Selected work

Representative Papers

Architectures for Robust Self-Organizing Energy Systems under Information and Control Constraints

Apr 23, 2026

This work addresses the challenge of achieving robust self-organization in cyber-physical energy systems under information constraints, control limitations, and network-induced disturbances such as cyber attacks. The authors propose an agent-based approach grounded in an observer/controller architecture that integrates controlled self-organization mechanisms with observation-control techniques tailored for limited-information environments. This framework enables controllable self-organization while preserving privacy, ensuring regulatory compliance, and respecting communication constraints. For the first time, the study systematically evaluates multiple architectural designs adapted to such constrained scenarios and demonstrates their effectiveness in significantly enhancing system robustness under realistic operational limitations, thereby providing both theoretical foundations and practical design guidelines for real-world deployment.

0 citationsRead paper

Towards a Metadata Schema for Energy Research Software

Jan 14, 2026

This study addresses a longstanding gap in energy research: the absence of a domain-specific metadata schema that simultaneously adheres to FAIR4RS principles and meets practical user needs, thereby hindering the discoverability and reusability of scientific software. Through systematic requirements analysis, the authors design and implement the first metadata schema tailored specifically for energy research software, integrating formal specifications, interoperability standards, and domain-specific characteristics. User testing demonstrates the schema’s usability and confirms its capacity to effectively balance generality with domain specificity. Furthermore, the research highlights the critical role of information presentation in enhancing the quality of metadata creation, offering valuable insights for future metadata schema development in scientific domains.

0 citationsRead paper

Outperforming Dijkstra on Sparse Graphs: The Lightning Network Use Case

Sep 16, 2025

Routing efficiency in payment channel networks (e.g., the Lightning Network) remains suboptimal. Method: This work pioneers the application of the theoretically superior Bounded Multi-Source Shortest Path (BMSSP) algorithm to real-world sparse topologies, implementing it in Rust and rigorously evaluating its path-finding performance on empirical Lightning Network data. Through large-scale randomized experiments and statistical hypothesis testing, we compare BMSSP against classical Dijkstra’s algorithm. Results: Our implementation fails to achieve the expected theoretical speedup—observed acceleration is marginal—primarily due to implementation complexity and non-negligible constant-factor overhead. This study fills a critical gap in the empirical evaluation of BMSSP, uncovering key bottlenecks in translating asymptotic advantages into practical routing performance. It establishes a reproducible benchmark, provides empirical evidence for algorithmic design trade-offs, and identifies concrete optimization directions for efficient routing in payment channel networks.

0 citationsRead paper

SMECS: A Software Metadata Extraction and Curation Software

Jul 24, 2025

To address the time-consuming and labor-intensive process of constructing metadata for research software—which hinders FAIR (Findable, Accessible, Interoperable, Reusable) implementation—this paper introduces a lightweight metadata enhancement tool. The tool automatically harvests basic metadata from platforms such as GitHub via web crawling, and integrates a responsive web interface enabling visual validation, interactive editing, and semantic enrichment. It natively exports standards-compliant JSON-LD files adhering to the CodeMeta schema. Its key innovation lies in synergistically combining automated metadata acquisition with human-in-the-loop refinement, thereby substantially lowering technical barriers for domain researchers. Empirical usability evaluation demonstrates that the tool reduces metadata curation time by an average of 62% and achieves a user satisfaction rating of 4.7/5.0, significantly improving the findability, reusability, and interoperability of research software.

0 citationsRead paper

Coordination of Electrical and Heating Resources by Self-Interested Agents

Jun 19, 2025

This paper addresses the distributed coordinated scheduling of electric–thermal resources in multi-energy coupled systems dominated by self-interested agents. We propose a novel distributed hybrid optimization algorithm integrating gossip-based communication with heuristic local search. The method ensures agent privacy preservation and adherence to physical constraints of energy devices, while simultaneously optimizing individual economic objectives and system-wide global performance. To the best of our knowledge, it is the first approach enabling distributed co-optimization across multiple time steps and coupled energy carriers (electricity and heat). Evaluated on two representative test cases—pure power systems and gas-fired combined heat and power (CHP) systems—the algorithm converges to near-global optimal solutions: average agent revenue increases by 12%, and total system operational cost decreases by 8.3%, demonstrating both effectiveness and practical applicability.

