<![CDATA[Electricity is an energy that is highly demanded by all of mankind. In Indonesia, the consumption of electricity increases every year. Therefore, there is a need for power plants that can supply the increasing electricity demand year after year. In 2020, out of the 65,236 MW generated by power plants in Indonesia, a total of 90.75% of the electricity in Indonesia was still supplied by fossil fuel power plants. In 2021, PLN (State Electricity Company) experienced a coal supply crisis due to extreme weather conditions in coal mining areas, delays in the coal procurement process, and the impact of coal export prices. This coal supply crisis resulted in20 coal-fired power plants with a capacity of 10,850 MW being at risk of blackouts. This highlights the importance of renewable energy power plants to reduce dependence on fossil fuels. The government is also striving to achieve a 25% utilization of renewable energy by 2025, including solar power plants. In electricity production, solar power plants rely heavily on solar radiation that can be captured by solar panels. Solar radiation on the surface of solar panels is a fundamental parameter for designing a well-integrated photovoltage (PV) system, both for load requirements and determining the amount of electricity produced by the panels, as well as for accurate operational simulations. Therefore, AI is expected to be used to assist in theanalysis of solar radiation. AI has advantages in certain tasks, making it possible for computers to make accurate decisions that result in more efficient operations. AI is highly suitable for processing solar radiation data in a particular location, especially considering the years of collected solar radiation data that form big 2 data. The use of artificial intelligence and big data can analyze the data and provide faster insights compared to conventional mathematical calculations. By employing various deep learning algorithms such as vector autoregression (VAR) and long-short term memory (LSTM), the prediction of solar radiation can become more accurate, facilitating optimal analysis in the design of solar power plants for households and industries. The AIalgorithm used for solar radiation prediction in this study is a combination of VAR and LSTM algorithms. The accuracy rate achieved by the combination of VAR and LSTM algorithms in this research exceeds 90%, indicating that this combination is highly suitable for predicting future solar radiation. Keywords: solar power plant, artificial intelligence (AI), solar radiation prediction.]]>
Copyrights © 2023
