• Energy Research

Time series forecasting in the energy sector is important to power utilities for decision making to ensure the sustainability and quality of electricity supply, and the stability of the power grid. Unfortunately, the presence of certain exogenous factors such as weather conditions, electricity price complicate the task using linear regression models that are becoming unsuitable. The search for a robust predictor would be an invaluable asset for electricity companies. To overcome this difficulty, Artificial Intelligence differs from these prediction methods through the Machine Learning algorithms which have been performing over the last decades in predicting time series on several levels. This work proposes the deployment of three univariate Machine Learning models: Support Vector Regression, Multi-Layer Perceptron, and the Long Short-Term Memory Recurrent Neural Network to predict the electricity production of Benin Electricity Community. In order to validate the performance of these different methods, against the Autoregressive Integrated Mobile Average and Multiple Regression model, performance metrics were used. Overall, the results show that the Machine Learning models outperform the linear regression methods. Consequently, Machine Learning methods offer a perspective for short-term electric power generation forecasting of Benin Electricity Community sources.

Time series forecasting in the energy sector is important to power utilities for decision making to ensure the sustainability and quality of electricity supply, and the stability of the power grid. Unfortunately, the presence of certain exogenous factors such as weather conditions, electricity price complicate the task using linear regression models that are becoming unsuitable. The search for a robust predictor would be an invaluable asset for electricity companies. To overcome this difficulty, Artificial Intelligence differs from these prediction methods through the Machine Learning algorithms which have been performing over the last decades in predicting time series on several levels. This work proposes the deployment of three univariate Machine Learning models: Support Vector Regression, Multi-Layer Perceptron, and the Long Short-Term Memory Recurrent Neural Network to predict the electricity production of Benin Electricity Community. In order to validate the performance of these different methods, against the Autoregressive Integrated Mobile Average and Multiple Regression model, performance metrics were used. Overall, the results show that the Machine Learning models outperform the linear regression methods. Consequently, Machine Learning methods offer a perspective for short-term electric power generation forecasting of Benin Electricity Community sources.

The Weibull distribution function is essential for planning and designing wind-farm implementation projects and wind-resource assessments. However, the Weibull distribution is limited for those areas with high frequencies of calm winds. One solution is to use the hybrid Weibull distribution. In fact, when the wind speed data present heterogeneous structures, it makes sense to group them into classes that describe the different wind regimes. However, the single use of the Weibull distribution presents fitting errors that should be minimized. In this context, mixture distributions represent an appropriate alternative for modelling wind-speed data. This approach was used to combine the distributions associated with different wind-speed classes by weighting the contribution of each of them. This study proposes an approach based on mixtures of Weibull distributions for modelling wind-speed data in the West Africa region. The study focused on mixture Weibull (WW-BAY) and mixture hybrid Weibull (WWH-BAY) distributions using Bayes' theorem to characterize the wind speed distribution over twelve years of recorded data at the Abuja, Accra, Cotonou, Lome, and Tambacounda sites in West Africa. The parameters of the models were computed using the expectation-maximization (E-M) algorithm. The parameters of the models were estimated using the expectation-maximization (E-M) algorithm. The initial values were provided by the Levenberg-Marquardt algorithm. The results obtained from the proposed models were compared with those from the classical Weibull distribution whose parameters are estimated by some numerical method such as the energy pattern factor, maximum likelihood, and the empirical Justus methods based on statistical criteria. It is found that the WWH-BAY model gives the best prediction of power density at the Cotonou and Lome sites, with relative percentage error values of 0.00351 and 0.01084. The energy pattern factor method presents the lowest errors at the Abuja site with a relative percentage error value of -0.54752, Accra with -0.55774, and WW-BAY performs well for the Tambacounda site with 0.19232. It is recommended that these models are useful for wind energy applications in the West African region.

The Weibull distribution function is essential for planning and designing wind-farm implementation projects and wind-resource assessments. However, the Weibull distribution is limited for those areas with high frequencies of calm winds. One solution is to use the hybrid Weibull distribution. In fact, when the wind speed data present heterogeneous structures, it makes sense to group them into classes that describe the different wind regimes. However, the single use of the Weibull distribution presents fitting errors that should be minimized. In this context, mixture distributions represent an appropriate alternative for modelling wind-speed data. This approach was used to combine the distributions associated with different wind-speed classes by weighting the contribution of each of them. This study proposes an approach based on mixtures of Weibull distributions for modelling wind-speed data in the West Africa region. The study focused on mixture Weibull (WW-BAY) and mixture hybrid Weibull (WWH-BAY) distributions using Bayes' theorem to characterize the wind speed distribution over twelve years of recorded data at the Abuja, Accra, Cotonou, Lome, and Tambacounda sites in West Africa. The parameters of the models were computed using the expectation-maximization (E-M) algorithm. The parameters of the models were estimated using the expectation-maximization (E-M) algorithm. The initial values were provided by the Levenberg-Marquardt algorithm. The results obtained from the proposed models were compared with those from the classical Weibull distribution whose parameters are estimated by some numerical method such as the energy pattern factor, maximum likelihood, and the empirical Justus methods based on statistical criteria. It is found that the WWH-BAY model gives the best prediction of power density at the Cotonou and Lome sites, with relative percentage error values of 0.00351 and 0.01084. The energy pattern factor method presents the lowest errors at the Abuja site with a relative percentage error value of -0.54752, Accra with -0.55774, and WW-BAY performs well for the Tambacounda site with 0.19232. It is recommended that these models are useful for wind energy applications in the West African region.

In this article, we introduced a new approach based on graphical method (GPM), maximum likelihood method (MLM), energy pattern factor method (EPFM), empirical method of Justus (EMJ), empirical method of Lysen (EML) and moment method (MOM) using the even or odd classes of wind speed series distribution histogram with 1 m/s as bin size to estimate the Weibull parameters. This new approach is compared on the basis of the resulting mean wind speed and its standard deviation using seven reliable statistical indicators (RPE, RMSE, MAPE, MABE, R2, RRMSE and IA). The results indicate that this new approach is adequate to estimate Weibull parameters and can outperform GPM, MLM, EPF, EMJ, EML and MOM which uses all wind speed time series data collected for one period. The study has also found a linear relationship between the Weibull parameters K and C estimated by MLM, EPFM, EMJ, EML and MOM using odd or even class wind speed time series and those obtained by applying these methods to all class (both even and odd bins) wind speed time series. Another interesting feature of this approach is the data size reduction which eventually leads to a reduced processing time.Article History: Received February 16th 2018; Received in revised form May 5th 2018; Accepted May 27th 2018; Available onlineHow to Cite This Article: Salami, A.A., Ajavon, A.S.A., Kodjo, M.K. , Ouedraogo, S. and Bédja, K. (2018) The Use of Odd and Even Class Wind Speed Time Series of Distribution Histogram to Estimate Weibull Parameters. Int. Journal of Renewable Energy Development 7(2), 139-150.https://doi.org/10.14710/ijred.7.2.139-150