Hybrid AI–Quantum Co-Design of a SiC-Based DAB Converter for Ultra-Fast EV Charging

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

1. This paper proposes an AI-driven quantum computing framework for modeling, design, and control of converter-dominated energy systems, with a focus on ultra-fast EV charging applications. The authors combine AI-based surrogate modeling with quantum-assisted optimization in a co-design architecture.
2. The topic is relevant and interesting, particularly in the context of future energy systems and intelligent power electronics.
3. I assume that the editorial board has already conducted a software-based plagiarism check prior to sending the manuscript for review and has determined that this is not a case of plagiarism or self-plagiarism. As a reviewer, I have not performed an independent plagiarism check.
4. Furthermore, I am not in a position to reliably assess the potential use of AI-assisted writing tools. If such use is subject to the journal’s policies, I leave this aspect to the editorial board’s procedures and expertise.
5. The article references 51 works, 12 of which were published in 2024 to 2026. The proposed references are sufficient for understanding the article and evaluating its relevance and originality. I do not assess that the authors should supplement the references with additional sources.
6. The paper is precisely written and easy to understand, with communication that is well-suited to the chosen journal and its readership.
7. I would only point out that the manuscript shows inconsistent and inappropriate use of list environments. In several places, numbered lists are used where itemized lists (bulleted lists) would be more correct.
8. I appreciate that the author clearly states that the work is a framework-oriented conceptual study rather than an immediate industrial solution. The manuscript is well structured and presents a well-defined multi-layer architecture, which is conceptually sound and easy to follow. I also acknowledge the author’s effort in identifying broader application areas beyond the specific case study, which strengthens the relevance of the proposed approach.
9. However, while the conceptual direction is well motivated, the manuscript does not clearly indicate a concrete pathway toward practical realization of the proposed framework. In particular, the transition from the presented conceptual architecture to a feasible implementation remains underdeveloped. Clarifying how the individual components could be realistically integrated or validated in practice would be very intersting for readers and would significantly strengthen the contribution.
10. The manuscript presents an interesting and relevant conceptual framework. With minor improvements in form, it could make a valuable contribution to the field.
11. I certainly suggest that the article be accepted and published in the scope and form after correcting the mentioned minor formatting and typographical errors.

Author Response

We sincerely thank the Reviewer for the careful reading of the manuscript, for the positive overall assessment, and for the constructive comments. We appreciate the Reviewer’s recognition of the relevance of the topic, the conceptual soundness of the proposed framework, and the clarity of the manuscript presentation. In the revised version, we have carefully addressed the Reviewer’s observations and further improved the manuscript in terms of structure, methodological clarity, practical realization pathway, and formatting.

Comment 1

This paper proposes an AI-driven quantum computing framework for modeling, design, and control of converter-dominated energy systems, with a focus on ultra-fast EV charging applications. The authors combine AI-based surrogate modeling with quantum-assisted optimization in a co-design architecture.

Response

We thank the Reviewer for the accurate summary of the manuscript and for correctly identifying the central idea of the proposed work. In the revised version, we have further improved the focus and clarity of the paper by narrowing the scope to a SiC-based dual active bridge (DAB) converter for ultra-fast EV charging, which is now explicitly reflected in the revised title, abstract, methodology, and case-study formulation. This revision was made to improve technical specificity and strengthen the engineering interpretability of the proposed framework.

Comment 2

The topic is relevant and interesting, particularly in the context of future energy systems and intelligent power electronics.

Response

We are grateful for this positive assessment. In the revised manuscript, we have preserved the broader motivation related to intelligent power electronics and future energy systems, while making the manuscript more technically focused through a concrete ultra-fast EV charging case study. This was done to better connect the broader relevance of the topic with a specific and technically interpretable converter-oriented application.

Comment 3

I assume that the editorial board has already conducted a software-based plagiarism check prior to sending the manuscript for review and has determined that this is not a case of plagiarism or self-plagiarism. As a reviewer, I have not performed an independent plagiarism check.

Response

We thank the Reviewer for this note. No response is required from our side regarding this point, and we remain fully aligned with the journal’s editorial procedures and publication policies.

Comment 4

Furthermore, I am not in a position to reliably assess the potential use of AI-assisted writing tools. If such use is subject to the journal’s policies, I leave this aspect to the editorial board’s procedures and expertise.

Response

We thank the Reviewer for this clarification. We fully respect the journal’s editorial policies and procedures regarding this matter.

Comment 5

The article references 51 works, 12 of which were published in 2024 to 2026. The proposed references are sufficient for understanding the article and evaluating its relevance and originality. I do not assess that the authors should supplement the references with additional sources.

Response

We thank the Reviewer for this positive evaluation of the literature base. Nevertheless, in the revised version we further improved the literature positioning in order to strengthen the introduction and respond to the comments of the other reviewers. In particular, we refined the literature review in the Introduction and incorporated more targeted recent references related to ultra-fast EV charging systems, charger design constraints, and converter-oriented charging architectures, while also improving the integration of the literature into the manuscript narrative.

Comment 6

The paper is precisely written and easy to understand, with communication that is well-suited to the chosen journal and its readership.

Response

We sincerely appreciate this encouraging comment. In the revised manuscript, we have made additional editorial improvements to preserve and further enhance the clarity of presentation, while also restructuring the manuscript in accordance with the journal template requirements into separate Introduction, Materials and Methods, Results, Discussion, and Conclusions sections.

Comment 7

I would only point out that the manuscript shows inconsistent and inappropriate use of list environments. In several places, numbered lists are used where itemized lists (bulleted lists) would be more correct.

