Toward Resilient Explainable AI–Digital Twin Frameworks for Cyber-Secure Renewable Energy Systems: An Uncertainty-Aware Lifecycle-Integrated Architecture

The accelerating penetration of variable renewable energy sources into modern power systems demands digital infrastructures capable of supporting predictive, autonomous, and trustworthy grid operation. While digital twins (DT) and artificial intelligence (AI)—particularly multi-agent and agentic architectures—are key enablers, their integration remains fragmented across three critical dimensions: explainability, cybersecurity, and uncertainty management. This paper proposes a conceptual, uncertainty-aware, lifecycle-integrated architecture that couples explainable AI (XAI) with digital twin modeling to support resilient, cyber-secure renewable energy operations. Based on a structured review spanning component-, building-, and urban-scale DT infrastructure, XAI-driven forecasting and fault diagnosis, distributed energy resource (DER) multi-agent control, and DER cybersecurity vulnerabilities, we identify a persistent scholarly disjunction. Few studies treat interpretability, adversarial resilience, and stochastic uncertainty as co-design requirements across the entire asset lifecycle. Using a design-science methodology and thematic synthesis, we develop a five-layer architecture: sensing and data acquisition, digital twin simulation, XAI-mediated agentic decision-making, cybersecurity and resilience, and lifecycle governance. The core contribution lies in defining the structural logic through which these layers interoperate to achieve transparency, security, and adaptability. This framework provides a foundation for future simulation-based validation and advances debates on digitalized energy infrastructure governance.