Digitalization and artificial intelligence for monitoring and optimization of building-integrated photovoltaic systems

The integration of photovoltaic modules into building envelopes like rooftops, façade and glazing systems is shown to be an effective solution to achieve sustainable energy generation in smart buildings: the Building-Integrated Photovoltaic system. The variability of environmental conditions, partial shading, temperature variations, mismatch losses, and dynamic building energy demand, however, play a significant role on the performance of BIPV systems. In real operating conditions, BIPV systems are nonlinear and stochastic, which can make it difficult to control and optimize the systems. In this context, the applications of Artificial Intelligence, Machine Learning and optimization techniques have become much highlighted to enhance the operational intelligence, reliability and energy efficiency of BIPV infrastructures. For the literature review, the PRISMA method is used in this review to achieve a comprehensive and systematic analysis of machine learning and optimizations techniques applied for smart BIPV systems between 2015 and 2025. The review critically discusses the use of AI in the design of BIPV systems, selection of materials and modules, optimization of design parameters, use of intelligent Maximum Power Point Tracking, energy predictions and building energy management, and fault detection and predictive maintenance. They are compared in terms of prediction accuracy, optimization capability, scalability and practical application with various AI techniques such as Artificial Neural Networks, Deep Learning, Support Vector Machines, Fuzzy Logic Controllers, Genetic Algorithms, and Particle Swarm Optimization. New technologies such as Digital Twins, Internet of Things based monitoring, Cloud and Edge intelligence for intelligent BIPV infrastructures are also addressed in the review. In addition, main research challenges and future research opportunities that can be associated to AI-enabled BIPV systems are systematically identified.