the number of ports n in the parameter formula needs to be changed. It can be directly backward compatible or upward extended to any multi-port MAB system, and has strong compatibility.

[0042] (3) The single-objective PI controller is used to achieve continuous and smooth adjustment of the duty cycle, avoiding the sudden change in duty cycle caused by the traditional lookup table method. When the system operating conditions meet the requirements of phase-shift modulation to achieve soft switching, the PI controller can smoothly transition the duty cycle to 0, realizing seamless switching between SDC-MDFM mode and PSM mode. Compared with the sudden change in duty cycle, the dynamic response stabilization time of the system is improved;

[0043] (4) Without increasing the system conduction loss, the method of this application reduces the conduction loss and can additionally realize zero-voltage soft switching or zero-current switching of at least 4 switching devices, which greatly reduces the overall switching loss of MAB under light load and voltage mismatch conditions. Attached Figure Description

[0044] The accompanying drawings are used to provide a further understanding of the technical solutions of this application and constitute a part of the specification. They are used together with the embodiments of this application to explain the technical solutions of this application and do not constitute a limitation on the technical solutions of this application.

[0045] Figure 1 A step diagram illustrating the online optimization control method for soft switching of a multi-active bridge converter provided in this application embodiment;

[0046] Figure 2 A sub-step diagram of step S300 provided in the embodiments of this application;

[0047] Figure 3 A diagram illustrating the sub-steps for adjusting the global common duty cycle based on the target pole current, provided in an embodiment of this application.

[0048] Figure 4 This is a diagram illustrating the sub-steps of dynamically adjusting the phase shift angle between ports using a decoupling matrix, as provided in an embodiment of this application.

[0049] Figure 5 This is a schematic diagram of the topology of a multi-active bridge converter provided in an embodiment of this application;

[0050] Figure 6 This is a schematic diagram of the H-bridge switching mechanism of a multi-active bridge converter under phase-shift modulation provided in an embodiment of the present invention;

[0051] Figure 7 A timing diagram showing the voltage waveform decomposition before and after the introduction of duty cycle in the H-bridge, provided in an embodiment of the present invention;

[0052] Figure 8 The overall system logic block diagram of the soft-switching online optimization control strategy provided in the embodiments of the present invention;

[0053] Figure 9 A curve showing the continuous change of load voltage during the smooth transition from a soft-switching online optimized control strategy to a phase-shift modulation mode, provided for embodiments of the present invention;

[0054] Figure 10 A curve showing the continuous change of duty cycle from a soft-switching online optimization control strategy to a phase-shifting modulation mode, provided for embodiments of the present invention;

[0055] Figure 11 The simulation diagram shows a comparison between the soft-switching online optimization control strategy provided in the embodiments of the present invention and the impact of sudden changes in duty cycle on the closed-loop stability of the system. Detailed Implementation

[0056] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0057] It should be noted that although functional modules are divided in the device schematic diagram and a logical order is shown in the flowchart, in some cases, the steps shown or described may be performed in a different order than the module division in the device or the order in the flowchart. The terms "first," "second," etc., in the specification, claims, or the aforementioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence.

[0058] To address the problems of high online computational burden, closed-loop instability caused by sudden duty cycle changes, and poor model scalability in existing MAB soft-switching optimization technologies, embodiments of this application provide an online optimization control method and system for soft switching of multiple active bridge converters. This method aims to restore soft-switching characteristics to ports that have lost zero-voltage soft-switching conditions by adjusting the global common duty cycle d in real time. It avoids the problems of poor practicality and closed-loop instability caused by sudden duty cycle changes, and has the advantages of low computational cost, high transmission efficiency, and strong scalability.

[0059] The embodiments of this application will be further described below with reference to the accompanying drawings.

[0060] A soft-switching online optimization control system for a multi-active bridge converter includes: a main circuit of the multi-active bridge converter and a controller connected to the main circuit.

[0061] The main circuit of the multi-active bridge converter includes a high-frequency transformer and multiple fully controlled H-bridge arms connected to the winding ports of the high-frequency transformer.

