Multiport DC-ac power converter and methods for the design optimization and loss minimization of the same

Abstract

A triple-active bridge (TAB) converter and methods to optimize the same is disclosed. The TAB converter includes a high-frequency three-winding transformer and three active full bridges, each coupled to the high-frequency transformer through a different passive network. The converter also includes a centralized controller providing gating pulses to semiconductor switches within the active full bridges, a power pulsating buffer (PPB) configured to reduce capacitor requirements on the DC port, and an unfolder stage to convert rectified AC voltage to a pure sine wave. The centralized controller may optimize performance by varying at least one of a switching frequency, a duty cycle of bridge voltages, and a phase shift between bridge voltages. The PPB may include a buck active power decoupler with a half-bridge, inductor, and capacitors, and also a primary full bridge and a power transfer impedance block.

Description

MULTIPORT DC-AC POWER CONVERTER AND METHODS FOR THE DESIGN OPTIMIZATION AND LOSS MINIMIZATION OF THE SAMERELATED APPLICATIONS

[0001] This application claims the benefit of U.S. provisional patent application 63 / 665,228, filed June 27, 2024, titled “Multiport DC-AC Power Converter and Methods for the Design Optimization and Loss Minimization of the Same,” the entirety of the disclosure of which is hereby incorporated by this reference.STATEMENT REGARDING FEDERALLY SPONSORED RESEARCH OR DEVELOPMENT

[0002] This invention was made with government support under DE-SC0022600 awarded by the Department of Energy. The government has certain rights in the invention.TECHNICAL FIELD

[0003] Aspects of this document relate generally to multiport DC-AC power converters.BACKGROUND

[0004] Photovoltaic (PV) cells have emerged as a prominent source of renewable energy, particularly valuable in remote residential areas where traditional grid infrastructure may be lacking or unreliable. However, integrating PV systems into these remote settings poses several challenges. These PV cells are often integrated with storage and grid sources to provide residential power using a generalized multi -active bridge (MAB) converter (e.g., a triple-active bridge converter, or TAB).

[0005] Multi-active bridge converters are widely used for their flexibility and capability to handle multiple power conversion stages (e.g., DC-AC -DC). Despite their advantages, these converters suffer from high losses during light load operations. The inefficiencies are mainly due to increased conduction and switching losses, which are exacerbated when the converter operates at non-unity gain, leading to further degradation in performance.

[0006] In remote residential areas, where maintenance resources are limited, the high maintenance costs of these converters become a substantial burden. Traditional converter designs often rely on electrolytic capacitors for energy storage and filtering purposes. Thesecapacitors, while effective, have relatively short lifespans and lower reliability, necessitating frequent replacements and leading to increased maintenance demands. Moreover, the necessity to achieve high efficiency across a wide range of load conditions remains unmet by conventional designs, further compounding the challenges.

[0007] Although triple-active bridge (TAB) converters have emerged as a prominent topology for three-port power conversion, accurate modeling is challenging and crucial for the design and control of TAB converters, especially when integrating an AC voltage waveform at the output, introducing a drastic and periodic voltage gain change in the converter's dynamics.SUMMARY

[0008] According to one aspect, a triple-active bridge (TAB) DC-AC-DC converter includes a high-frequency three-winding transformer; three active full bridges, each coupled to the high-frequency transformer through a different passive network; a centralized controller providing gating pulses to semiconductor switches within the active full bridges; a power pulsating buffer (PPB) configured to reduce capacitor requirements on a DC port; and an unfolder stage to convert rectified AC voltage to a pure sine wave.

[0009] In some embodiments, the centralized controller optimizes performance by varying at least one of a switching frequency, a duty cycle of bridge voltage, and a phase shift between bridge voltages. In some embodiments, the centralized controller optimizes performance based on at least one of the following calculated parameters: minimum conduction loss, minimum switching loss, minimum total loss, minimum magnetic loss, or maximum zerovoltage switching (ZVS) events. In some embodiments, the TAB converter includes a plurality of triple-active bridge (TAB) converters connected in parallel, wherein each of the plurality of TAB converters shares the same centralized controller.

[0010] In some embodiments, the PPB includes a buck active power decoupler with a half-bridge, inductor, and capacitors; an output voltage sensor; and an inductor current sensor; wherein the output voltage sensor and the inductor current sensor are communicatively coupled to the centralized controller; wherein the centralized controller is configured to provide PPB control logic for double-line frequency attenuation. In some embodiments, the converter is configured to eliminate high value electrolytic capacitors by using ceramic capacitors and the power pulsating buffer. In some embodiments, the TAB converter includes an islanding protection mechanism to safely disconnect an AC grid side during power outages.

[0011] In some embodiments, the passive networks include non-resonant LC circuits to minimize the higher-order harmonic RMS of the bridge currents to enhance an efficiency profile and AC output voltage total harmonic distortion (THD) for a given power transfer. In some embodiments, the passive network at the AC-side of the converter includes an LC circuit, the LC circuit having an effective inductance that is adaptively varied by the centralized controller using frequency modulation based on at least one of an operating point parameter, ports’ voltages, or a power levels of the converter, wherein such variation of inductance is performed to extend or preserve the zero-voltage switching (ZVS) operation range or minimize 60 Hz RMS current through at least one passive network of the converter over dynamic load and line conditions. In some embodiments, the centralized controller utilizes a hybrid pulsewidth modulation (PWM) control scheme to provide a seamless transition between different control modes depending on voltage gains and load conditions.

[0012] In some embodiments, the centralized controller is further configured to generate auxiliary modulation variables (AMVs) and an operating frequency in real time by executing a two-dimensional computational method including: (i) applying a Fourier-based curve fitting algorithm offline to obtain modulation waveforms corresponding to different combinations of port voltages and power levels, (ii) performing a two-layer polynomial regression on the coefficients of the fitted waveforms to produce a compact model mapping port conditions to modulation parameters, and implementing the resulting regression model within the digital controller to dynamically compute optimized AMVs and switching frequency as a function of converter operating conditions, thereby reducing onboard memory usage, improving control resolution, and enhancing real-time efficiency.

[0013] According to one aspect, a triple-active bridge (TAB) DC-AC-DC converter system includes at least one TAB converter having: a high-frequency three-winding transformer; three active full bridges, each coupled to the high-frequency transformer through a different passive network; a centralized controller providing gating pulses to semiconductor switches within the active full bridges; a power pulsating buffer (PPB) configured to reduce capacitor requirements on a DC port; and an unfolder stage to convert rectified AC voltage to a pure sine wave.