0 citationsRead paper

Recent publications

Latest Papers

Architectures for Robust Self-Organizing Energy Systems under Information and Control Constraints

Apr 23, 2026

This work addresses the challenge of achieving robust self-organization in cyber-physical energy systems under information constraints, control limitations, and network-induced disturbances such as cyber attacks. The authors propose an agent-based approach grounded in an observer/controller architecture that integrates controlled self-organization mechanisms with observation-control techniques tailored for limited-information environments. This framework enables controllable self-organization while preserving privacy, ensuring regulatory compliance, and respecting communication constraints. For the first time, the study systematically evaluates multiple architectural designs adapted to such constrained scenarios and demonstrates their effectiveness in significantly enhancing system robustness under realistic operational limitations, thereby providing both theoretical foundations and practical design guidelines for real-world deployment.

0 citationsRead paper

Towards a Metadata Schema for Energy Research Software

Jan 14, 2026

This study addresses a longstanding gap in energy research: the absence of a domain-specific metadata schema that simultaneously adheres to FAIR4RS principles and meets practical user needs, thereby hindering the discoverability and reusability of scientific software. Through systematic requirements analysis, the authors design and implement the first metadata schema tailored specifically for energy research software, integrating formal specifications, interoperability standards, and domain-specific characteristics. User testing demonstrates the schema’s usability and confirms its capacity to effectively balance generality with domain specificity. Furthermore, the research highlights the critical role of information presentation in enhancing the quality of metadata creation, offering valuable insights for future metadata schema development in scientific domains.

0 citationsRead paper

Outperforming Dijkstra on Sparse Graphs: The Lightning Network Use Case

Sep 16, 2025

Routing efficiency in payment channel networks (e.g., the Lightning Network) remains suboptimal. Method: This work pioneers the application of the theoretically superior Bounded Multi-Source Shortest Path (BMSSP) algorithm to real-world sparse topologies, implementing it in Rust and rigorously evaluating its path-finding performance on empirical Lightning Network data. Through large-scale randomized experiments and statistical hypothesis testing, we compare BMSSP against classical Dijkstra’s algorithm. Results: Our implementation fails to achieve the expected theoretical speedup—observed acceleration is marginal—primarily due to implementation complexity and non-negligible constant-factor overhead. This study fills a critical gap in the empirical evaluation of BMSSP, uncovering key bottlenecks in translating asymptotic advantages into practical routing performance. It establishes a reproducible benchmark, provides empirical evidence for algorithmic design trade-offs, and identifies concrete optimization directions for efficient routing in payment channel networks.

0 citationsRead paper

SMECS: A Software Metadata Extraction and Curation Software

Jul 24, 2025

To address the time-consuming and labor-intensive process of constructing metadata for research software—which hinders FAIR (Findable, Accessible, Interoperable, Reusable) implementation—this paper introduces a lightweight metadata enhancement tool. The tool automatically harvests basic metadata from platforms such as GitHub via web crawling, and integrates a responsive web interface enabling visual validation, interactive editing, and semantic enrichment. It natively exports standards-compliant JSON-LD files adhering to the CodeMeta schema. Its key innovation lies in synergistically combining automated metadata acquisition with human-in-the-loop refinement, thereby substantially lowering technical barriers for domain researchers. Empirical usability evaluation demonstrates that the tool reduces metadata curation time by an average of 62% and achieves a user satisfaction rating of 4.7/5.0, significantly improving the findability, reusability, and interoperability of research software.

0 citationsRead paper

Coordination of Electrical and Heating Resources by Self-Interested Agents

Jun 19, 2025

This paper addresses the distributed coordinated scheduling of electric–thermal resources in multi-energy coupled systems dominated by self-interested agents. We propose a novel distributed hybrid optimization algorithm integrating gossip-based communication with heuristic local search. The method ensures agent privacy preservation and adherence to physical constraints of energy devices, while simultaneously optimizing individual economic objectives and system-wide global performance. To the best of our knowledge, it is the first approach enabling distributed co-optimization across multiple time steps and coupled energy carriers (electricity and heat). Evaluated on two representative test cases—pure power systems and gas-fired combined heat and power (CHP) systems—the algorithm converges to near-global optimal solutions: average agent revenue increases by 12%, and total system operational cost decreases by 8.3%, demonstrating both effectiveness and practical applicability.

0 citationsRead paper