Cognitive Digital Twin Framework with Virtual Reality for Interactive Control of Solar PV Infrastructure

The rapid growth of photovoltaic (PV) installations demands intelligent, transparent, and resilient control frameworks that can adapt to fluctuating atmospheric and grid conditions. Conventional digital twin solutions, though effective in monitoring and forecasting, often fall short when facing sensor degradation, communication noise, or operator workload challenges. Their static architectures and narrowly scoped machine learning models typically lack both adaptive resilience and seamless integration of human oversight—critical requirements for mission-essential PV systems. To overcome these limitations, we introduce a Cognitive Digital Twin with Virtual Reality (CDT–VR), a framework that combines physics-informed models, deep learning surrogates, and reinforcement learning–based policy control. The design incorporates hybrid-fidelity monitoring, real-time drift detection, and immersive VR interfaces, enabling operators to validate or adjust system trajectories interactively through explainable overlays. This dual focus on algorithmic robustness and operator usability distinguishes CDT–VR from prior digital twin deployments in solar energy. Experimental results demonstrate forecasting accuracy with NRMSE as low as 8.9%, drift detection within 90 seconds, and P95 response latency of 382 ms, ensuring grid-compliant performance. Furthermore, operator studies using VR interfaces revealed a 27% reduction in task completion time and a 22% decrease in workload compared with conventional SCADA platforms. By unifying adaptive control with human-in-the-loop decision support, CDT–VR establishes a resilient pathway for digital twin deployment in renewable energy operations, with broader relevance to smart grids and critical cyber-physical infrastructures.