978-1-6654-3613-7/21/$31.00 ©2021 IEEE Artificial Intelligence (AI) in Renewable Energy Systems: A Condensed Review of its Applications and Techniques Jeffrey T. Dellosa Department of Elecronics Engineering Caraga State University Butuan City, Philippines jtdellosa@carsu.edu.ph Eleonor C. Palconit College of Engineering and Architecture Ateneo de Davao University Davao City, Philippines ecpalconit@addu.edu.ph Abstract— This paper's main objective is to examine the state of the art of artificial intelligence (AI) techniques and tools in power management, maintenance, and control of renewable energy systems (RES) and specifically to the solar power systems. The findings would allow researchers to innovate the current state of technologies and possibly use the standard and successful techniques in building AI-powered renewable energy systems, specifically for solar energy. Various peer-reviewed journal articles were examined to determine the condition and advancement of the AI techniques in the field of RES, specifically in solar power systems. Different theoretical and experimental AI techniques often used and reliable techniques determined were the Artificial Neural Network (ANN), Backpropagation Neural Network (BPNN), Adaptive Neuro- Fuzzy Inference System (ANFIS), and Genetic Algorithm (GA). These techniques are widely used in different types of solar predictions based on the findings of this review. However, ANN stood out as the best of these techniques. ANN's specific advantages over its competition include short computing time, higher accuracy, and generalization capabilities over other modeling techniques. This would translate to cost efficiency over other modeling techniques. Keywords— Artificial Intelligence (AI), Renewable Energy Systems (RES), Solar PV systems I. INTRODUCTION The use of renewable energy systems (RES) to supplement energy insufficiency was demonstrated, according to previous studies [1-7]. Renewable energy, unlike fossil fuels, deliver clean energy to households with a minimal carbon footprint. More importantly, several studies have shown the use of RES reduces greenhouse gases (GHGs) in the atmosphere [8-10]. Renewable energy systems (RES) include systems that generate energy from environmental sources. Sources include the sun (solar), water (hydroelectric), heat derived from the grounds of the earth (geothermal), wind, plants (biomass), waves from the ocean, and even the temperature difference of the sea or the ocean (ocean thermal energy conversion or otherwise known as OTEC). Solar photovoltaic (PV) energy is a renewable energy source that captures available sunlight and converts them to electric power. Solar is widely used in both developed and developing countries in the world, as reported by several authors [11-13]. There are at least 402 Gigawatts of combined solar capacity for all countries in the world as of the end of 2017 [14]. China has the highest number of solar power deployments among all developed countries [15]. These massive deployments in solar power in developed countries are also being translated as a solution to the problems being encountered by those countries with no access to electricity, especially in rural areas. A. Environmental, social, and technical impact of RES Several studies have already been done on the RES’ environmental, social and technical feasibility, making RES a viable option for use going forward to the future [16-23]. Akella, et al. thoroughly discussed the technical and environmental impact of the deployment of renewable energy systems. The advantages and benefits of using clean energy systems compared to the traditional oil, natural gas, or coal- based energy generation systems were presented. In this case study, the renewable energy systems were installed to supply clean energy to the unelectrified villages [24]. B. The Need for AI in RES The use of RES has also evolved and advanced. Still, there exist opportunities with regards to the use of smart and highly intelligent systems, particularly the application of artificial intelligence (AI) to address the present evolving challenges and opportunities in renewable energy and solar photovoltaic systems' power management, maintenance, and control [25- 29]. AI has several applications in broad scope or area of concern that includes not only the energy sector but in food, agriculture, education, health & safety, and even in business and the art, among others [30]. Before AI, there exist basic decision systems in renewable energy that includes data acquisition and monitoring systems [31-34]. The review's main objective is to examine state of the art using artificial intelligence (AI) techniques and tools in power management, maintenance, and control of renewable energy and specifically to the solar power systems. The last review related to AIs [35]. This study's findings shall allow researchers to innovate the current state of technologies and possibly use the standard and successful techniques in building AI-powered renewable energy systems. II. METHODOLOGY A literature review from highly regarded journals and reviews of the reports related to the study in the last ten years was conducted to draw meaningful and significant results from artificial intelligence (AI) studies, specifically in the field of power management, maintenance, and control of solar energy systems. This study covered the current trends in AI as applied in renewable energy and specifically to solar energy systems. This study considered the different techniques in using artificial intelligence to improve solar energy systems' performance and efficiency. Explored in this literature review, are those AI techniques and descriptive analysis of renewable energy techniques, such as solar energy. III. RESULTS AND DISCUSSIONS A. AI Applications in General AI is a computer science and engineering field that focuses on creating smart or intelligent machines, devices,