「AI駆動の都市全体エネルギーガバナンスのための状況認識アーキテクチャとリスク優先マルチドローン検証」のためのデータ・再現性支援成果物

Unofficial AI-generated summary based on the public title and abstract. Not an official translation.

🔬研究者:AIとエネルギーガバナンスの統合アーキテクチャの設計と評価方法を参考にできる。

🏢実務担当者:都市エネルギー管理における異常検知とドローン検証の実装可能性を評価するための指標を提供する。

🏛政策担当者:スマートシティ政策やエネルギーインフラのレジリエンス強化策を検討する際の参考になる。

This repository provides the supporting research artifacts for the study entitled “An AI-Driven Situational Awareness Architecture for Citywide Energy Governance with Risk-Prioritized Multi-Drone Verification.” The study proposes a citywide AI-driven situational awareness architecture that integrates heterogeneous urban energy sensing, artificial intelligence, digital twin simulation, risk-based prioritization, multi-drone physical verification, and governance controls into a closed-loop decision-support framework. The research follows a Design Science Research (DSR) approach and evaluates the proposed architecture through analytical scenario modelling using a representative 100 km² mixed-use urban environment with approximately 4,500 monitored energy nodes . The validation considers low-, medium-, and high-risk disturbance scenarios and evaluates the feasibility of selective multi-drone activation under realistic UAV endurance constraints. The repository is intended to support transparency, reproducibility, and future extension of the analytical evaluation presented in the paper. It may include datasets, scenario parameters, calculation sheets, simulation outputs, figures, model definitions, and supporting analytical scripts used to examine: citywide monitored-node density and risk activation; risk-prioritized drone verification; multi-drone fleet requirements; drone fleet energy consumption; false anomaly confirmation reduction; Situational Awareness Index improvement; response latency reduction; and governance-oriented interpretation of verification results. The analytical results reported in the study indicate that selective drone activation requires approximately 3–12 drones across the modeled disturbance scenarios, while maintaining low fleet-level energy overhead. Drone-assisted verification reduces false anomaly confirmations from 16% to 6% , increases the Situational Awareness Index from 0.79 to 0.92 , and reduces modeled response latency from 50 minutes to 19 minutes . The repository is provided to facilitate independent inspection, replication of the analytical calculations, comparison with alternative situational awareness models, and future development of real-world pilot deployments involving smart meters, SCADA systems, EV charging infrastructure, public lighting, renewable energy assets, digital twins, and UAV-based verification.

• Zenodo https://zenodo.org/records/22215729first seen 2026-09-01 04:33:19 · last seen 2026-09-03 04:37:44