Response

We thank the Reviewer for this useful formatting observation. The manuscript has been carefully checked and the list formatting has been revised for consistency. Inappropriate numbered lists were corrected where necessary, and the presentation of enumerations was aligned more closely with the expected journal style. In addition, the manuscript underwent broader formatting cleanup to improve consistency across sections, figures, tables, and structured items.

Comment 8

I appreciate that the author clearly states that the work is a framework-oriented conceptual study rather than an immediate industrial solution. The manuscript is well structured and presents a well-defined multi-layer architecture, which is conceptually sound and easy to follow. I also acknowledge the author’s effort in identifying broader application areas beyond the specific case study, which strengthens the relevance of the proposed approach.

Response

We thank the Reviewer for this positive and thoughtful assessment. In the revised manuscript, we have retained the framework-oriented character of the work while making the practical scope more explicit. In particular, we now frame the work consistently as a simulation-based proof-of-concept study, and we have strengthened the link between the conceptual multi-layer architecture and the selected engineering case study. This was done to preserve the broader relevance of the framework while improving technical clarity and practical interpretability.

Comment 9

However, while the conceptual direction is well motivated, the manuscript does not clearly indicate a concrete pathway toward practical realization of the proposed framework. In particular, the transition from the presented conceptual architecture to a feasible implementation remains underdeveloped. Clarifying how the individual components could be realistically integrated or validated in practice would be very interesting for readers and would significantly strengthen the contribution.

Response

We sincerely thank the Reviewer for this important and constructive observation. We agree that, in the original version, the pathway from the conceptual framework to a practically interpretable implementation was not sufficiently explicit.

To address this point, the revised manuscript has been substantially strengthened in the following ways:

1. A specific engineering case study was introduced and consistently adopted throughout the manuscript.
   Rather than keeping the framework at a broad conceptual level, we now focus on a SiC-based DAB converter in a representative 350 kW ultra-fast EV charging system, which provides a concrete application context for the proposed methodology.
2. An overall system schematic was added.
   The revised manuscript now includes a complete architecture of the considered ultra-fast charging system, including the AFE stage, DC-link, isolated DAB charging stage, battery-side interface, and supervisory controller. This clarifies where the proposed AI–quantum co-design framework interacts with the charging system in practice. See Figure 2.
3. The converter topology was made explicit.
   To avoid remaining at an abstract converter-block level, we introduced the selected SiC-based DAB topology and explained why it is used as the concrete realization platform for the proposed co-design study. See Figure 3 and Section 2.3.
4. A practical co-design workflow and validation pathway were added.
   The revised manuscript now includes a dedicated Co-Design Workflow and Validation Procedure subsection and an explicit workflow diagram showing the sequence of simulation-based data generation, surrogate-model training, quantum-assisted candidate search, high-fidelity re-validation, and iterative refinement. This was added specifically to clarify how the conceptual framework could be implemented and validated in practice. See Section 2.8 and Figure 4.
5. The manuscript now consistently frames the study as a simulation-based proof-of-concept validation strategy.
   We explicitly clarify that the present work is not claimed as an immediate industrial solution or hardware-demonstrated deployment, but as a technically interpretable computational workflow that can serve as a practical bridge between conceptual co-design and future implementation-oriented studies. This clarification is made in the Methods, Results, Discussion, and Conclusions sections.

We believe that these revisions substantially improve the practical realization pathway of the manuscript and directly address the Reviewer’s concern.

Comment 10

The manuscript presents an interesting and relevant conceptual framework. With minor improvements in form, it could make a valuable contribution to the field.

Response

We thank the Reviewer for this positive assessment. Following this recommendation, we have improved the manuscript both structurally and editorially. In addition to formatting corrections, we also introduced substantial clarifications regarding the case study, converter topology, workflow, validation logic, and manuscript structure. We hope that these revisions further strengthen the contribution and overall readability of the paper.

Comment 11

I certainly suggest that the article be accepted and published in the scope and form after correcting the mentioned minor formatting and typographical errors.

Response

We are grateful for the Reviewer’s supportive recommendation. In response, we carefully revised the manuscript and corrected formatting, structural, and typographical issues throughout the text. We also took the opportunity to improve the technical focus and the practical interpretability of the work beyond purely formal corrections. We thank the Reviewer again for the encouraging and constructive feedback.

 

Once again, we sincerely thank Reviewer 1 for the positive evaluation and constructive suggestions. The comments were very helpful in improving the clarity, formatting consistency, and practical interpretability of the revised manuscript.

Reviewer 2 Report

Comments and Suggestions for Authors During the digitization of EPS characterized by the distributed renewables,
energy communities, EVs, BMS, converter-dominated energy systems is the
key feature of the future smart grids. The goal of submitted text is highly ambitious focusing on the AI–quantum
framework designed for the modeling, design, and control of the
converter-dominated power and energy systems. It starts with high quality
background description moving to the detailed requirements for development of the AI and quantum computing listing main issues with
energy and power systems transition. Next section deals with the hybrid AI–quantum co-design framework for the
modeling, design, and control of converter-dominated power and energy
systems. In the following section, proposed frameworks was validated on the task of
adaptive charging for an ultra-fast EV charging system. Despite the important topic addressed by the authors and interesting ideas
to use the quantum-assisted optimization, this text can't published. suggestions. 1. Authors must remove (or extend) the combined references like [4-6], [12-18], etc. Each must be cited separately and its contribution to SATA must be explained. 2. in line 281 the continuous DAE system is presented in very general form. What is objective of this math formalism here? from which class if function f? Can you introduce it for the addressed problem? if not what the goal? 3. Lines 297 - 301 introduce the similar discrete formalism for AI. and again, its goal is not very clear here 4. Next subsection is really confusing and mathematically vulgar. What is functional J ? is it convex problem? what means "this may be instantiated as" ? pls be specific. This seems to be MILP? 5. Why do you need proposed solver here? please compare your results with other solvers

