A REVIEW ON SOLAR IRRADIANCE FORECASTING MODELS FOR OPTIMAL PHOTOVOLTAICPOWER GENERATION USING ARTIFICIAL INTELLIGENCE

Increased energy demands and power consumption can be attributed to a number of factors, including global population inflation and changing consumer preferences. Other important causes for employing clean and renewable energy sources for generating power and increasing energy efficiency are the rapid depletion of fossil fuels, the alarming rise in air pollution, and global warming contributing to frequent natural disasters. Solar energy power generation is a consequential origin of renewable energy. Irradiance is defined as the amount of power received from the Sun in the form of electromagnetic radiation per unit area, as measured in the wavelength range of the measuring apparatus. The watts per sq meter (W/m2) metric is used to measure solar irradiance in SI units. and it performs a leading job in determining photovoltaic system’s power generation potential. Solar irradiance is monitored and assessed for a wide range of purposes, including predicting energy output from photovoltaic panels, climatology, and weather forecasting. This paper presents a general outlook of several tools and approaches for forecasting solar irradiance utilizing data-driven methods constructed on algorithms like machine learning and deep learning. Depending on the input information used to 'train' the algorithm, it may be used to a wide variety of forecast horizons. Although these algorithms have been proved to be capable of estimating solar radiation, variances in their performance make comparing and choosing the optimal technique an interesting task. Artificial neural network, SupportVector Machines, and their comparison with Deep Learning approaches are among the most often used ML methods in this work