[0062] The controller is used to generate pulse width modulation drive signals according to the online optimization control method for soft switching of multi-active bridge converters, and control the conduction and turn-off of each switching device in the fully controlled H-bridge arm to achieve soft switching of the switching devices under light load or voltage mismatch conditions.

[0063] The controller includes a soft-switching online optimization control module and a power decoupling module.

[0064] The soft-switching online optimization control module is used to calculate the high-frequency link pole current at the start and end rising edges of the H-bridge output square wave voltage at each port based on a unified time-domain high-frequency link current analytical model and circuit parameters. It then determines the state of each port based on the high-frequency link pole current, and selects the target port whose DC voltage meets the preset voltage conditions as the time-domain optimization target. Finally, it determines the smaller of the high-frequency link pole current at the start and end rising edges of the target port's H-bridge output square wave voltage as the target pole current, and adjusts the global common duty cycle based on the target pole current.

[0065] The power decoupling module is used to dynamically adjust the phase shift angle between ports using a decoupling matrix to compensate for power drops caused by changes in duty cycle.

[0066] A multi-active bridge converter includes a high-frequency transformer (HFT) with multiple windings, and multiple fully controlled H-bridge arms connected to the ports of each winding. (See reference...) Figure 5 Taking a four-port active bridge converter as an example, it includes a high-frequency transformer with four windings and four fully controlled H-bridge arms connected to each winding port respectively.

[0067] Each port can be flexibly connected to a DC power supply, DC load, or energy storage battery according to energy interaction requirements, realizing bidirectional energy flow and electrical isolation. In the equivalent circuit model, the equivalent series inductance (ESL) and equivalent series resistance (ESR) of each port branch are taken into account. Among them, ESL is a key physical parameter that determines the transient waveform of high frequency chain current (HFCC) and the power interaction characteristics between ports. The controller collects the voltage and current information of each port in real time and generates a pulse width modulation (PWM) signal to drive the H-bridge switching transistors, thereby realizing soft switching operation.

[0068] Reference Figure 1 The online optimization control method for soft switching of multi-active bridge converters includes the following steps:

[0069] Step S100: Obtain the circuit parameters of each port of the multi-active bridge converter;

[0070] Step S200: Based on the unified time-domain high-frequency link current analytical model, calculate the high-frequency link pole current at the start and end rising edges of the H-bridge output square wave voltage at each port according to the circuit parameters.

[0071] Step S300: Determine the state of each port based on the high-frequency link pole current, and select the target port whose DC voltage meets the preset voltage condition as the time domain optimization target based on the state of each port.

[0072] Step S400: Determine the smaller of the high-frequency link pole current at the start rising edge and the end rising edge of the target port H-bridge output square wave voltage as the target pole current, and adjust the global common duty cycle according to the target pole current.

[0073] Step S500: The phase shift angle between ports is dynamically adjusted using a decoupling matrix to compensate for the power drop caused by the change in duty cycle.

[0074] For step S100, obtain the circuit parameters of each port of the multi-active bridge converter. The circuit parameters include the DC voltage V. i Inductance parameter L i Switching frequency f s and the target transmission power P of each port ref .

[0075] For step S200, based on the unified time-domain high-frequency link current analytical model, the high-frequency link pole current at the start and end rising edges of the H-bridge output square wave voltage at each port is calculated according to the circuit parameters.

[0076] The unified time-domain high-frequency link current analytical model is derived based on the improved discrete-time model (IDTM). It adopts a common duty cycle multi-degree-of-freedom modulation strategy and uses a PI controller to perform online smooth optimization of the duty cycle of each port H-bridge.

[0077] The unified time-domain high-frequency link current analytical model decomposes a multi-degree-of-freedom modulated square wave voltage with duty cycle into two sets of sub-square waves with halved amplitude, different phases, and containing only phase-shift modulation (e.g., Figure 7 The v shown h1,I and v h1,II The converter under multi-degree-of-freedom modulation is decomposed into two sets of sub-converters containing only phase-shift modulation. Then, the high-frequency link current pole values ​​at the corresponding time are obtained by solving the two sets of sub-square waves and superimposing them using the linear combination interpolation method.

