Federated digital twins and explainable AI for sustainable urban energy end-use: governance, trust, and equity

Urban energy end-use is becoming harder to manage as cities electrify transport, heat, and buildings while integrating renewables, distributed energy resources, storage, and flexible demand. Consumption patterns are now shaped by households, workplaces, mobility, and the built environment interacting in real time, raising challenges of intermittency, coordination, scalability, interoperability, cybersecurity, data governance, and demand-side decision-making. Centralized management is often inadequate when energy-use data are distributed across households, utilities, transport, and buildings, and when privacy rules limit data sharing. Federated learning, digital twins, edge intelligence, and explainable AI (XAI) are emerging as complementary responses for understanding and steering end-use: federated learning enables collaborative demand forecasting and optimization without exchanging sensitive consumption data; digital twins support continuous monitoring and simulation of buildings, districts, and energy assets; and XAI improves the interpretability, accountability, and public acceptance of demand-side decisions. Gaps remain in integrating these technologies, validating digital twins in dynamic consumption settings, ensuring interoperability and robustness, and assessing the environmental, social, and cybersecurity implications of managing urban energy demand.These technologies also sit within live debates about how cities and energy systems shape one another: whether federated architectures genuinely redistribute control and data sovereignty over consumption data or relocate existing power asymmetries; whether explainability delivers real accountability and contestability for households and consumers or merely legitimises automated demand management; what and whose realities digital twins encode when energy use is governed through models; and whether efficiency-driven demand management narrows or widens distributional inequality in energy access and affordability. This connects to wider critiques of technocratic “smart city” visions and calls for citizen-centric, socially accountable approaches to how cities consume energy.This Research Topic advances methods, architectures, and applications combining federated intelligence, digital twins, and XAI to improve the efficiency, resilience, transparency, and sustainability of urban energy end-use, while examining the governance, policy, managerial, and citizen-facing dimensions of demand-side management. Contributions may test whether integrated federated digital-twin frameworks improve demand forecasting, demand response, flexibility, building and district energy performance, and carbon reduction versus centralized or non-interpretable approaches, and evaluate their consequences for data governance, fairness, security, and public trust. Multidisciplinary work including horizon scanning, scenario building, and analysis of the power relations and everyday experiences around urban energy use is especially welcome, alongside technical validation in realistic end-use settings, in support of SDGs 3, 6, 7, 9, 11, and 12.Themes:• Federated learning for urban energy demand management• Digital-twin architectures for buildings, districts, and end-use energy systems• Explainable AI for demand-side management and sustainable cities• Privacy-preserving energy-use forecasting and analytics• AI-enabled demand response and load management• Flexible demand and its integration with renewable supply• Federated intelligence for distributed energy resources at the point of use• Digital-twin-based optimisation of building and district energy performance• Governance, policy, and regulatory frameworks for AI-driven demand management• Data governance, ownership, and sovereignty over consumption data• Citizen engagement, participation, and co-design in smart energy use• Energy justice, equity, and distributional impacts on energy access and affordability• Human-centric, trustworthy, and responsible AI (accountability, contestability)• Electric-vehicle charging, vehicle-to-grid, and demand-side flexibility• Carbon-aware and net-zero demand management• Edge, cloud, and energy-efficient AI for managing urban energy use• Cybersecurity, blockchain, and privacy in end-use energy data• Real-time monitoring and control of building and district energy assets• Multi-agent systems for demand-side coordination• Climate-resilient and sustainable urban energy infrastructure• Validation, benchmarking, and interoperability of integrated end-use frameworks• Social, economic, and policy implications of intelligent urban energy-use systemsAuthor note: every contribution should make explicit its relevance to urban governance or urban sustainability, and its connection to the intersection of technology and urban energy end-use. Topic editors: Aparna Kumari, Riaz Ullah Khan Source: https://www.frontiersin.org/research-topics/84109/federated-digital-twins-and-explainable-ai-for-sustainable-urban-energy-end-use-governance-trust-and-equity