Wind Power Potential Assessment by Optimization Algorithms at Offshore and Coastal Sites using Reanalysis Data

Accurately assessing wind resources is vital for wind energy projects, this study proposes a methodology for evaluates wind resources at coastal and offshore locations in India through Weibull distribution parameter optimization. Employing a 22-year dataset from NASA MERRA-2 reanalysis, wind characteristics are analysed at a 50-meter height using the Weibull probability distribution. Three numerical techniques (Modified Maximum Likelihood Method, Empirical Method of Lysen, and Method of Moment) and two metaheuristic optimization algorithms (Particle Swarm Optimization and Genetic Algorithm) determine shape and scale parameters. Goodness of fit tests, RMSE, and MBE assess method accuracy. Results indicate Modified Maximum Likelihood Method as the most accurate for wind potential evaluation. The offshore wind industry's expansion necessitates reliable predictions, making this research crucial for estimating India's offshore wind resource using reanalysis data. Three numerical approaches are applied, with Maximum Likelihood Method standing out as the most precise for determining wind potential. The study contributes valuable insights into probabilistic modelling of wind resources and MERRA-2 utility. The findings underscore the role of early information in determining the feasibility of wind energy generation at specific coastal and offshore locations in India, providing a foundation for future offshore wind industry developments.