[0014] In some embodiments, the centralized controller optimizes performance by varying at least one of a switching frequency, a duty cycle of bridge voltage, and a phase shift between bridge voltages. In some embodiments, the centralized controller optimizes performance based on at least one of the following calculated parameters: minimum conduction loss, minimum switching loss, minimum total loss, minimum magnetic loss, or maximum zero-voltage switching (ZVS) events. In some embodiments, 15. the centralized controller optimizes performance based on generalized harmonic approximation all-harmonics-inclusive domain circuit modeling of the TAB converter system’s operating parameters including: the relationship of the conduction loss and the switching loss with the RMS and switching transient values of the active bridge currents and power device parameters.

[0015] In some embodiments, the at least one TAB converter includes a plurality of the at least one TAB converters connected in parallel. In some embodiments, each of the at least one TAB converters share the same centralized controller. In some embodiments, the PPB of each of the at least one TAB converters includes a buck active power decoupler with a half-bridge, inductor, and capacitors; an output voltage sensor; and an inductor current sensor; wherein the output voltage sensor and the inductor current sensor are communicatively coupled to the centralized controller; wherein the centralized controller is configured to provide PPB control logic for double-line frequency attenuation.

[0016] According to one aspect, a triple-active bridge (TAB) DC-AC-DC converter includes a high-frequency three-winding transformer; three active full bridges, each coupled to the high-frequency transformer through a different passive network; a centralized controller providing gating pulses to semiconductor switches within the active full bridges, wherein the centralized controller provides the gating pulses based on circuit modeling to optimize at least two of: minimum conduction loss, minimum switching loss, minimum total loss, minimum magnetic loss, or maximum zero-voltage switching (ZVS) events; a power pulsating buffer (PPB) configured to reduce capacitor requirements on a DC port; and an unfolder stage to convert rectified AC voltage to a pure sine wave.

[0017] In some embodiments, the centralized controller optimizes performance by varying at least one of a switching frequency, a duty cycle of bridge voltage, and a phase shift between bridge voltages.

[0018] Aspects and applications of the disclosure presented here are described below in the drawings and detailed description. Unless specifically noted, it is intended that the words and phrases in the specification and the claims be given their plain, ordinary, and accustomed meaning to those of ordinary skill in the applicable arts. The inventors are fully aware that they can be their own lexicographers if desired. The inventors expressly elect, as their own lexicographers, to use only the plain and ordinary meaning of terms in the specification and claims unless they clearly state otherwise and then further, expressly set forth the “special” definition of that term and explain how it differs from the plain and ordinary meaning. Absent such clear statements of intent to apply a “special” definition, it is the inventors’ intent anddesire that the simple, plain and ordinary meaning to the terms be applied to the interpretation of the specification and claims.

[0019] The inventors are also aware of the normal precepts of English grammar. Thus, if a noun, term, or phrase is intended to be further characterized, specified, or narrowed in some way, then such noun, term, or phrase will expressly include additional adjectives, descriptive terms, or other modifiers in accordance with the normal precepts of English grammar. Absent the use of such adjectives, descriptive terms, or modifiers, it is the intent that such nouns, terms, or phrases be given their plain, and ordinary English meaning to those skilled in the applicable arts as set forth above.

[0020] Further, the inventors are fully informed of the standards and application of the special provisions of 35 U.S.C. § 112(f). Thus, the use of the words “function,” “means” or “step” in the Detailed Description or Description of the Drawings or claims is not intended to somehow indicate a desire to invoke the special provisions of 35 U.S.C. § 112(f), to define the invention. To the contrary, if the provisions of 35 U.S.C. § 112(f) are sought to be invoked to define the inventions, the claims will specifically and expressly state the exact phrases “means for” or “step for”, and will also recite the word “function” (i.e., will state “means for performing the function of [insert function]”), without also reciting in such phrases any structure, material or act in support of the function. Thus, even when the claims recite a “means for performing the function of . . . “ or “step for performing the function of . . . ,” if the claims also recite any structure, material or acts in support of that means or step, or that perform the recited function, then it is the clear intention of the inventors not to invoke the provisions of 35 U.S.C. § 112(f). Moreover, even if the provisions of 35 U.S.C. § 112(f) are invoked to define the claimed aspects, it is intended that these aspects not be limited only to the specific structure, material or acts that are described in the preferred embodiments, but in addition, include any and all structures, materials or acts that perform the claimed function as described in alternative embodiments or forms of the disclosure, or that are well known present or later-developed, equivalent structures, material or acts for performing the claimed function.

[0021] The foregoing and other aspects, features, and advantages will be apparent to those artisans of ordinary skill in the art from the DESCRIPTION and DRAWINGS, and from the CLAIMS.BRIEF DESCRIPTION OF THE DRAWINGS

[0022] The disclosure will hereinafter be described in conjunction with the appended drawings, where like designations denote like elements, and:

[0023] FIG. 1 is a schematic view of a triple-active bridge (TAB) DC-AC-DC converter architecture;

[0024] FIG. 2 is a schematic view of a controller interfacing with a TAB converter architecture;

[0025] FIG. 3 is a schematic view of a power pulsating buffer for a TAB converter architecture;

[0026] FIG. 4 is a schematic view of two configurations of a power transfer impedance block used in a TAB converter architecture;

[0027] FIG. 5 is a waveform diagram showing gate signals and corresponding voltage waveforms at different switching nodes of a TAB converter;

[0028] FIG. 6 is a schematic view of a delta equivalent model of a TAB converter switching network;

[0029] FIG. 7 is a process flow of an algorithm to provide optimized modulation variables and the final regression model, based on a TAB converter’s operating point; and

[0030] FIG. 8 is a process flow for the implementation of a control algorithm in a DSP controller for a TAB converter.DETAILED DESCRIPTION

[0031] This disclosure, its aspects and implementations, are not limited to the specific material types, components, methods, or other examples disclosed herein. Many additional material types, components, methods, and procedures known in the art are contemplated for use with particular implementations from this disclosure. Accordingly, for example, although particular implementations are disclosed, such implementations and implementing components may comprise any components, models, types, materials, versions, quantities, and / or the like as is known in the art for such systems and implementing components, consistent with the intended operation.

[0032] The words "exemplary," "example," or various forms thereof are used herein to mean serving as an example, instance, or illustration. Any aspect or design described herein as "exemplary" or as an "example" is not necessarily to be construed as preferred or advantageous over other aspects or designs. Furthermore, examples are provided solely forpurposes of clarity and understanding and are not meant to limit or restrict the disclosed subject matter or relevant portions of this disclosure in any manner. It is to be appreciated that a myriad of additional or alternate examples of varying scope could have been presented, but have been omitted for purposes of brevity.