[0078] Taking a four-port active bridge converter as an example, after introducing the same common duty cycle d, the pole current analytical expressions of each port at the beginning rising edge (0-time) and the end rising edge (0+time) of the square wave voltage 0 level are as follows:

[0079] ;

[0080] .

[0081] In the formula, A i The electrical coefficient of port i (related to the number of turns, inductance, and frequency). To normalize the equivalent series inductance, To calculate the port k voltage referred to the reference port, D jk Let be the phase shift angle between port j and port k.

[0082] The unified time-domain high-frequency link current analytical model has a port number parameter n. When the topology changes to n ports, it can be adapted to multiple active bridge converters with any number of ports by changing the port number parameter. That is, the upper limit of the summation sign in the analytical expression can be changed to n. There is no need to rebuild the complex state space matrix, which has high scalability.

[0083] Reference Figure 2 For step S300, the state of each port is determined based on the high-frequency link pole current, and the target port whose DC voltage meets the preset voltage condition is selected as the time-domain optimization target based on the state of each port, including the following steps:

[0084] Step S310: Determine whether the state of each port meets the zero-voltage soft-switching condition based on the high-frequency chain pole current.

[0085] Step S320: Select the target port with the smallest DC voltage from the ports that have lost the zero-voltage soft-switching condition as the time-domain optimization target.

[0086] Understandably, the mechanism by which an H-bridge achieves zero-voltage switching (ZVS) under traditional phase-shift modulation (PSM) is based on... Figure 6 As shown, the necessary condition is that the high-frequency link current flowing into the H-bridge at the rising edge of the output square wave voltage at this port must be less than 0. However, when there is a voltage mismatch at multiple ports, the current at the rising edge of the port with the lower voltage is very likely to become positive, leading to the loss of the zero-voltage soft-switching condition and a sharp increase in switching losses.

[0087] The high-frequency chain pole current of a port that has lost the zero-voltage soft-switching condition is greater than 0.

[0088] For step S400, compare the high-frequency link pole current at the start rising edge of the target port H-bridge output square wave voltage with the high-frequency link pole current at the end rising edge of the target port H-bridge output square wave voltage. The smaller of the high-frequency link pole currents at the start rising edge and the end rising edge of the target port H-bridge output square wave voltage is I. Min , will I Min Set as the target pole current.

[0089] Reference Figure 3 Adjusting the global common duty cycle based on the target pole current includes the following steps:

[0090] Step S410: When the target pole current is greater than 0, smoothly increase the global common duty cycle;

[0091] Step S420: When the target pole current is less than 0, reduce the global common duty cycle according to the set safety margin;

[0092] In step S430, when the voltages at each port are restored to matching and soft switching can be achieved under pure phase-shift modulation, the common duty cycle is smoothly adjusted to 0.

[0093] Specifically, refer to Figure 8 The target pole current I Min The feedback error signal is input to the single-target duty cycle PI regulator (PI_d).

[0094] If I Min A value >0 indicates that the switching device is in a hard-switching state, and the PI controller outputs a positive integral signal to smoothly increase the common duty cycle d.

[0095] If I Min If the value is less than 0, it indicates that the soft-switching condition has been met. The PI controller will appropriately reduce the duty cycle d according to the set safety margin to prevent over-modulation from causing deterioration of return power and conduction loss.

[0096] Meanwhile, to ensure that power transmission does not get out of control due to excessive duty cycle, an upper limit constraint is imposed on the common duty cycle to avoid port power loss due to excessive duty cycle.

[0097] The upper limit of the global duty cycle is expressed as: ;

[0098] In the formula, f is the upper limit of the global duty cycle. s L is the switching frequency. ij，Max For the maximum equivalent series inductance, I ref，Max For the maximum target average current value, V' Min The minimum port voltage after being reduced to the reference port, where n is the number of ports.