Author Response

We sincerely thank the Reviewer for the detailed and technically focused comments. We appreciate the Reviewer’s recognition of the importance of the topic and of the potentially interesting role of quantum-assisted optimization in converter-dominated energy systems. At the same time, we fully acknowledge that the original version of the manuscript was too ambitious in scope and insufficiently precise in several key methodological and mathematical parts. In response, the manuscript has been substantially revised to improve technical specificity, mathematical clarity, literature positioning, workflow interpretability, and consistency of the engineering case study. Most importantly, the revised manuscript now adopts a much more focused scope, centered on a SiC-based dual active bridge (DAB) converter for a representative 350 kW ultra-fast EV charging system, rather than a broad and highly abstract converter-dominated systems framework.

Below we respond to each comment in detail.

General comment

During the digitization of EPS characterized by the distributed renewables, energy communities, EVs, BMS, converter-dominated energy systems is the key feature of the future smart grids. The goal of submitted text is highly ambitious focusing on the AI–quantum framework designed for the modeling, design, and control of the converter-dominated power and energy systems. It starts with high quality background description moving to the detailed requirements for development of the AI and quantum computing listing main issues with energy and power systems transition. Next section deals with the hybrid AI–quantum co-design framework for the modeling, design, and control of converter-dominated power and energy systems. In the following section, proposed frameworks was validated on the task of adaptive charging for an ultra-fast EV charging system. Despite the important topic addressed by the authors and interesting ideas to use the quantum-assisted optimization, this text can't published.

Response

We thank the Reviewer for the careful overall assessment. We agree that the original manuscript was too ambitious in scope and insufficiently focused from a technical and methodological standpoint. This concern was taken very seriously in the revised version.

To address this, the manuscript has been fundamentally tightened in scope and structure:

1. The title has been changed to
   “Hybrid AI–Quantum Co-Design of a SiC-Based DAB Converter for Ultra-Fast EV Charging”,
   which better reflects the actual content and no longer overstates the breadth of the contribution.
2. The abstract has been completely revised to emphasize the specific case study, the methodological structure, and the proof-of-concept nature of the work. The revised abstract now explicitly states that the manuscript presents a simulation-based proof-of-concept evaluation, rather than a universal framework for all converter-dominated systems.
3. The Introduction has been rewritten to provide a more focused literature review and a more explicit gap statement connecting ultra-fast EV charging, DAB converter selection, surrogate learning, and quantum-assisted optimization.
4. The manuscript structure has been reorganized into the standard journal format:
   Introduction – Materials and Methods – Results – Discussion – Conclusions,
   in line with the journal requirement.
5. A concrete engineering realization pathway has been introduced through:
   • an overall system schematic (Figure 2),
   • an explicit selected converter topology (Figure 3),
   • a clarified co-design workflow (Figure 4),
   • simulation-based results (Figures 5–6),
   • and structured comparative tables (Tables 1–4).

We believe that these revisions significantly reduce the excessive generality of the original manuscript and make the revised version technically clearer and more defensible.

Comment 1

Authors must remove (or extend) the combined references like [4-6], [12-18], etc. Each must be cited separately and its contribution to SATA must be explained.

Response

We thank the Reviewer for this important observation. We agree that the use of combined citations in the original manuscript reduced the clarity of the literature positioning and did not sufficiently explain the contribution of each reference.

In the revised manuscript, the literature review has been significantly improved in the following way:

1. Combined references were removed or substantially reduced, and the citation style was revised so that key sources are cited more explicitly and individually rather than as broad grouped ranges.
2. The Introduction was rewritten to organize the literature into distinct thematic groups:
   • ultra-fast EV charging systems and charger constraints,
   • charger architectures and isolated DC/DC converter relevance,
   • AI/surrogate modeling in power electronics,
   • and quantum/quantum-inspired optimization for engineering search problems.
3. The role of the cited literature is now explained more clearly. Rather than citing large groups of references without distinction, the revised text now clarifies which references support charger design constraints, which motivate 400-V/800-V fast-charging adaptability, which support AI-based performance approximation, and which motivate the use of advanced search mechanisms.
4. In addition, the literature base itself was updated and refined, including more targeted charging-related references to improve the relevance and specificity of the revised Introduction.

We agree that this point was important, and we believe the revised literature review is now considerably more transparent and better aligned with the manuscript scope.

Comment 2

in line 281 the continuous DAE system is presented in very general form. What is objective of this math formalism here? from which class if function f? Can you introduce it for the addressed problem? if not what the goal?

Response

We sincerely thank the Reviewer for this very important comment. We agree that, in the original manuscript, the mathematical formalism was introduced in too general a way and without a sufficiently clear methodological role.