[0033] While this disclosure includes a number of embodiments in many different forms, there are shown in the drawings and will herein be described in detail particular embodiments with the understanding that the present disclosure is to be considered as an exemplification of the principles of the disclosed methods and systems, and is not intended to limit the broad aspect of the disclosed concepts to the embodiments illustrated.

[0034] Contemplated herein are methods for modeling, control, design optimization and loss minimization in multiport DC-AC power converters, as well as a novel hardware implementation. The contemplated methods aim to overcome the challenges discussed above by introducing and optimizing a novel multiport (N-port) power electronic converter. According to various embodiments, this converter employs a three-winding transformer, multiple active bridge cells, and optimized passive networks, coupled with advanced control strategies and design-level optimizations. By doing so, it provides a cost-effective, reliable, and efficient solution for integrating PV systems into remote residential areas, significantly improving performance under light load conditions and reducing the maintenance burden traditionally associated with such systems.

[0035] Various modeling techniques have been proposed, each offering distinct advantages and limitations. Generalized Harmonic Approximation (GHA) is a computationally efficient method that approximates the converter's voltage and current waveforms with higher- order harmonics of the switching frequency, enabling efficient loss analysis. The average model approach provides a simplified method capturing the converter's steady state behavior, suitable for control design but lacks sufficient accuracy for predicting transient responses. State-Space Averaging (SSA), however, offers a more accurate representation than the average model by considering the converter's switching dynamics, making it suitable for advanced control strategies. Selecting the modeling technique sometimes depends on the specific application and desired accuracy margins. While SSA provides a more comprehensive model at a high complex control design cost, the GHA technique remains a popular choice for loss analysis due to its computational efficiency.

[0036] Similarly, various control strategies have been developed to regulate the output voltage and power flow in TAB-based converters, each of which, offers distinct performance characteristics and computational complexity. Linear control techniques, such as Proportional-Integral (PI) controllers, are commonly used due to their simplicity. However, it can degrade the converter's performance under varying load conditions. Model Predictive Control (MPC) is an advanced technique that offers an improved dynamic performance by considering future system behavior, but requires higher computational and time-intensive resources. Similarly, the Sliding Mode Control (SMC) provides robustness against system uncertainties but can introduce chattering in the control signal, requiring careful design with a hysteresis band and sometimes a higher computational control technique. Selecting a proper control strategy depends on factors like desired performance, required computational complexity, and system robustness. Properly designed linear techniques offer a good balance between simplicity and performance, while advanced methods like MPC and SMC can be beneficial for demanding applications.

[0037] In addition to the modelling and control techniques, efficiency is a critical performance metric in TAB-based converters. Efforts focus on various techniques to minimize losses and improve overall converter efficiency. Zero-Voltage Switching (ZVS) techniques significantly reduce conduction losses by achieving ZVS during switching events, leading to improved efficiency. However, its implication when integrating an AC voltage on the TAB remains significant. An optimal control minimizes conduction and switching losses by optimizing the voltage duty cycle of the active bridges' switching nodes voltages.

[0038] The present disclosure is directed to a unified approach encompassing modeling, circuit analysis, power flow optimization techniques, and a corresponding hardware implementation methodology for a three-port, three-active bridge DC-AC-DC converter. According to various embodiments, the converter utilizes three active full bridges and a high- frequency (HF) three-winding transformer, where the proposed technology aims to enhance the cost-effectiveness, reliability, and efficiency of three-port converters across a wide load and port voltage gain range. This is achieved by introducing an optimal phase-duty control strategy coupled with a multi-variable design-level optimization technique.

[0039] According to various embodiments, the implemented loss optimization technique is a two-stage process. The first stage meticulously quantifies power losses (conduction and switching) within the HF switching network. This quantification is achieved by modeling the converter using the generalized harmonic approximation (GHA) technique. Additionally, a universal loss minimization criterion for three-port converters is derived by proposing a portequivalent converter model for accurate switching loss estimation. In some embodiments, this criterion may be Zero-Voltage Switching (ZVS). The second stage establishes acomprehensive mathematical framework that synthesizes the circuit model to identify optimal design-level parameters, ensuring both high efficiency and reduced production cost.

[0040] While the mathematical framework provides the optimal operating point, the former developed models are implemented within a digital controller alongside decoupled PI controllers that regulate the voltage phase duty of each active bridge, according to various embodiments. Experimental results employing the proposed optimal phase-duty control strategy and the design-level optimization have demonstrated a 15% reduction in conduction losses compared to the conventional design on a dual active bridge converter. Furthermore, a cost reduction exceeding 50% for the three-port converter has also been achieved. These results demonstrate the effectiveness of the contemplated methods for enhancing the performance of three-port converters, essential for the proper integration of renewable energies and their storage in remote residential areas.

[0041] It should be noted that the following discussion is centered around triple-active bridge (TAB) converters, specifically a three-port, three-active bridge DC-AC-DC converter. However, those skilled in the art will recognize that the designs and methods discussed herein may be adapted for application to other multiport power converters, and that the examples used below are non-limiting embodiments.

[0042] FIG. 1 is a schematic view of a non-limiting example of a triple-active bridge (TAB) single-stage converter (i.e., conversion does not require an intermediate conversion step). Specifically, FIG. 1 shows a three-port, three-active bridge DC-AC-DC converter 100. This converter 100 is well adapted for the seamless interconnection of DC-AC-DC voltages, where the input side has a DC input 102 (which maybe, for example, intended for PV panel connections), the AC output side has AC output 104 (which maybe, for example, intended for grid connections), and the DC output side has DC output 106 (which maybe, for example, intended for energy storage system connections). This is advantageously provided as a single converter topology solution, eliminating the need for higher number of isolated converters and subsequent active bridges for such voltage level integrations.

[0043] According to various embodiments, the converter 100 comprises three active full bridges (active full bridges 110, 120, and 130) that control voltage transformation and power flow to and from a port (e.g., 102, 104, 106) using active switches SI to S12 (e.g., semiconductor switching devices). These full bridges are connected to a high-frequency three- winding transformer 150 through passive networks (e.g., an inductor, LC tank, etc.). The three- winding transformer 150 facilitates power transfer between the ports 102, 104, and 106 using power transfer impedance blocks 152, 154, and 156, as represented in FIG. 1. The activeswitches SI to S 12 control the power flow and voltage transformation using an optimal phaseduty control strategy to minimize losses and enhance efficiency, as will be discussed further, below. According to various embodiments, the converter also comprises a centralized controller 200 (shown in FIG. 2) that provides the gating pulses to the semiconductor switches SI to S 12 in all active bridges.