[0099] Reference Figure 4 For step S500, the phase shift angle between ports is dynamically adjusted using the decoupling matrix, including the following steps:

[0100] Step S510: Compare the average current setpoint of each port with the actual sampled value to obtain the current error;

[0101] Step S520: Multiply the current error by the output of the proportional-integral regulator and the decoupling matrix to obtain the phase shift compensation amount;

[0102] Step S530: Dynamically adjust the phase shift angle between ports according to the phase shift angle compensation amount.

[0103] Specifically, refer to Figure 8 Introducing the duty cycle d will cause the active power originally transmitted by the MAB through phase shifting to drop. Therefore, the system adds a power feedforward decoupling network H based on the inversion of the small signal gain matrix.

[0104] The decoupling matrix H is derived as follows:

[0105] ;

[0106] .

[0107] In the formula, H is the power decoupling matrix; G is the gain matrix; ΔI is the average current error increment at each port; and ΔD is the phase shift angle control increment.

[0108] Average current setpoint at each port (e.g., I) 2ref The current error is generated by comparing the actual sampled value (e.g., ΔI2) with the actual sampled value. After passing through their respective PI regulators, the error is multiplied by the decoupling matrix H to output the phase shift compensation amount for each port. This compensation amount is superimposed with the basic phase shift angle and output to the PWM generator of the four active bridge converter to compensate for the power drop caused by the change in duty cycle and to offset the interference of duty cycle d fluctuations on the power loop.

[0109] Reference Figure 9 and Figure 10 Based on the transition simulation waveform, when the load or bus conditions change and the system can achieve ZVS by satisfying the traditional PSM (i.e., the pole current is less than 0), the PI_d controller will automatically and smoothly adjust the duty cycle to zero, and the system will return to pure PSM mode. There are no abrupt changes in the duty cycle throughout the entire process, resulting in a smooth voltage transition.

[0110] Reference Figure 11If the traditional offline lookup table method is used, the duty cycle will suddenly change to 0 or a certain value when the operating condition is switched, resulting in a sharp drop and oscillation of the load voltage. In contrast, the SSOOCS strategy of this invention eliminates the sudden change in duty cycle and speeds up the dynamic response stabilization time of the system by about 5ms. At the same time, since it adopts the analytical expression of multiply-accumulate operation, the calculation time of a single control cycle in a conventional digital signal processor is less than 3μs. Under the premise of reducing the conduction loss by 0.27%, it can additionally realize the soft switching of multiple devices, completely freeing up computing power and improving operating efficiency.

[0111] The above is a detailed description of the preferred embodiments of this application, but this application is not limited to the embodiments. Those skilled in the art can make various equivalent modifications or substitutions without departing from the spirit of this application, and these equivalent modifications or substitutions are all included within the scope defined by the claims of this application.

Claims

1. A soft-switching online optimization control method for a multi-active bridge converter, characterized in that, include: Obtain the circuit parameters of each port of the multi-active bridge converter; Based on the unified time-domain high-frequency link current analytical model, the high-frequency link pole current at the start and end rising edges of the H-bridge output square wave voltage at each port is calculated according to the circuit parameters. The state of each port is determined based on the high-frequency chain pole current, and the target port whose DC voltage meets the preset voltage condition is selected as the time domain optimization target based on the state of each port. The smaller of the high-frequency link pole currents at the start and end rising edges of the target port H-bridge output square wave voltage is determined as the target pole current, and the global common duty cycle is adjusted according to the target pole current. The phase shift angle between ports is dynamically adjusted using a decoupling matrix to compensate for the power drop caused by changes in duty cycle.

2. The online optimization control method for soft switching of a multi-active bridge converter according to claim 1, characterized in that, The circuit parameters include DC voltage, inductance parameters, switching frequency, and target transmission power.

