Artificial Intelligence (AI) in relation to environmental life-cycle assessment, photovoltaics, smart grids and small-island economies

Considering the gaps in the literature on Artificial Intelligence (AI) modelling, this article aims to: i) present models that combine AI and environmental Life Cycle Assessment (LCA), ii) analyse the role of AI in photovoltaics, smart grids and small-island economies. The methodology used is based upon selection of publications and analysis. Regarding LCA/AI models, the results show that AI can anticipate environmental impacts but model performance depends on the amount of data available. LCA/AI models can be used for eco-design and decision-making. However, it is necessary to develop standardised methodologies to evaluate AI environmental impacts. Regarding AI and photovoltaics, AI provides remarkably interesting applications: design, optimisation and prediction of parameters related to different kinds of photovoltaics (concentrating, building-integrated, etc.). As for AI and smart grids, AI offers advantages such as integration of intermittent renewable energy sources and decentralised-grid management. With respect to AI and small-island economies, factors such as effective energy storage, energy plans and estimation of the degree of susceptibility to disasters are important. Generally speaking, and considering the above-mentioned issues, it can be argued that AI poses multiple challenges: machine-learning models on a large-scale basis; the internet of things; options to reduce negative environmental impacts and so on.