[0044] In some embodiments, the passive networks may comprise non-resonant LC configurations to minimize the higher-order harmonic RMS of the bridge currents and hence enhance the efficiency profile and AC output voltage total harmonic distortion (THD) for a given power transfer.

[0045] In some embodiments, the contemplated TAB converter 100 may have a stacked design, which simplifies troubleshooting and reduces maintenance costs. In the stacked design, each power stage (e.g., a single TAB converter) is realized individually and irrespective of the other connected ports (e.g., 102, 104, or 106). A galvanically isolated connection is maintained among the three ports (i.e., between any two of ports 102, 104, or 106), enabling a safe, single converter-based solution for DC-AC-DC voltage level integration. This stacked design can reduce production costs, due to the lower number of utilized full active bridges 110, 120, or 130. According to various embodiments, the stacked design is also associated with low maintenance costs due to an islanded-power-stages design, which will be discussed below.

[0046] In some embodiments, the TAB converter 100 may have a modular design that allows for easy scalability and more convenient troubleshooting due to plug-and-play fabrication structure of each active bridge 110, 120, or 130. This would enable building larger PV systems as the input to TAB converter 100 by adding additional TAB converter 100 modules connected to each other in parallel. According to various embodiments, the system efficiency could be maximized by an optimal power split control algorithm, in which each active module is operated near a loss-minima point that is obtained from standalone efficiency vs. load characterization of each module.

[0047] According to various embodiments, the contemplated TAB converter 100 comprises an islanding protection. Islanding protection may be provided by developing a controlling method for the TAB converter 100 to safely disconnect the AC output 104 on the AC grid side in case of a power outage, preventing potential safety hazards for utility workers.

[0048] The single-stage TAB converter 100 contemplated herein comprises a centralized controller 200 that provides the gating pulses to the semiconductor switches SI to S 12 in all active-bridges 110, 120, and 130. Specifically, the controller provides gate signals to vary one or many of the converter 100 control variables to optimize performance: (i) switchingfrequency, (ii) duty cycle of the bridge voltages, and (iii) phase shift between the bridge voltages. This alleviates the data communication, security, and synchronization challenge typically faced by a decentralized control architecture, according to various embodiments. In some embodiments, these converter 100 control variables may vary as a function of load power and / or AC voltage phase angle.

[0049] As shown in FIG. 1, 51 116, 52 126, 53 136, q>2 166, and cp3 168 are phase shift control signals for the switches (S1-S12). Specifically, the main modulation variables, (p2166 and <p3168, represent the main phase shifts between the input port 102 leading leg’s gate signal and the AC side 104 leading leg’s gate signal and the DC side 106 leading leg’s gate signal, respectively. Similarly, the auxiliary modulation variables 116, 82126, and <53136, provide the intra-phase shift between the leading and lagging legs of each full bridge 110, 120, and 130. Hence, the output voltages can be controlled in a closed-loop, and the active power flow requirements can be met. Vin 170 and V'3 172 indicate the DC voltages and V'2 171 is the rectified AC voltage at different stages, while vl 173, v'2 174, and v'3 175 are the switching nodes’ voltages.

[0050] According to various embodiments, the multi-constraint multi-variable optimization framework identifies the optimal control variables for different converter 100 operating conditions, targeting objectives or objective functions such as minimizing conduction loss, switching loss, total loss, magnetic loss, and maximizing the soft-switching range of the converter 100 (i.e., maximizing ZVS events). In some embodiments, the implementation of the optimal duty cycle or frequency tracking block can be carried out using multivariate polynomial regression functions of port-voltage gain and port power at any given sampling instant or by incorporating a gradient descent algorithm.

[0051] A digital signal processing-based implementation of the contemplated closed-loop three-port TAB control system 200 includes optimal duty tracking features, which control the mismatch between PV output (e.g., PV connected to DC input 102) and battery needs (e.g., batteries connected to DC output 106) at different charging stages, to increase battery lifespan by minimizing the battery side full-bridge RMS currents. Two proportional-integral and / or proportional-integral-derivative (PI / PID) controllers 200 output the phase shift control variables, and optimal auxiliary modulation variables are fetched as feed-forward variables from the implemented optimal duty cycle models, according to various embodiments.

[0052] According to various embodiments, the AC output is generated by an unfolder stage 180, which converts the rectified AC voltage V'2 172 to a pure sine wave. In someembodiments, the input 102 from the photovoltaic (PV) source is processed through a buckboost stage and transferred via the high-frequency inverter to the output, which includes an HF inverter stage to convert the ripple-free high DC voltage to a grid-compatible 60Hz AC. Advantageously, the unfolder stage 180 offers lower control complexity and reduced losses compared to high-frequency (HF) inverters. The unfolder stage 180 facilitates an intermediate AC link realizable by electrolytic-less capacitor solution, unlike a conventional architecture with a higher intermediate DC link capacitance.

[0053] As previously discussed, conventional TAB converters can be problematic in use cases where the load is low and the gain is non-unitary. In the contemplated converter 100 topology, the AC waveform at the output port 104 is generated from a full-wave-rectified (FWR) voltage waveform across the interlinked capacitor between the unfolder stage 180 and the TAB's secondary bridge 120 as opposed to the conventional method where the AC waveform is generated through a terminal HF inverter stage. Therefore, the interlinking capacitor requirements are reduced and can be realized by ceramic capacitors. An unfolder stage 180 also has the advantage of lower switching losses and simpler digital control implementation due to line frequency switching.

[0054] Put differently, employing an unfolder stage 180 to generate a pure sinewave from a full-wave-rectified sinewave eliminates requiring high-value capacitors at the input of the unfolder stage 180. This allows the troublesome electrolytic capacitors to be replaced by reliable ceramic capacitors. Another electrolytic capacitor can be eliminated through the use of a power pulsating buffer 300, as will be discussed with respect to FIG. 3, below.

[0055] These two approaches (i.e., the unfolder 180 and the buffer 300) lead to an electrolytic capacitor-less design of a single converter 100 solution that can integrate solar panels to the AC grid as well as an energy storage system. This converter 100 design offers superior performance in terms of cost-effectiveness, reliability, longevity, and efficiency compared to conventional designs.

