On Hybrid Health Monitoring of Photovoltaics

During the past few years, adaptation to renewable energy and utilizing solar power have attracted an exponentially increasing interest in interdisciplinary research domains as well as large-scale economies. However and due to many environmental factors, providing a reliable and fault tolerant control solution that can accomplish the main objectives of the power generation, even in the presence of faults, is an inevitable requirement. The main objective of this paper is to review the challenges of fault diagnosis of solar power systems and to present a hybrid and cloud-enabled architecture for a health monitoring system for photovoltaic (PV) farms, where both model-based and data-driven methods are utilized in a unified framework. The main focus of our proposed architecture is to explain the main components of a practical solution such that it can be easily integrated into currently available cloud technologies. This solution can significantly improve reliability of the new generations of power supply networks. The key enabler of our proposed solution is its modular structure; particularly, it can be augmented with any available or future control systems such as automated PV cleaning systems in order to provide a fully autonomous fault tolerant control solution that can (i) detect, (ii) localize, and (iii) rectify various types of faults in PV systems (such as shade faults). Furthermore, we discuss data privacy concerns and how our architecture addresses this issue.