A Structured Analyse of Artificial Intelligence Techniques in IoT-Based Solar Photovoltaic Systems
Authors: Tewfik BENCHATTI, Abdeljalil HocineCHENAFI, khaledLAMINE
DOI: 10.37326/ajsev8.10/2057
Page No: 95-118
Abstract
The increasing deployment of photovoltaic (PV) systems has intensified the need for intelligent monitoring, diagnostics, and optimization solutions capable of operating under dynamic environmental and operational conditions. In this context, the integration of Artificial Intelligence (AI) techniques with Internet of Things (IoT) infrastructures has emerged as a promising paradigm for enhancing the performance, reliability, and autonomy of solar energy systems. This paper presents a structured and analytical review of AI-enabled IoT approaches applied to photovoltaic monitoring and management, with a particular focus on system architectures, data-driven intelligence, and operational functionalities. The review systematically categorizes existing studies according to their targeted applications, including real-time performance monitoring, fault detection and diagnosis, predictive maintenance, energy forecasting, and adaptive control mechanisms. Machine learning and deep learning techniques are analyzed in relation to their data requirements, computational complexity, and deployment layers within IoT-based PV systems. In addition, the role of collaborative experimental environments, such as OASIS Colab, is discussed as a practical support tool for developing, training, and validating AI models for photovoltaic data analysis in research-oriented settings. Beyond summarizing existing solutions, this paper identifies key technical challenges related to data quality, model generalization, scalability, and real-time implementation constraints. Finally, open research directions are outlined to guide future developments toward more robust, explainable, and scalable intelligent IoT architectures for photovoltaic energy systems.
