Estimation Wind Energy Potential Using Artificial Neural Network Model in West Lampung Area

International Conference on Science, Infrastructure Technology and Regional Development

自然科学（总论）

Putro, W.S.^1^2 ; Prahmana, R.A.^3 ; Yudistira, H.T.^3 ; Darmawan, M.Y.^4 ; Triyono, D.^2 ; Birastri, W.^1^2

Department of Atmospheric and Planetary Sciences, Institut Teknologi Sumatera (ITERA), Lampung, Sumatera

35365, Indonesia^1

Center of Meteorology, Climatology and Geophysics, Institut Teknologi Sumatera (ITERA), Jati Agung, Lampung

35365, Indonesia^2

Department of Mechanical Engineering, Institut Teknologi Sumatera (ITERA), Lampung, Sumatera

35365, Indonesia^3

Department of Physics, Institut Teknologi Sumatera (ITERA), Lampung, Sumatera

35365, Indonesia^4

关键词: Artificial neural network modeling;    Levenberg-Marquardt algorithm;    Local characteristics;    Multi-layer perceptron architecture (MLP);    Observation data;    Root mean square errors;    Wind directions;    Wind energy development;   

Others  :  https://iopscience.iop.org/article/10.1088/1755-1315/258/1/012006/pdf
DOI  :  10.1088/1755-1315/258/1/012006

学科分类：自然科学（综合）

来源: IOP

【 摘 要 】

Wind energy is a famous green energy after solar energy realized. The highest potential green energy especially wind energy is depending from local characteristic study. However, survey and observation of local characteristic study it's very expensive to obtain wind potential energy. Thus, in this study aimed to estimate wind potential energy using Artificial Neural Network (ANN) over Krui, West Lampung, Sumatera, Indonesia. Here, the observation data such as wind speed, wind direction, and elevation taken from local survey while the wind potential energy taken from NASA LaRC POWER project. Here, all the data was processed using Levenberg Marquardt algorithm and ANN back propagation to estimate wind energy potential for Multilayer Perceptron (MLP) architecture with Root Mean Square Error (RMSE) and Variance Accounted For (VAF) wind energy potential energy value less than 0.04% and 95%, respectively. Based on the result, we have recommedation model to estimate wind energy potential for wind energy development over Krui, West Lampung in near future.