3. The online optimization control method for soft switching of a multi-active bridge converter according to claim 1, characterized in that, The unified time-domain high-frequency link current analytical model decomposes the multi-degree-of-freedom modulated square wave voltage with duty cycle into two sets of sub-square waves with halved amplitude, different phases, and only phase-shift modulation. This decomposes the converter under multi-degree-of-freedom modulation into two sets of sub-converters containing only phase-shift modulation. Then, by solving the two sets of sub-square waves and superimposing them using the linear combination interpolation method, the high-frequency link current pole values ​​at the corresponding time are obtained.

4. The online optimization control method for soft switching of a multi-active bridge converter according to claim 3, characterized in that, The unified time-domain high-frequency link current analytical model is configured with a port number parameter, which can be changed to be applicable to multiple active bridge converters with any number of ports.

5. The online optimization control method for soft switching of a multi-active bridge converter according to claim 1, characterized in that, The step of determining the state of each port based on the high-frequency chain pole current, and selecting target ports whose DC voltage meets preset voltage conditions as time-domain optimization targets based on the state of each port, includes: Determine whether the state of each port meets the zero-voltage soft-switching condition based on the high-frequency chain pole current. Select the target port with the smallest DC voltage from the ports that have lost the zero-voltage soft-switching condition as the time-domain optimization target.

6. The online optimization control method for soft switching of a multi-active bridge converter according to claim 1, characterized in that, The step of adjusting the global common duty cycle based on the target pole current includes: When the target pole current is greater than 0, the global common duty cycle is smoothly increased. When the target pole current is less than 0, the global common duty cycle is reduced according to the set safety margin. When the voltages at each port are restored to matching and soft switching can be achieved under pure phase-shift modulation, the common duty cycle is smoothly adjusted to 0.

7. The online optimization control method for soft switching of a multi-active bridge converter according to claim 6, characterized in that, In the process of adjusting the global common duty cycle according to the target pole current, an upper limit constraint is imposed on the global common duty cycle; The upper limit of the global common duty cycle is expressed as: ; In the formula, f is the upper limit of the global duty cycle. s L is the switching frequency. ij，Max For the maximum equivalent series inductance, I ref，Max For the maximum target average current value, V' Min The minimum port voltage after being reduced to the reference port, where n is the number of ports.

8. The online optimization control method for soft switching of a multi-active bridge converter according to claim 1, characterized in that, Dynamically adjusting the phase shift angle between ports using a decoupling matrix includes: The current error is obtained by comparing the average current setpoint at each port with the actual sampled value. The phase shift compensation amount is obtained by multiplying the output of the proportional-integral regulator of the current error by the decoupling matrix; The phase shift angle between ports is dynamically adjusted based on the phase shift angle compensation amount.

9. A soft-switching online optimization control system for a multi-active bridge converter, characterized in that, include: The main circuit of the multi-active bridge converter and the controller connected to the main circuit; The main circuit of the multi-active bridge converter includes a high-frequency transformer and multiple fully controlled H-bridge arms connected to the winding ports of the high-frequency transformer. The controller is used to generate a pulse width modulation drive signal according to the online optimization control method for soft switching of the multi-active bridge converter according to any one of claims 1 to 8, and to control the on and off of each switching device in the fully controlled H-bridge arm, so as to realize soft switching of the switching devices under light load or voltage mismatch conditions.

10. The online optimization control system for soft switching of a multi-active bridge converter according to claim 9, characterized in that, The controller includes a soft-switching online optimization control module and a power decoupling module; The soft-switching online optimization control module is used to calculate the high-frequency link pole current at the start and end rising edges of the square wave voltage output by each port of the H-bridge based on the unified time-domain high-frequency link current analytical model and the circuit parameters; determine the state of each port based on the high-frequency link pole current; and select the target port whose DC voltage meets the preset voltage conditions as the time-domain optimization target based on the state of each port. The smaller of the high-frequency link pole currents at the start and end rising edges of the target port H-bridge output square wave voltage is determined as the target pole current, and the global common duty cycle is adjusted according to the target pole current. The power decoupling module is used to dynamically adjust the phase shift angle between ports using a decoupling matrix to compensate for power drops caused by changes in duty cycle.

Images