To address this, the corresponding section has been substantially revised and rewritten. In the revised manuscript:

1. The previously overly generic system representation has been replaced by a more clearly framed reduced-order nonlinear state-space representation for the selected DAB charging converter. The text now explains explicitly that this representation is not intended as a complete full-order analytical model, but rather as a compact interface model linking the physical charging converter to the surrogate-learning and optimization workflow. See Section 2.4.
2. The role of the mathematical formalism is now stated explicitly:
   • it defines the dependency between design variables, operating conditions, and observable performance metrics;
   • it supports simulation-based data generation;
   • and it provides the physical basis for subsequent surrogate-assisted optimization.
3. We now clarify the meaning of the functions f(⋅) and g(⋅):
   they are not assumed to belong to a specific convex or closed-form function class, but represent the nonlinear converter behavior induced by the chosen DAB topology and operating conditions. The revised text explicitly states that the manuscript does not claim a universal analytical model class and that the formulation is used as a physically meaningful reduced-order representation for co-design purposes.
4. The revised section also introduces more concrete problem-dependent interpretation of the variables:
   • the state vector includes representative electrical and thermal quantities;
   • the control vector includes the phase-shift command and regulation signals;
   • the operating-point vector captures battery voltage, requested charging power, and charging-region variation;
   • and the design parameter vector includes transfer inductance, transformer ratio, switching frequency, and controller gains.

In short, we fully agree that the original version was too abstract here. The revised Section 2.4 was rewritten precisely to answer the Reviewer’s question: the mathematical formalism is now clearly introduced as a reduced-order physical interface model for the selected DAB case study, rather than as a vague universal dynamic formalism.

Comment 3

Lines 297 - 301 introduce the similar discrete formalism for AI. and again, its goal is not very clear here

Response

We thank the Reviewer for this comment and agree that, in the original manuscript, the AI-related formalism was introduced without sufficient explanation of its practical role.

This section has been substantially clarified in the revised manuscript, especially in Section 2.6 (AI Surrogate-Learning Layer). The main improvements are as follows:

1. The AI layer is now introduced explicitly as a surrogate-learning block whose role is to reduce the computational burden of repeated converter evaluation during multi-objective optimization. This practical role is stated directly at the beginning of Section 2.6.
2. The revised manuscript now defines a feature vector and a target vector for the DAB case study, with clear engineering meaning:
   • input features include battery voltage, charging power, switching frequency, phase shift, transfer inductance, transformer ratio, controller gains, dead time, and ambient temperature;
   • target outputs include efficiency, loss, ripple-related current variation, transient error, and junction-temperature rise.
3. The surrogate-learning task is now explicitly written as the approximation
   r^~=F_ϕ(z),
   and the revised text explains that the AI model is used not as a replacement for final validation, but as a fast predictor embedded in the optimization loop.
4. To reinforce this point, we also added Table 3, which summarizes the interaction between:
   • physical simulation,
   • dataset generation,
   • surrogate training,
   • candidate screening,
   • and high-fidelity re-validation.

Thus, the AI formalism now has a clearly stated methodological purpose: it defines the input–output mapping learned by the surrogate model and justifies the use of AI as a computational acceleration layer within the co-design workflow.

Comment 4

Next subsection is really confusing and mathematically vulgar. What is functional J ? is it convex problem? what means "this may be instantiated as" ? pls be specific. This seems to be MILP?

Response

We sincerely thank the Reviewer for this direct and important criticism. We agree that the optimization problem formulation in the original manuscript was not sufficiently precise and that the earlier wording created unnecessary ambiguity.

This part of the manuscript has been substantially rewritten in Section 2.5 (Multi-Objective Co-Design Problem Formulation). The following clarifications were introduced:

1. The revised manuscript now explicitly states that the problem is treated as a nonlinear constrained multi-objective optimization problem and not as a MILP formulation.
2. The decision vector ξ is now defined explicitly and concretely for the DAB case study as containing:
   • transfer inductance L_k​,
   • switching frequency f_s​,
   • transformer turns ratio n,
   • phase shift δ,
   • controller gains K_p​, K_i​,
   • and dead time d_t​.
3. The objective J(ξ) is now described clearly as a weighted scalarized engineering performance index used for comparative co-design. The revised manuscript explicitly states that:
   • J(ξ) is not assumed to be convex;
   • the problem is not reduced to a classical linear or mixed-integer linear program;
   • the scalarized formulation is used to compare candidate solutions under multiple coupled engineering criteria.
4. A representative form of the scalarized objective is now given explicitly in Equation (4), where the terms correspond to:
   • efficiency degradation,
   • losses,
   • thermal stress,
   • ripple,
   • dynamic response quality,
   • and implementation-related penalties.
5. The engineering meaning of the optimization variables has been further clarified through Table 2, which summarizes the variables, their bounds, and their role in the DAB ultra-fast charging case study.

We fully acknowledge that this was one of the weakest parts of the original manuscript. The revised text was rewritten precisely to eliminate the ambiguity noted by the Reviewer. We hope that the optimization problem is now presented in a sufficiently specific and technically consistent way.

Comment 5

Why do you need proposed solver here? please compare your results with other solvers

Response

We thank the Reviewer for this very important question. We agree that the original manuscript did not explain clearly enough why the proposed solver layer was needed and how its relevance should be assessed.

