A Direct Iterative Technique for Weibull Parameters Estimation Optimized for lowest Root Mean Square Error

Document Type : Original Article

Author

Mechanical Eng. Depart., NRC

Abstract

Classical Weibull distribution (Weibull minima) is gaining momentum due to its employment in wind speed estimation for the growing renewable energy field. The main difficulty of Weibull distribution is that it is hard to accurately estimate its parameters for a set of wind speed data. Much research has been involved in this task with varying complexity, success, and accuracy. Statistical analysis is usually used to compare the accuracy of proposed Weibull parameters estimation methods. As the Weibull parameters of a set of data are not normally estimated, for simulation, more than once, this research considered the processing time to be of far less importance than accuracy. In this research, efforts were made to make the best use of the current PC processing power to achieve the best possible accuracy of the Weibull parameters. In the current research, an iterative technique is proposed to directly estimate the Weibull parameters that yield the best results with the most employed statistical analysis; least Root Mean Square Errors of estimation (RMSE). The results of the new technique were compared to that of the Maximum likelihood method (MLM), and a considerable improvement in accuracy, 9.3 % was achieved.

Keywords

Main Subjects

Files

Share

How to cite

Statistics