Development of an enhanced parametric model for wind turbine power curve

•A new parametric model (MHTan) for wind turbine power curved is proposed.•Probability density function of the output power based on MHTan is obtained.•Parametric models yield better performance than do the nonparametric models.•Proposed MHTan model using BSA algorithm outperforms all other models....

Full description

Saved in:
Bibliographic Details
Published in:Applied energy 2016-09, Vol.177, p.544-552
Main Authors: Taslimi-Renani, Ehsan, Modiri-Delshad, Mostafa, Elias, Mohamad Fathi Mohamad, Rahim, Nasrudin Abd
Format: Article
Language:English
Subjects:
Adaptive neuro fuzzy inference system
Data mining
Estimates
Mathematical analysis
Mathematical models
Maximum likelihood estimation
Metaheuristic algorithm
Modelling
Parametric models
Power curve
Searching
Wind turbine
Wind turbines
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:•A new parametric model (MHTan) for wind turbine power curved is proposed.•Probability density function of the output power based on MHTan is obtained.•Parametric models yield better performance than do the nonparametric models.•Proposed MHTan model using BSA algorithm outperforms all other models. Modeling of wind turbine power curve is greatly important in performance monitoring of the turbine and also in forecasting the wind power generation. In this paper, an accurate parametric model called modified hyperbolic tangent (MHTan) is proposed to characterize power curve of the wind turbine. The paper also presents the development of both parametric and nonparametric models of wind turbine power curve. In addition, least square error (LSE) and maximum likelihood estimation (MLE) are employed to estimate vector parameter of the proposed model. Here, three evolutionary algorithms, namely, particle swarm optimization, Cuckoo search, and backtracking search algorithm aid LSE and MLE. The performance of all presented methods is evaluated by a real data collected from a wind farm in Iran as well as three statistically generated data sets. The results demonstrate the efficiency of the proposed model compared to some other existing parametric and nonparametric models.
ISSN:0306-2619
1872-9118
DOI:10.1016/j.apenergy.2016.05.124