Artificial Intelligence Techniques for Estimation of Optimum Weibull Parameters for Wind Speed Distribution
N. Natarajan1, S. Dinesh Kumar2, M. Shyam Sundar3, M. Santhosh Kumar4
1N. Natarajan, Department of Civil Engineering, Dr. Mahalingam College of Engineering and Technology, Pollachi (Tamil Nadu), India.
2S. Dinesh Kumar, Department of Civil Engineering, Dr. Mahalingam College of Engineering and Technology, Pollachi (Tamil Nadu), India.
3M. Shyam Sundar, Department of Civil Engineering, Dr. Mahalingam College of Engineering and Technology, Pollachi (Tamil Nadu), India.
4M. Santhosh Kumar, Department of Civil Engineering, Dr. Mahalingam College of Engineering and Technology, Pollachi (Tamil Nadu), India.
Manuscript received on 27 November 2019 | Revised Manuscript received on 16 December 2019 | Manuscript Published on 31 December 2019 | PP: 150-156 | Volume-8 Issue-4S2 December 2019 | Retrieval Number: D10331284S219/2019©BEIESP | DOI: 10.35940/ijrte.D1033.1284S219
Open Access | Editorial and Publishing Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: The main objective of this study is to estimate the optimum Weibull scale and shape parameters for wind speed distribution at three stations of the state of Tamil Nadu, India using Nelder-Mead, Broyden–Fletcher–Goldfarb–Shanno, and Simulated annealing optimization algorithms. An attempt has been made for the first time to apply these optimization algorithms to determine the optimum parameters. The study was conducted for long term wind speed data (38 years), short term wind speed data (5 years) and also with single year’s wind speed data to assess the performance of the algorithm for different quantum of data. The efficiency of these algorithms are analyzed using various statistical indicators like Root mean square error (RMSE), Correlation coefficient (R), Mean absolute error (MAE) and coefficient of determination (R2). The results suggest that the performance of three algorithms is similar irrespective of the quantum of the dataset. The estimated Weibull parameters are almost similar for short term and long term dataset. There is a marginal variation in the obtained parameters when only single year’s wind data is considered for the analysis. The Weibull probability distribution curve fits very well on the wind speed histogram when only single year’s wind speed data is considered and fits marginally well when short term and long term wind speed data is considered.
Keywords: Optimum Weibull Parameters Probability Distribution Curve Nelder-Mead, Broyden–Fletcher–Goldfarb–Shanno, Simulated Annealing.
Scope of the Article: Civil and Environmental Engineering