Digital Twins offer significant potential for assessing and enhancing reliability of energy systems by enabling continuously-updated models of reliability-relevant system behavior, which can be used to predict failures and identify critical components. While our previous and ongoing research demonstrates the capability of Digital Twins to discover reliability models from system data, the use of these models to improve system reliability remains underexplored. This paper aims to address this research gap by presenting a methodology and a proof-of-concept use case of a Digital Twin for a compact photovoltaic system, using Fault Trees as underlying models. The Digital Twin is used to identify components’ importance with respect to the overall reliability of the system, inform maintenance strategies, and guide decisions to enhance system reliability while considering maintenance expenditures. Using data from system faults and failures, this study builds upon previous work on discovery of reliability models and applying them to a real-world system to make decisions that enhance system reliability.
Mostafa et al. (Wed,) studied this question.