[0056] While the discussion has been focused on the non-limiting example shown in FIG. 1, a three-port converter 100, in other embodiments the design improvements contemplated herein may be applied to a multiport (i.e., N-port) power electronic converter, to enhance the efficiency, cost-effectiveness, and reliability. According to various embodiments, the contemplated N-port converter comprises an N-winding transformer, (N+l) active bridge cells with semiconductor switching devices, and N passive networks that couple each active bridge cell with the transformer. A centralized controller provides gating pulses to the semiconductor switches in all active bridges.

[0057] In conventional TAB converters, to suppress the double line frequency reflected from the output AC port 104 to the input port 102, a high-value capacitor is required. Such high-value capacitances can be realized by electrolytic capacitors, leading to low reliability and low power density. According to various embodiments, the contemplated converter 100 design can reduce or eliminate the need for such problematic capacitors.

[0058] FIG. 3 is a schematic view of a non-limiting example of a buck derived power pulsating buffer (PPB) 300. Specifically, FIG. 3 shows the power pulsating buffer 300 setup, which is implemented to reduce the capacitor requirements on the DC link 102 (i.e., the PV input). As shown, the PPB 300 includes a buck active power decoupler comprising half-bridge 310 (i.e., switches SI 312 and S2 314), an inductor Lp 320, and capacitors Cdc 330 and Cp 322. The inductor Lp 320 is in series with a lower-value capacitor Cp 322 connected to the half bridge's 310 switching node 316. The PPB 300 parallelly connects to the PV side input 102 as well as the primary side full-active bridge 110 to mitigate the double-line frequency ripple reflected from the AC grid side 104 to the input side 102 (e.g., the PV side).

[0059] This configuration allows for a significant reduction in the capacitor size needed at the input 102, replacing traditional electrolytic capacitors with more reliable ceramic capacitors. This buffer 300 smooths out power fluctuations, helps manage the pulsating power, and mitigates the double-line frequency ripple reflected from the AC output side 104, improving the overall reliability and power density of the converter 100. Also, the required voltage rating of the capacitors in a power pulsating buffer 300 is lower than the electrolytic capacitors in the conventional design, making it a cost-effective solution. In use, the input side 102 DC link capacitors are replaced with this power-pulsating buffer 300.

[0060] FIG. 4 is a schematic view of two configurations of a non-limiting example of a power transfer impedance block (e.g., power transfer impedance blocks 152, 154, or 156) used in the TAB converter (see FIG. 3). The left side shows a single inductor configuration 410, while the right side depicts a non-resonant LC circuit configuration 420, comprising an inductor Lr and a capacitor Cr (LC circuit 420 can be a variety of different resonant, tank, or tuned LC circuits other than the sample circuit shown as an example in FIG. 4). These configurations are part of the design-level optimization to enhance the efficiency and performance of the converter 100 by properly tuning the impedance network for optimal power transfer. Both configurations manage the power transfer impedance but offer different characteristics for power flow and impedance matching. In some embodiments, the single inductor configuration 410 is preferred, for example, in case of fixed frequency converter 100 operation that might ease the electromagnetic interference (EMI) filter design process, whilein certain embodiments the non-resonant LC circuit configuration 420 is more advantageous, for example, when switching frequency is modulated as a function of load power and AC line cycle phase angle.

[0061] The contemplated converter's 100 passive components such as the non-resonant LC tank parameters and HF three winding transformer's 150 turns ratios can greatly affect its operating point. As the transformer 150 turns ratio deviates from the nominal voltage conversion ratios while assuming the AC RMS voltage to be the nominal value for the AC port 104, the sum of bridge mean square currents and hence the circulating reactive power increases. Therefore, to achieve an optimal efficiency profile, turns ratio is maintained equal to the nominal port voltage ratios, according to various embodiments. As for the passive components selection for a non-resonant LC configuration, same equivalent inductance at switching fundamental is obtained by a higher L and lower C or a lower L and higher C. If the power transfer inductance needs to be integrated as a controllable leakage as part of the transformer 150, a lower L e.g., less thanfor a 3.5kW three-port converter 100 design and higher C is preferred in some embodiments. If overall bridge RMS currents need to be minimized, the equivalent inductance values at higher harmonics are expected to be higher, which is facilitated by a choice of higher L and lower C, in which case, the power transfer inductance can be realized in the form of an external magnetic core. According to various embodiments, design optimization of power transfer inductances and transformer turns ratio in a three-port TAB 100 can lead to maximum weighted efficiency profile across the desired load power range.

[0062] In some embodiments, one aspect of optimization contemplated herein is the design optimization of power transfer inductances, transformer turns ratio, and / or absolute turns count, leading to a maximum weighted efficiency profile across the desired load power range. These design-level parameters may be optimized for one or more objective functions including, but not limited to, minimum conduction loss, minimum switching loss, minimum total loss, minimum magnetic loss, and maximum ZVS events (i.e., maximum soft-switching range of the converter 100).

[0063] In some embodiments, a design optimization toolbox which could be realized in the form of a graphical user interface (GUI) or a software facilitates the identification of optimized parameters, maximizing soft-switching events and efficiency in the converter 100. The design of non-resonant LC tank parameters and transformer 150 turns ratio in a phase- frequency-controlled three-port TAB converter 100 may also be optimized to achieve high efficiency and reliability across various load conditions.

[0064] According to various embodiments, the implemented loss optimization technique contemplated herein is a two-stage process. The first stage meticulously quantifies power losses (i.e., conduction and switching) within the HF switching network. This quantification is achieved by modeling the converter 100 using the generalized harmonic approximation (GHA) technique. Additionally, a universal loss minimization criterion for three-port converters 100 is derived by proposing a port-equivalent converter 100 model for accurate switching loss estimation. In some embodiments, the universal loss minimization criteria may be Zero- Voltage Switching (ZVS). The second stage establishes a comprehensive mathematical framework that synthesizes the circuit model to identify optimal design-level parameters, ensuring both high efficiency and reduced production cost. The conduction and switching losses are related to the transformer 150 winding RMS currents and their peaks, derived from the generalized harmonic approximation (GHA)-oriented all-harmonics-inclusive domain circuit modeling of the TAB converter 100, as will be discussed below.

[0065] According to various embodiments, the power impedance block 152 / 154 / 156 connects the full active bridges 110 / 120 / 130 of each side to the three winding transformer 150 and is connected in series with the transformer’s 150 windings at its respective side. The impedance blocks 152 / 154 / 156 can play an important role in enabling the active power transfer among the converter’s ports 102 / 104 / 106, and their individual values affect the HF current waveshape. In some embodiments, their values affect the ZVS range in each of the active bridges 110 / 120 / 130 as well as conduction losses in the switching network and transformer 150 windings.

