Skip to main content

Power Systems Researchers of Royal Holloway University of London and Industry Collaborators have Developed a Novel Framework for Behind-the-Meter Energy Resources Forecasting

Power Systems Researchers of Royal Holloway University of London and Industry Collaborators have Developed a Novel Framework for Behind-the-Meter Energy Resources Forecasting

  • Date20 November 2026

Power Systems researchers of Royal Holloway University of London (RHUL) and their co-authors from TNEI Group, UK, and National Energy System Operator (NESO), UK, have developed a novel framework for behind-the-meter energy resources forecasting. Their study focused on solar photovoltaics and battery energy storage units installed at residential sites. The innovative framework named X-PiTCT is based on deep learning model embedded with power systems domain knowledge.

Datacollection

Fig. A: Illustration of a residential site with energy resources (Ref. doi: 10.1109/TIA.2026.3753921)

According to the Corresponding author (Dr Onyema Nduka, Associate Professor of Power Sustainability and Head of Power Systems Group at RHUL): "The X-PiTCT framework has useful application in energy resources forecasting and hence an enabler for flexibility services. Our framework promotes the deep decarbonisation of modern electricity networks. It demonstrates that combining deep learning (a subset of machine learning) with power systems domain knowledge improves forecasting performance while respecting engineering and physical constraints. The publication of this technical work by the prestigious IEEE Transactions on Industry Applications is a clear testament to the work’s novelty and scientific rigour. The Power Systems Group of Royal Holloway University of London remains committed to contributing to the UK and global energy systems decarbonisation agenda through relevant framework and tool development for current and future electricity networks." 

The first author, Mr Amir Adib (a PhD Researcher at RHUL) has this to say: "Delighted that our paper has been accepted in IEEE Transactions on Industry Applications. Building on our award-winning research first presented at IEEE icSmartGrid 2025 Glasgow, this milestone reflects the excellent guidance of my supervisor, Dr Onyema Nduka, alongside invaluable support from the Royal Holloway Doctoral School. Our work addresses an important challenge: standard deep learning and AI forecasts often predict net power exchange at prosumer’s point of connection to the utility’s mains meter that violate physical hardware limits. By embedding engineering rules directly into the deep learning model, our forecast is guaranteed to respect real-world constraints. Ultimately, I hope this framework helps grid operators safely integrate renewables for a sustainable future."

Co-author, Dr Wentao Zhu of NESO writes: “Drawing on my experience in power system engineering, machine learning, data centre grid integration and IEEE standards development, I believe collaboration between academia and industry is vital for bridging cutting-edge research with real-world operational challenges. This work demonstrates how AI-driven forecasting can improve the visibility of behind-the-meter energy resources, supporting a more flexible, intelligent and resilient electricity system.”

Another co-author, Dr Yueqi Yu of TNEI Group writes: “As a specialist energy consultancy, TNEI recognises the value of academic collaboration in developing innovative methodologies that enhance the quality and effectiveness of our technical studies. Improved forecasting of behind-the-meter energy resources can help reduce uncertainty in network modelling, capacity assessments and investment planning, enabling us to provide more robust, evidence-based advice to our clients.”

A. Adib, O. S. Nduka, Y. Wu and W. Zhu, "X-PiTCT: A Cross-Resolution Physics-Informed Temporal Convolutional Transformer for Behind-the-Meter Energy Resources Forecasting," in IEEE Transactions on Industry Applications, Available on Early Access (doi: 10.1109/TIA.2026.3753921).

Explore Royal Holloway