This issue has been explicitly addressed in the revised manuscript, especially in Section 2.7 (Quantum-Assisted Optimization Layer), Section 2.9 (Baseline Strategy and Evaluation Metrics), and Section 3.4 (Comparative Evaluation Against the Baseline Workflow). The main revisions are as follows:

1. The revised manuscript now states clearly that the proposed optimization layer is not introduced as a universally superior black-box solver, nor as a claim of guaranteed global optimality. Instead, it is framed as a quantum-assisted search component embedded in a hybrid engineering workflow. Its role is to improve the search structure of a high-dimensional, mixed-variable, multi-objective co-design problem.
2. We now explain explicitly why such a solver layer is relevant in the selected DAB charging application:
   • the converter must operate across a wide battery-voltage range;
   • the design space includes coupled electrical, magnetic, and control-related variables;
   • and acceptable operating points must balance efficiency, loss, thermal stress, ripple, and dynamic response.
     Under these conditions, the value of the solver is not that classical tuning is impossible, but that a more structured search mechanism may be beneficial for converter-aware multi-condition co-design.
3. Regarding comparison, we agree that the original manuscript was insufficient. In the revised version, we introduced a clear conventional baseline workflow consisting of:
   • fixed CC–CV charging logic,
   • conventional controller tuning,
   • and classical parameter-sweep-based operating-point evaluation.
     This baseline is now described in Section 2.9.
4. The results are now explicitly compared against this baseline in:
   • Section 3.1,
   • Section 3.4,
   • and Table 4,
     where representative engineering and workflow-level tendencies are summarized. These include charging-time tendency, loss-related tendency, efficiency, ripple, thermal stress, settling time, robustness to operating-point changes, retuning burden, and computational efficiency.
5. We also deliberately clarify the scope of this comparison. The revised manuscript does not claim an exhaustive benchmark against all state-of-the-art optimizers. Instead, it establishes a technically realistic comparison against a conventional engineering baseline and explicitly states this limitation in both the Results and Discussion sections.

We acknowledge that a broader optimizer benchmark would be valuable in future work. However, for the present revised manuscript, we believe that the newly introduced baseline comparison and clarified solver justification substantially address the Reviewer’s concern and make the role of the proposed optimization layer much more transparent.

 

We thank Reviewer 2 again for the highly valuable technical comments. These remarks were instrumental in improving the mathematical clarity, methodological precision, and engineering focus of the manuscript. We believe that the revised version is substantially stronger and more technically consistent as a result of these revisions.

Reviewer 3 Report

Comments and Suggestions for Authors

The article provides no detailed information and remains very vague.

The author introduces several topics — from AI‑Quantum frameworks to power converters and ultrafast EV charging — but without any technical depth or explanation associated with each section.

The article requires significant improvement as follows:

1. The title should be changed.
2. The abstract should clearly highlight the originality of this work.
3. The introduction should include a relevant and up‑to‑date literature review.
4. The methodology and approach should be clearly explained and demonstrated.
5. Results should be presented through simulations and/or experimental validation.
6. A discussion section should be included.
7. A conclusion and future research directions should be provided.

The author should also include an overall schematic of the proposed ultrafast charging system.

The associated converter topologies are missing.

Some simulation or experimental results are required to validate the proposed idea.

The references need to be improved.

The author may use the two articles below to strengthen the final version of the manuscript.

-Important Technical Considerations in Design of Battery Chargers of Electric Vehicles, Energies 2021.

-A Comprehensive Review of Universal Fast-Charging Techniques for 400-V and 800-V Electric Vehicles, IEEE 2025.

Author Response

We sincerely thank the Reviewer for the careful reading of the manuscript and for the direct and constructive comments. We fully acknowledge that the original version of the paper was too broad in scope, insufficiently specific in technical focus, and not adequately developed in terms of methodological clarity, validation structure, and engineering detail. We took these concerns very seriously.

In response, the manuscript has been substantially revised and refocused. The revised version is no longer presented as a broad and highly abstract AI–quantum framework for converter-dominated energy systems in general. Instead, it is now formulated as a focused simulation-based proof-of-concept study centered on a SiC-based dual active bridge (DAB) converter for a representative 350 kW ultra-fast EV charging system. This revision affected the title, abstract, introduction, methodology, results, discussion, conclusions, figures, tables, and literature review. We believe that the revised manuscript is now significantly more specific, technically interpretable, and better aligned with the Reviewer’s expectations.

Below we respond to each point in detail.

General comment

The article provides no detailed information and remains very vague. The author introduces several topics — from AI-Quantum frameworks to power converters and ultrafast EV charging — but without any technical depth or explanation associated with each section.

Response

We thank the Reviewer for this direct assessment. We agree that the original manuscript was too broad and insufficiently detailed. This concern became one of the main drivers of the revision.

To address it, the manuscript has been substantially tightened and made more concrete in the following ways:

1. The scope was narrowed from a broad converter-dominated systems framework to a specific ultra-fast EV charging case study.
   The revised manuscript now focuses explicitly on a SiC-based DAB converter within a representative 350 kW ultra-fast EV charging system. This shift was made to improve technical specificity and reduce conceptual vagueness.
2. The methodological structure was rewritten and expanded.
   The revised manuscript now contains a dedicated and detailed Materials and Methods section, including:
   • methodological overview,
   • system architecture,
   • selected converter topology,
   • physical and mathematical representation,
   • co-design problem formulation,
   • AI surrogate-learning layer,
   • quantum-assisted optimization layer,
   • co-design workflow and validation procedure,
   • and baseline strategy with evaluation metrics.
3. Results, Discussion, and Conclusions are now fully separated and explicitly developed.
   The revised manuscript now contains dedicated:
   • Results,
   • Discussion,
   • and Conclusions sections,
     which were either absent or insufficiently developed in the original version.
4. Figures and tables were substantially expanded and clarified to give the manuscript a concrete engineering backbone. These now include:
   • an overall framework figure (Figure 1),
   • a complete ultra-fast charging system schematic (Figure 2),
   • an explicit DAB topology figure (Figure 3),
   • a workflow and validation figure (Figure 4),
   • simulation-based result figures (Figures 5 and 6),
   • and multiple structured tables (Tables 1–4).