[0066] In some embodiments, FIG. 1 can be equivalently utilized as a hybrid series- resonant triple active bridge (HSR-TAB) DC-AC-DC topology, featuring variable frequency modulation based on the average powers of the ports 102 / 104 / 106. To avoid excessive circulating current arising from port voltage mismatch at ports 102 / 104 / 106, only one resonant LC tank (e.g., power transfer impedance block 154, LC circuit 420, etc.) is placed at the AC port 104. This LC tank circuit 420 operates near resonance under nominal load to maximize power transfer and shifts to higher switching frequencies at lighter load to enhance ZVS and reduce RMS currents. A time-domain harmonic decomposition method based on superposition can model the system (e.g., using controller 200) as three full -active bridges, enabling accurate estimation of inductor currents and switching dynamics without a high computational burden. Controller 200, for example, can optimize the passive networks (e.g., power transfer impedance blocks 152 / 154 / 156), switching, and other parameters discussed herein to optimize transient values of the active bridge currents and power device parameters, including its on-stateresistance and dynamic voltage-dependent body capacitance. A charge-based model (e.g., using controller 200) accounts for nonlinear output capacitance during dead time, and a combined loss model guides modulation optimization.

[0067] The synthesized modulation models are integrated into a closed-loop control architecture (e.g., using controller 200) for the TAB converter 100, where controller 200 regulates output port voltages 104 / 106 by adjusting the primary modulation variables (e.g., controller 200 is a proportional-integral controller in some embodiments). To minimize converter 100 losses under varying operating conditions, auxiliary modulation variables (AMVs) are precomputed as functions of AC line cycle phase and port 102 / 104 / 106 powers rather than storing memory-intensive lookup tables, employing a two-stage Fourier-based and polynomial regressions to model AMVs as functions of the AC voltage phase. This approach allows real-time generation of AMVs beyond predefined discretization steps, ensuring both computational efficiency and control resolution within the hardware constraints of embedded systems.

[0068] FIG. 5 is a waveform diagram that shows the voltage waveforms of the switching nodes under hybrid voltage duty and phase shift control within a non-limiting example of the contemplated converter 100. Specifically, FIG. 5 shows the gate signals (i.e., GS1, GS4) and the corresponding voltage waveforms (i.e., vl 173, v'2 174, and v'3 175) at different switching nodes. This control method optimizes the switching events to achieve minimal conduction and switching losses by detailing the dynamics of the multi-port converter in an all-harmonic- inclusive domain to mathematically present and minimize the HF current flowing through the switching networks, HF inductors, and transformer winding leveraging the generalized harmonic approximation (GHA) technique for accurate modeling and efficient control of the converter. Additionally, this shows how the control signals (i.e., 51 116, cp2 166, and cp3 168) affect the voltage waveforms, ensuring efficient power conversion with minimal losses across various operating conditions.

[0069] Conventional methods for minimizing conduction and switching losses often rely on pre-calculated data, straining memory and processing power of controller 200 with timeintensive approaches. Advantageously, the contemplated methods provide four alternative approaches to optimize the TAB-based converters' 100 efficiency performance.

[0070] The first utilizes a search algorithm that dynamically adjusts the converter's 100 operation to minimize losses during the steady-state operation, reducing memory demands on controller 200 compared to traditional methods. The second approach leverages mathematical models to predict optimal modulation variables based on real-time voltage and loadingconditions, enabling efficient control without overwhelming the digital controller 200. The third approach provides a design methodology, providing the optimal design-level parameters that perform most efficiently. The fourth approach includes designing a power pulsating buffer 300 to reduce the capacitor needs at the converter's 100 input 102, eliminating the required electrolytic capacitor at the DC link 102, leading to higher reliability and power density.

[0071] In some embodiments, the optimized 51 116, 52 126, and 53 136 auxiliary modulation variables in FIG. 5 are stored in a look-up table and / or predefined multivariate polynomial regression models and fed to the converter 100 during the 60Hz line cycle in each interval of 1° of the output voltage phase to ensure loss minima tracking. As a specific, nonlimiting example, in one embodiment, the algorithmic computation time for generating the control variables using a digital signal processor (DSP) with 100MHz clock for a three-port DC-AC-DC converter is found to be less than 5 / J.S, which combined with other proportionalintegral (PI) controller programs, turns out less than 7 / is, hence supporting any switching frequencies up to 140kHz.

[0072] Meanwhile, the principal phase shifts between the primary bridge 110 and secondary bridge 120, (|)2 166, and between the primary bridge 110 and tertiary bridge 130, (|)3 168, are governed through the PI closed-loop controller 200 such that the output ports 104 / 106 are regulated at grid voltage (for 104) and battery voltage (for 106) throughout the 60Hz line cycle. These modulation variables govern the semi-square waveshape at the full bridge 110 / 120 / 130 switching nodes, which is then applied to the power transfer impedances and govern the current-voltage relation of the power transfer impedance blocks 152 / 154 / 156.

[0073] FIG. 6 is a schematic view of a non-limiting example of a delta equivalent model of the contemplated TAB switching networks, and the generalized harmonic approximation (GHA) model of semi-square voltage waveforms. The converter 100 is modeled in the allharm onics-inclusive domain using the Fourier transform, and the semi-square wave voltage waveform across the switching nodes of each full active bridge 110 / 120 / 130 is defined as a summation of the sinusoid voltage sources of the odd higher-order harmonics of the switching frequency, according to various embodiments. The three-winding transformer 150 in series with the power transfer impedances at each side may be modeled with their equivalent delta model to facilitate the mathematical representation of the HF current flowing through the transformer 150 windings at each side.

[0074] According to various embodiments, the delta equivalent model simplifies the analysis of the converter 100 by representing the switching networks in a delta configuration, while the GHA model provides a detailed representation of the voltage waveforms, includinghigher-order harmonics. This model aids in accurately quantifying power losses (i.e., conduction and switching) within the HF switching network and optimizing the converter's 100 performance by facilitating the mathematical representation of the converter’s 100 governing equations, i.e., time-varying HF inductor’s current and its active power flow, in the allharm onics-inclusive domain using the Fourier transform, which enables the applicability of the constrained non-linear optimization techniques.

