Weibull Parameter Estimation Using Empirical and AI Methods: A Wind Energy Assessment in İzmir

Area of Science:

• Renewable Energy Systems
• Statistical Modeling
• Artificial Intelligence Applications

Background:

• Accurate wind speed modeling is crucial for effective wind energy potential assessments.
• Traditional empirical methods for Weibull distribution parameter estimation have limitations in complex wind data.
• Artificial Intelligence Optimization Algorithms (AIOAs) offer potential for enhanced accuracy.

Purpose of the Study:

• To compare the performance of traditional empirical methods with AIOAs for Weibull parameter estimation in wind speed modeling.
• To assess the techno-economic viability of wind energy sites based on accurate parameter estimation.
• To provide a decision-support tool for sustainable wind energy investments.

Main Methods:

• Estimation of Weibull distribution parameters (shape k, scale c) using empirical methods (JEM, PDM, EPFM, LAM, SEM) and AIOAs (GA, GSA, SCA, TLBA, GWA, RFA, RPA).
• Analysis of hourly wind speed data from four Turkish locations (Foça, Urla, Karaburun, Çeşme).
• Techno-economic analysis including capacity factors, unit energy costs, and payback periods.

Main Results:

• AIOAs, particularly GA, GSA, SCA, TLBA, and GWA, demonstrated superior performance over empirical methods, achieving low RMSE (0.0071) and high R² (0.9755).
• SEM and LAM showed competitive results among empirical methods; PDM and EPFM exhibited higher errors.
• Foça and Urla were identified as optimal investment sites, while Çeşme was deemed unviable.

Conclusions:

• AIOAs provide a more accurate framework for Weibull parameter estimation in wind energy assessments.
• The study offers a robust decision-support tool for selecting optimal wind energy investment locations.
• Accurate modeling significantly impacts the techno-economic feasibility of wind energy projects.