We agree with the Reviewer that the original version lacked sufficient technical depth. We believe that the revised manuscript now addresses this concern in a concrete and visible way.

Comment 1

The title should be changed.

Response

We thank the Reviewer for this important suggestion. We fully agreed with this point and changed the title accordingly.

The original broad title was replaced with the more specific and technically focused title:

“Hybrid AI–Quantum Co-Design of a SiC-Based DAB Converter for Ultra-Fast EV Charging”

This revised title better reflects the actual content of the manuscript, the selected converter topology, and the case-study-oriented nature of the revised work. It also avoids overstating the scope beyond what is actually demonstrated in the manuscript.

Comment 2

The abstract should clearly highlight the originality of this work.

Response

We thank the Reviewer for this comment. In response, the abstract was completely rewritten.

The revised abstract now makes the originality of the work clearer in several ways:

1. it explicitly states that the work proposes a hybrid AI–quantum co-design framework for a SiC-based DAB converter;
2. it identifies the three main layers of the framework:
   • physical converter model,
   • AI surrogate-learning layer,
   • and quantum-assisted optimization layer;
3. it clarifies that the contribution is demonstrated through a representative 350 kW ultra-fast EV charging case study;
4. it states explicitly that the revised manuscript includes:
   • a complete system schematic,
   • an explicit converter topology,
   • a clarified workflow,
   • and simulation-based proof-of-concept results.

We believe that the originality of the work is now much more visible already at the abstract level.

Comment 3

The introduction should include a relevant and up-to-date literature review.

Response

We thank the Reviewer for this important suggestion. We agree that the original Introduction did not provide a sufficiently focused and up-to-date literature review.

To address this, the Introduction was substantially rewritten. In the revised version, the literature review is now organized more clearly around the following themes:

1. ultra-fast EV charging systems and charger constraints;
2. wide battery-voltage adaptability and 400-V / 800-V charging context;
3. intelligent power-electronics methods and AI-based surrogate modeling;
4. the remaining gap between charger-oriented literature, AI-driven prediction, and advanced search frameworks.

In addition, the literature base was refined and updated, including the incorporation of more targeted charging-related references. We also followed the Reviewer’s suggestion and used the proposed papers to strengthen the revised Introduction and references. Specifically, the revised literature review now explicitly includes and benefits from the following recommended works:

• Important Technical Considerations in Design of Battery Chargers of Electric Vehicles (Energies, 2021);
• A Comprehensive Review of Universal Fast-Charging Techniques for 400-V and 800-V Electric Vehicles (IEEE Access, 2025).

We believe that the Introduction is now much more relevant, focused, and current.

Comment 4

The methodology and approach should be clearly explained and demonstrated.

Response

We thank the Reviewer for this key comment. We fully agree that the original version did not explain the methodology with sufficient clarity.

This issue has been addressed by major restructuring and expansion of the manuscript. The revised methodology is now presented in a dedicated Materials and Methods section and includes the following concrete subsections:

• methodological overview and scope,
• overall charging-system architecture,
• selected DAB topology,
• physical and mathematical representation,
• multi-objective co-design problem formulation,
• AI surrogate-learning layer,
• quantum-assisted optimization layer,
• co-design workflow and validation procedure,
• and baseline strategy with evaluation metrics.

In addition, the methodology is now demonstrated through a specific engineering case study, rather than remaining only at a conceptual level. The manuscript now clearly connects the conceptual framework to:

• a selected converter topology,
• a charging architecture,
• a simulation-based data-generation procedure,
• a surrogate-model training process,
• a candidate-search stage,
• and a high-fidelity re-validation step.

We believe that the revised manuscript now explains and demonstrates the methodology much more clearly.

Comment 5

Results should be presented through simulations and/or experimental validation.

Response

We thank the Reviewer for this important point. We agree that the original version did not provide a sufficiently structured validation section.

In response, we introduced a dedicated Results section in the revised manuscript and explicitly framed the work as a simulation-based proof-of-concept study. The revised paper now includes:

• a simulation-based evaluation overview,
• converter-level performance trends,
• dynamic and charging-profile results,
• comparison against a conventional baseline workflow,
• and a summary of the main quantitative tendencies.

The simulation-based results are supported by:

• Figure 5, which summarizes representative converter-level quantitative tendencies;
• Figure 6, which compares the conventional fixed CC–CV charging strategy and the proposed adaptive DAB-aware charging profile;
• and Table 4, which summarizes the comparative engineering and workflow-level tendencies between the baseline and the proposed framework.

At the same time, we explicitly clarify in the manuscript that the present work remains a simulation-based proof-of-concept and does not claim hardware-level experimental validation. This limitation is now clearly acknowledged in the Discussion and Conclusions sections.

Comment 6

A discussion section should be included.

Response

We thank the Reviewer for this recommendation. A dedicated Discussion section has now been added to the revised manuscript.

The new Discussion section includes:

• interpretation of the hybrid co-design advantage,
• explanation of why the proposed workflow improves design-space exploration,
• engineering relevance for ultra-fast EV charging,
• and limitations of the present study.

This addition was made specifically to separate result reporting from interpretation and to improve the manuscript structure in line with the journal style.

Comment 7

A conclusion and future research directions should be provided.

Response

We thank the Reviewer for this important suggestion. A fully developed Conclusions section has now been included in the revised manuscript.