[0075] According to various embodiments, the mathematical converter 100 modeling contemplated herein may be leveraged to provide a graphical-user-interface (GUI) based toolbox that outputs the three-port converter’s 100 design-level parameters and identifies all losses happening within the converter 100. This user-friendly software toolbox calculates optimal design-level parameter settings for any TAB-based converter 100 and can be integrated with existing power electronics circuit simulation software such as MATLAB and Plexim or used as a standalone simulator. The engineering insights provided by this software into the proposed three-port converter’s 100 efficiency performance as a function of the parameter selection and modulation techniques can automate and streamline the design optimization workflow, facilitating a rapid prototyping process. Major loss categories can be identified in an existing converter at all possible operating conditions and then modified with a selection of optimum passive components and optimized modulation scheme and control variables at each operating sample, in some embodiments. Similarly, the contemplated software can be used to troubleshoot the control logic, compute peak current, RMS current, and switching instantaneous currents that are useful for loss function formulation at each sample, and mitigate any given type of loss function, including the conduction losses or switching losses or their combination, according to various embodiments.

[0076] The contemplated software enables the identification and optimization of losses, provides the loss division of each power stage as a function of design parameters, and blends the converter's 100 design parameters into a unified approach for performance optimization, according to various embodiments. In some embodiments, this unified design-level and switching modulation optimization for loss minima tracking in multi-port DC-AC-DC converters 100 can perform four to six times faster than traditional circuit simulators. Additionally, the optimized modulation variables may be integrated as a submodule within the software, enhancing efficiency, cost-effectiveness, and reliability, making it a valuable contribution to the field of power electronics.

[0077] The methods contemplated herein present a design-level optimization approach alongside an optimized multi-variable PWM control technique for minimizing the sum of meansquare currents for a TAB DC-AC-DC converter 100, according to various embodiments. This provides several advantages over existing, conventional technologies.

[0078] First, it is mathematically proven that applying the TAB design-level optimization reduces the converter's 100 conduction losses, especially at nominal load conditions, by providing a previously obtained operating benchmark to maximize efficiency. The application of design-level optimization improves the light-load efficiency as well and reduces the designed transformer size by minimizing the winding current peaks and RMSs solely by selecting the converter's 100 parameters. The application of design-level optimization of a TAB converter 100 is novel.

[0079] Second, in some embodiments, the theoretical modeling of the non-resonant TAB network is performed using the proposed all-harmonics-inclusive GHA model. Due to the higher degree of freedom to shape the switching nodes' voltages applied to the transformer 150 in series with the non-resonant LC tank, the TAB converter 100 has several tens of operating modes in the optimized design. The complexity makes using existing time-domain-based converter's analysis computationally heavy and time-consuming, and devoid of advantage for AC voltage generation where one of the output side voltages varies from 0V to 240 2V that requires updating modulation variables at each switching cycle. The digital implementation becomes more complicated as the number of modulation variables increases, shifting the controller from the DPS modulation up to the Penta Phase Shift (PPS) modulation. On the contrary, the contemplated GHA-based voltage and currents model facilitates the synthesis of a unified interpretation of the transformer 150 current regardless of the switching modulation pattern and the number of control variables involved, leading to a concise expression and easing down the digital control implementation.

[0080] Additionally, a hybrid and optimized PWM modulation scheme that depends on the converter's 100 operating point may leverage the advantages of the PWM control strategies to optimize system loss. The hybrid PWM modulation provides a seamless transition between different control modes depending on voltage gains and load conditions (e.g., from DPS control at unity voltage gains to TPS and PPS control at other load conditions and voltage gains, etc.), in some embodiments.

[0081] According to various embodiments, the three full active bridges 110 / 120 / 130 are modulated such that the three winding transformer 150 providing galvanic isolation between any two ports (e.g., 102 to 104 or 102 to 106) will safely disconnect from the AC grid side 104 in case of a power outage, preventing potential safety hazards for utility workers.

[0082] In some embodiments, stackable and modular design can allow for easy scalability using optimal power split control algorithm, which splits the reference power in an optimum ratio among the participating modules based on the efficiency vs. load characteristics with an objective to operate each module as closely as possible to the maximum efficiency point. This would enable building larger PV systems leveraging the modular TAB architecture by adding additional converter 100 modules in parallel.

[0083] According to various embodiments, loss optimization objective functions are formulated by relating the conduction loss and the switching loss with the transformer 150 winding RMS currents and their peaks, synthesized from the GHA-oriented all-harmonics- inclusive domain circuit modeling of the TAB converter 100.

[0084] In some embodiments, a digital signal processing-based implementation of the contemplated closed-loop three-port TAB control system 200 having an optimal duty tracking feature may be carried out to control the mismatch between PV output fed into DC input 102 and the battery needs at DC output 106 as an energy storage system at different charging stages. This will minimize the RMS current profile on the battery side full-bridge 130 and hence increase the battery’s life span.

[0085] In some embodiments, the proposed loss optimization approach includes the design optimization toolbox of power transfer impedances in a three-port TAB converter 100, GUI, wherein the passive components and design-level parameters are provided in advance in mathematical analysis software and simulation environment, leading to the minimized conduction losses and maximized soft-switching events in a switching cycle across the desired load power range.

[0086] FIG. 7 is a process flow of a non-limiting example of an algorithm to provide the optimized modulation variables and the final regression model, based on the converter’s 100 operating point. The objective function is to optimize the modulation variables in Penta Phase Shift (PPS) modulation in order to minimize the losses in the proposed three-port converter 100. The DC-AC dual active bridge (DAB) stage of the contemplated converter 100 generates a full-wave rectified (FWR) voltage at its output 104 where the modulation variables are subjected to the power balance constraints to meet the instantaneous active power delivery of the converter 100 and ZVS activation, respectively. The resultant optimized 116, 62126, and <53136 modulation variables are stored in a lookup table and fetched by the converter 100 in each interval of 1° during the 60Hz cycle of the output voltage. In some embodiments, an additional approach is to generate a regression model of the auxiliary modulation variablesbased on the input voltage and load demands. Such an optimization algorithm for a specific output power and voltage is shown in FIG. 7.

[0087] FIG. 8 is a process flow for the implementation of a non-limiting example of a control algorithm in a DSP controller 200 belonging to an embodiment of the contemplated converter 100. According to various embodiments, the principal phase shift between the input and output full-active bridges' leading legs, (p2166 or <p3168, is governed through the PI closed- loop controller 200 to maintain the synchronization throughout the AC line cycle. In some embodiments, the regression model of the auxiliary modulation variables are calculated in advance based on the load demands and the input voltage and are readily available in the DSP controller 200. In the case of using a look-up table storing the modulation variables, three auxiliary modulation variables may be stored in the lookup table, fetched at each 1° interval of the AC line cycle.

