IEEE PVSC 49
Search
SPLTRAK Abstract Submission
Intelligent cloud-based monitoring and control digital twin for photovoltaic power plants
Andreas Livera1, Georgios Paphitis1, Loucas Pikolos1, Ioannis Papadopoulos1, Javier Lopez-Lorente1, George Makrides1, Juergen Sutterlueti2, George E. Georghiou 1
1PV Technology Laboratory, FOSS Reseacrh Centre for Sustainable Energy, Department of Electrical and Computer Engineering, University of Cyprus, Nicosia, Cyprus
/2Gantner Instruments GmbH, Schruns, Austria

A main challenge in the scope of integrating higher shares of photovoltaic (PV) systems is to ensure optimal operations. This can be achieved through next-generation monitoring with automatic data-driven functionalities. The purpose of this paper is to address the fundamental challenges of developing robust and accurate monitoring and control solutions for PV power plants. The proposed digital twin enables the digital-enhancement of PV power plants for real-time asset observability and control (use of robust grid-edge Linux devices and communicative interfaces to stream high-resolution to advanced cloud database), and acts as a high-level health-state monitor to timely prognose/diagnose failures. Moreover, the paper presents a novel unified approach for constructing high-performing digital twin health-state models based entirely on synthetic data. Overall, the results showcase that the proposed digital twin trained with actual datasets exhibited high predictive accuracies, approximately 2% given by the root mean square error (nRMSE) relative to the capacity of the system, at granularities of 30- and 60-min. Moreover, the application of synthetic data for training proved a robust alternative in the absence of actual data since the prediction error was 3.5%. Finally, the advanced digital twin is expected to have significant impact on the value chain of the technology given the reduction of PV electricity costs by increasing the lifetime output and enabling higher integration shares.