The revised Conclusions section summarizes:

• the proposed hybrid AI–quantum co-design framework,
• the selected SiC-based DAB ultra-fast charging case study,
• the main observed simulation-based tendencies,
• and the methodological contribution of the work.

In addition, the revised Conclusions section now explicitly states several directions for future work, including:

• experimental or hardware-in-the-loop validation,
• broader comparisons with classical optimization methods,
• and extension of the methodology toward full charging architectures beyond the isolated DAB stage.

Comment 8

The author should also include an overall schematic of the proposed ultrafast charging system.

Response

We thank the Reviewer for this very useful suggestion. This point has been fully addressed in the revised manuscript.

A complete overall schematic of the considered 350 kW ultra-fast EV charging system has now been added as Figure 2. The revised figure shows:

• the three-phase active front-end AC/DC stage,
• the stabilized DC-link,
• the SiC-based isolated DAB DC/DC charging stage,
• the battery-side interface,
• and the supervisory controller.

The text in Section 2.2 was also expanded to explain the role of the charging architecture and to clarify why the isolated DC/DC stage is the main subsystem targeted by the proposed co-design methodology.

Comment 9

The associated converter topologies are missing.

Response

We thank the Reviewer for identifying this important weakness in the original manuscript. We fully agree that the absence of a clearly selected and illustrated converter topology reduced the technical credibility of the original version.

To address this point, the revised manuscript now includes:

1. an explicit subsection titled “Selected SiC-Based DAB Converter Topology”;
2. a dedicated topology figure (Figure 3);
3. and an explanation of why the DAB topology was selected for the revised case study.

The revised manuscript now makes clear that the study is not built around a generic converter block, but around a concrete SiC-based DAB isolated DC/DC charging converter, whose main power-transfer variables are then used in the co-design formulation.

Comment 10

Some simulation or experimental results are required to validate the proposed idea.

Response

We thank the Reviewer for this comment. This concern has been directly addressed through the introduction of a dedicated simulation-based results section.

The revised manuscript now includes:

• simulation-based converter-level performance trends,
• dynamic charging-profile comparison,
• and comparative baseline-versus-proposed workflow results.

We also explicitly state that the present validation is simulation-based and should be interpreted as a proof-of-concept validation strategy rather than as full experimental confirmation. This clarification is now made consistently throughout the manuscript.

Comment 11

The references need to be improved.

Response

We thank the Reviewer for this important observation. We agree that the literature positioning of the original manuscript needed improvement.

In response, the references and literature integration were revised in the following ways:

1. the Introduction was rewritten to include a more relevant and up-to-date literature review;
2. the literature was reorganized around ultra-fast EV charging, charger constraints, AI-based surrogate modeling, and advanced search methods;
3. more targeted and recent charging-related references were added;
4. and the reference usage in the text was made more specific and informative.

We believe that the revised reference base is now significantly more relevant to the final focused scope of the manuscript.

Comment 12

The author may use the two articles below to strengthen the final version of the manuscript.

• Important Technical Considerations in Design of Battery Chargers of Electric Vehicles, Energies 2021.
• A Comprehensive Review of Universal Fast-Charging Techniques for 400-V and 800-V Electric Vehicles, IEEE 2025.

Response

We sincerely thank the Reviewer for these highly relevant recommendations. We followed this suggestion and used these references to strengthen the revised manuscript, especially the Introduction and literature-positioning part.

These references were particularly useful because:

• the Energies 2021 paper helped improve the charger-design and engineering-constraint perspective of the revised Introduction;
• the IEEE Access 2025 paper helped strengthen the discussion of 400-V / 800-V fast-charging adaptability and the relevance of wide battery-voltage operating ranges in modern charging systems.

We are grateful to the Reviewer for these suggestions, which helped improve the final focus and technical grounding of the revised manuscript.

 

We sincerely thank Reviewer 3 again for the direct and highly constructive feedback. Although the comments were critical, they were extremely helpful in reshaping the manuscript into a more focused, technically explicit, and journal-appropriate form. We believe that the revised version is substantially improved in response to these remarks.

Round 2

Reviewer 2 Report

Comments and Suggestions for Authors

The resulting manuscript (with obvious not-declared AI - assist)  remained vague, see e.g. DFD fig 4 with "(examples)" when it comes to the specific models.  Author must re-think the whole paper and resubmit the human text.

Author Response

We thank the Reviewer for the direct comment. We understand the concern that the previous version of the manuscript still contained overly generic framework-style wording and did not sufficiently demonstrate that the proposed workflow was tied to a concrete engineering model.

In response, the manuscript has been carefully revised again with the specific aim of reducing generic and AI-like phrasing and improving technical specificity. The text has been rewritten to present the study as a simulation-based proof-of-concept for one clearly defined application: a modular 350 kW ultra-fast EV charging system based on four parallel 87.5 kW SiC-based DAB converter modules.

The following changes were made:

1. The manuscript was re-focused on the selected SiC-based DAB converter topology rather than on a broad abstract framework.

2. The system architecture and converter topology were clarified through explicit figures and parameter tables. The revised text now defines the DC-link voltage, battery-side voltage, module power, switching frequency, transformer ratio, transfer inductance, output filtering parameters, controller gains, and thermal constraints.

3. The previous vague “example”-style wording in the workflow description has been removed or replaced with specific model elements, variables, and validation steps used in the DAB case study.

4. The AI layer is now described specifically as a surrogate-learning block trained on simulation-generated DAB converter data, with defined input variables and output performance indicators.

