Comparison of methods in estimating Weibull parameters for wind energy applications

Tizgui, Ijjou; Guezar, Fatima El; Bouzahir, Hassane; Benaid, Brahim

doi:10.1108/ijesm-06-2017-0002

Cited by 42 publications

(29 citation statements)

References 18 publications

(19 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…These are the Weibull, Hybrid Weibull, Gamma, Rayleigh and inverse Gaussian distribution functions. Several methods are used to determine the distribution parameters [15,16,17]. The one used in this study to calculate the parameters of the different distribution functions is the maximum likelihood method because of its accuracy at our study site [4].The theoretical desciptions of probability density functions are discussed in the following sections.…”

Section: Modelling the Wind Speed Frequency Distributionmentioning

confidence: 99%

Assessing the Best Fit Probability Distribution Model for Wind Speed Data for Different Sites of Burkina Faso

Boro¹,

Thierry²,

Kieno³

et al. 2020

CJAST

View full text Add to dashboard Cite

In order to estimate the power output of a wind turbine, optimise its sizing and forecast the economic rate of return and risks of a wind energy project, wind speed distribution modelling is crucial. For which, Weibull distribution is considered as one of the most acceptable model. However, this distribution does not fit certain wind speed regimes. The objective of this study is to model the frequency distribution of the three-hourly wind speed at ten sites of Burkina Faso. In this context, we compared the accuracy of five distributions (Weibull, Hybrid Weibull, Rayleigh, Gamma and inverse Gaussian) which gave satisfactory results in this field. The maximum likelihood method was used to fit the distributions to the measured data. According to the statistical analysis tools (the coefficient of determination and the root mean square error), it was found that the Weibull distribution is most suited to the Bobo, Dédougou, Ouaga and Ouahigouya sites. On the other hand, for the sites of Bogandé, Fada and Po, the hybrid Weibull distribution is the most suitable one. As to the inverse Gaussian distribution, it is the most suitable for the Boromo, Dori and Gaoua sites. In addition, the analysis focused on comparing the mean absolute error of the annual wind power density estimation using the distributions examined. The Hybrid Weibull distribution was found to have a minimal mean absolute error for most study sites.

show abstract

Section: Modelling the Wind Speed Frequency Distributionmentioning

confidence: 99%

Assessing the Best Fit Probability Distribution Model for Wind Speed Data for Different Sites of Burkina Faso

Boro¹,

Thierry²,

Kieno³

et al. 2020

CJAST

View full text Add to dashboard Cite

show abstract

“…However, the selection of effective distributions is more important compared to the selection of parameter estimation methods (Tizgui, et al, 2017). In this work, the Maximum Likelihood method (ML) is applied.…”

Section: Estimation Of Distributions Parametersmentioning

confidence: 99%

Modeling of hospital admissions for respiratory diseases as a function of probability distribution functions

Souza

Santos

Oliveira‐Júnior

et al. 2020

RSD

View full text Add to dashboard Cite

Objective: To analyze the adjustments of the weibull, gamma, normal and logistic probability density distributions of the historical series of hospitalizations for respiratory diseases (childhood and adult pneumonia) from 2011 to 2015, in Campo Grande, MS. Methods: The shape and scale parameters of the distributions were determined to verify the quality of the data fit. Results: Four probability density functions (Table 2) were fitted and the R2, MAE, RSME, MAPE tests were used to verify the best density function for hospitalization data. Conclusion: The best fit was the Gamma distribution; the distribution can be used as an alternative distribution that adequately describes the data on hospital admissions for respiratory diseases in Campo Grande.

show abstract

“…A widely used probability distribution for underlying wind speed is the Weibull distribution (Sohoni et al, 2016) which is used by Celik (2003), for example. With respect to parameter estimation based on hourly wind speed data, Tizgui et al (2017) provide a comparison of different estimation methods. However, there are also other approaches.…”

Section: Literature Reviewmentioning

confidence: 99%

Model risk regarding monthly wind energy production for the valuation of a wind farm investment

Krömer

2019

IJESM

View full text Add to dashboard Cite

Purpose The purpose of this paper is to assess model risk with regard to wind energy output in monthly cash flow models for the purpose of valuation and risk assessment of wind farm investments, where only a few approaches exist in the literature. Design/methodology/approach This paper focuses on the risk-return characteristics of this investment from the perspective of private and institutional investors and takes into account several risks, in particular the resource risk related to the uncertainty of the monthly wind energy produced. To this end, this paper presents different approaches for modeling monthly wind power output and assesses the impact of three selected models with different properties on the investment’s risk-return characteristics by means of a stochastic discounted cash flow model. In addition, the model considers the possibility of a joint operation of the wind farm with a pumped hydro storage system to reduce risk and improve profits. Findings The results show that the (non-)consideration of seasonality of the monthly wind energy produced considerably influences the risk-return characteristics, but that principal developments dependent on input parameters and model variables remain similar. Originality/value This paper contributes to the literature by presenting different approaches for modeling the monthly wind energy produced based on direct models of the wind energy output, which are rare in the existing literature. Further, their impact on risk-return characteristics of a wind farm investment is analyzed, and thus, related model risk is assessed.

show abstract

Comparison of methods in estimating Weibull parameters for wind energy applications

Cited by 42 publications

References 18 publications

Assessing the Best Fit Probability Distribution Model for Wind Speed Data for Different Sites of Burkina Faso

Assessing the Best Fit Probability Distribution Model for Wind Speed Data for Different Sites of Burkina Faso

Modeling of hospital admissions for respiratory diseases as a function of probability distribution functions

Model risk regarding monthly wind energy production for the valuation of a wind farm investment

Contact Info

Product

Resources

About