[0088] It will be understood that implementations are not limited to the specific components disclosed herein, as virtually any components consistent with the intended operation of a method and / or system implementation for an optimized multiport DC-AC power converter 100 may be utilized. Accordingly, for example, although particular methods, models, and converters may be disclosed, such components may comprise any shape, size, style, type, model, version, class, grade, measurement, concentration, material, weight, quantity, and / or the like consistent with the intended operation of a method and / or system implementation for an optimized multiport DC-AC power converter 100 may be used. In places where the description above refers to particular implementations of an optimized multiport DC-AC power converter 100, it should be readily apparent that a number of modifications may be made without departing from the spirit thereof and that these implementations may be applied to other power converters.

Claims

CLAIMSWhat is claimed is:

1. A triple-active bridge (TAB) DC-AC-DC converter, comprising: a high-frequency three-winding transformer; three active full bridges, each coupled to the high-frequency transformer through a different passive network; a centralized controller providing gating pulses to semiconductor switches within the active full bridges; a power pulsating buffer (PPB) configured to reduce capacitor requirements on a DC port; and an unfolder stage to convert rectified AC voltage to a pure sine wave.

2. The TAB converter of claim 1, wherein the centralized controller optimizes performance by varying at least one of a switching frequency, a duty cycle of bridge voltage, and a phase shift between bridge voltages.

3. The TAB converter of claim 1, wherein the centralized controller optimizes performance based on at least one of the following calculated parameters: minimum conduction loss, minimum switching loss, minimum total loss, minimum magnetic loss, or maximum zero-voltage switching (ZVS) events.

4. The TAB converter of claim 1, further comprising a plurality of triple-active bridge (TAB) converters connected in parallel, wherein each of the plurality of TAB converters shares the same centralized controller.

5. The TAB converter of claim 1, wherein the PPB comprises: a buck active power decoupler with a half-bridge, inductor, and capacitors; an output voltage sensor; and an inductor current sensor; wherein the output voltage sensor and the inductor current sensor are communicatively coupled to the centralized controller; wherein the centralized controller is configured to provide PPB control logic for doubleline frequency attenuation.

6. The TAB converter of claim 1, wherein the converter is configured to eliminate high value electrolytic capacitors by using ceramic capacitors and the power pulsating buffer.

7. The TAB converter of claim 1, further comprising an islanding protection mechanism to safely disconnect an AC grid side during power outages.

8. The TAB converter of claim 1, wherein the passive networks include non-resonant LC circuits to minimize the higher-order harmonic RMS of the bridge currents to enhance an efficiency profile and AC output voltage total harmonic distortion (THD) for a given power transfer.

9. The TAB converter of claim 1, wherein the passive network at the AC-side of the converter includes an LC circuit, the LC circuit having an effective inductance that is adaptively varied by the centralized controller using frequency modulation based on at least one of an operating point parameter, ports’ voltages, or a power levels of the converter, wherein such variation of inductance is performed to extend or preserve the zero-voltage switching (ZVS) operation range or minimize 60 Hz RMS current through at least one passive network of the converter over dynamic load and line conditions.

10. The TAB converter of claim 1, wherein the centralized controller utilizes a hybrid pulsewidth modulation (PWM) control scheme to provide a seamless transition between different control modes depending on voltage gains and load conditions.

11. The TAB converter of claim 1, wherein the centralized controller is further configured to generate auxiliary modulation variables (AMVs) and an operating frequency in real time by executing a two-dimensional computational method comprising: (i) applying a Fourier-based curve fitting algorithm offline to obtain modulation waveforms corresponding to different combinations of port voltages and power levels, (ii) performing a two-layer polynomial regression on the coefficients of the fitted waveforms to produce a compact model mapping port conditions to modulation parameters, and implementing the resulting regression model within the digital controller to dynamically compute optimized AMVs and switching frequency as a function of converter operating conditions, thereby reducing onboard memory usage, improving control resolution, and enhancing real-time efficiency.

12. A triple-active bridge (TAB) DC-AC-DC converter system, comprising at least one TAB converter having: a high-frequency three-winding transformer; three active full bridges, each coupled to the high-frequency transformer through a different passive network; a centralized controller providing gating pulses to semiconductor switches within the active full bridges; a power pulsating buffer (PPB) configured to reduce capacitor requirements on a DC port; and an unfolder stage to convert rectified AC voltage to a pure sine wave.

13. The at least one TAB converter of claim 12, wherein the centralized controller optimizes performance by varying at least one of a switching frequency, a duty cycle of bridge voltage, and a phase shift between bridge voltages.

14. The at least one TAB converter of claim 13, wherein the centralized controller optimizes performance based on at least one of the following calculated parameters: minimum conduction loss, minimum switching loss, minimum total loss, minimum magnetic loss, or maximum zero-voltage switching (ZVS) events.

15. The at least one TAB converter of claim 14, wherein the centralized controller optimizes performance based on generalized harmonic approximation all-harmonics-inclusive domain circuit modeling of the TAB converter system’s operating parameters including: the relationship of the conduction loss and the switching loss with the RMS and switching transient values of the active bridge currents and power device parameters.

16. The TAB converter system of claim 12, wherein the at least one TAB converter comprises a plurality of the at least one TAB converters connected in parallel.

17. The TAB converter system of claim 16, wherein each of the at least one TAB converters share the same centralized controller.

18. The TAB converter system of claim 17, wherein the PPB of each of the at least one TAB converters comprises: a buck active power decoupler with a half-bridge, inductor, and capacitors; an output voltage sensor; and an inductor current sensor; wherein the output voltage sensor and the inductor current sensor are communicatively coupled to the centralized controller; wherein the centralized controller is configured to provide PPB control logic for doubleline frequency attenuation.

19. A triple-active bridge (TAB) DC-AC-DC converter, comprising: a high-frequency three-winding transformer; three active full bridges, each coupled to the high-frequency transformer through a different passive network; a centralized controller providing gating pulses to semiconductor switches within the active full bridges, wherein the centralized controller provides the gating pulses based on circuit modeling to optimize at least two of: minimum conduction loss, minimum switching loss, minimum total loss, minimum magnetic loss, or maximum zerovoltage switching (ZVS) events; a power pulsating buffer (PPB) configured to reduce capacitor requirements on a DC port; and an unfolder stage to convert rectified AC voltage to a pure sine wave.

20. The TAB converter of claim 19, wherein the centralized controller optimizes performance by varying at least one of a switching frequency, a duty cycle of bridge voltage, and a phase shift between bridge voltages.

Images