5. The optimization layer is now framed more cautiously as a quantum-assisted or quantum-inspired search component within a hybrid workflow, without claiming hardware-level quantum advantage or universal superiority.

6. The comparison against the baseline strategy has been clarified as a simulation-based comparison with a conventional CC–CV/classical tuning workflow.

7. The manuscript language has been manually revised throughout to improve readability and reduce generic AI-generated phrasing.

Regarding the Reviewer’s concern about AI assistance, we clarify that any language-support tool was used only for editorial polishing and clarity improvement. The technical content, methodology, simulations, interpretation, and conclusions remain the responsibility of the author. A disclosure statement has been added to the manuscript in accordance with the journal policy.

We hope that the revised manuscript now presents a clearer, more specific, and more technically grounded contribution.

Reviewer 3 Report

Comments and Suggestions for Authors

The paper is well revised and improved. However, it is necessary to provide the actual numerical values used in this research work, especially for Figures 2 and 3, as well as Tables 1 and 2.

The authors must clearly indicate all parameter values for at least one specific case — or for several representative cases — for example, for a 400 V or 800 V battery system. In addition, the values of all components shown in Figures 2 and 3 should be explicitly provided.

Author Response

We sincerely thank the Reviewer for the positive assessment of the revised manuscript and for this important and constructive comment.

We fully agree that providing explicit numerical values for a well-defined representative case is essential for technical clarity and reproducibility. In response, the manuscript has been further clarified to ensure that all relevant parameter values are explicitly stated and consistently linked to the figures and tables.

The following improvements have been made:

1. Explicit definition of representative operating cases  
The manuscript now clearly distinguishes between:
- a nominal full-power 800 V battery case (V_dc = 900 V, V_bat = 800 V, P_out = 350 kW), and  
- a current-limited 400 V battery case (P_out ≤ 200 kW).  

These cases are explicitly described in Section 2.2 and are used consistently throughout the results section.

2. Complete parameter specification  
All main electrical, control, and thermal parameters used in the study are now explicitly summarized in Table 1. This includes:
- DC-link voltage, battery voltage, power levels, and currents  
- switching frequency, transformer ratio, and inductances  
- filter components, controller gains, and thermal limits  

3. Converter-level parameter clarification  
The parameters of the SiC-based DAB converter shown in Figures 2 and 3 are now fully specified and linked to Table 1 and Table 2. In particular, the revised manuscript clearly provides values for:
- transfer/leakage inductance  
- output filter inductance and capacitance  
- DC-link capacitance  
- switching frequency and dead time  
- controller gains (K_p, K_i)  

4. Nominal validation parameter set  
For clarity, a specific validated parameter set is now explicitly stated in the Results section (Section 3), including:
L_k = 9.5 µH, f_s = 55 kHz, n = 1.0, d_t = 150 ns, K_p = 0.75, K_i = 140,  
which corresponds to the nominal 800 V validation case with 350 kW output power.

5. Distinction between design space and validation values  
To avoid ambiguity, the manuscript now clearly distinguishes between:
- the broader design/simulation envelope (Tables 1 and 2), and  
- the specific parameter values used for representative validation cases.

We believe that these revisions address the Reviewer’s concern and ensure that all figures, tables, and results are now fully supported by explicit and traceable numerical parameter definitions.

We thank the Reviewer again for this valuable suggestion, which significantly improved the clarity and technical completeness of the manuscript.

Round 3

Reviewer 2 Report

Comments and Suggestions for Authors

I have reexamined the manuscript and still below the academic standards. Moreover, chatGPT usage in text  and DFD preparation must be  formally declared  

Author Response

We would like to sincerely thank the reviewer for the careful re-evaluation of the manuscript and for the valuable critical remarks. The comments significantly contributed to improving both the technical clarity and the academic quality of the revised version.

In response to the reviewer’s concerns regarding academic rigor, the manuscript has been substantially revised and expanded. The revised version now includes:

• a clarified and technically structured co-design methodology;
• an explicit SiC-based DAB converter topology and modular 350 kW charging architecture;
• formalized multi-objective optimization formulation;
• surrogate-model training and quantitative validation metrics;
• representative optimization convergence analysis;
• comparative efficiency, thermal, waveform, and ripple-performance evaluation;
• sensitivity analysis of the principal design variables;
• additional figures, tables, and quantitative engineering indicators supporting the reported trends.

Special attention was given to improving methodological transparency and reducing excessive conceptual narration. Several sections were substantially rewritten in a more concise engineering-oriented style, while the overall workflow was reorganized to improve technical coherence and reproducibility.

Regarding the reviewer’s comment on AI-assisted content generation, a formal “Declaration on the Use of AI-Assisted Tools” section has now been included in the revised manuscript. The declaration explicitly clarifies that AI-assisted tools were used only for limited language refinement, readability improvement, and partial support in conceptual workflow figure drafting, while all scientific concepts, simulation procedures, engineering analysis, numerical evaluation, interpretation of results, and conclusions were independently developed, verified, and approved by the author.

In addition, the revised manuscript explicitly clarifies that the presented results correspond to a simulation-based proof-of-concept investigation rather than to experimental hardware validation or claims of industrial quantum-computing deployment. The scope and limitations of the study are now stated more clearly in the revised “Limitations” section.

We believe that these revisions have substantially strengthened the technical depth, engineering relevance, methodological clarity, and academic consistency of the manuscript. We are grateful for the reviewer’s constructive comments, which helped significantly improve the overall quality of the work.