Wind speed forecasting based on Quantile Regression Minimal Gated Memory Network and Kernel Density Estimation

2019 ◽  
Vol 196 ◽  
pp. 1395-1409 ◽  
Author(s):  
Zhendong Zhang ◽  
Hui Qin ◽  
Yongqi Liu ◽  
Liqiang Yao ◽  
Xiang Yu ◽  
...  
Keyword(s):  
Wind Speed ◽  
Quantile Regression ◽  
Density Estimation ◽  
Kernel Density Estimation ◽  
Kernel Density ◽  
Wind Speed Forecasting ◽  
Memory Network

scholarly journals Probability Density Forecasting of Wind Speed Based on Quantile Regression and Kernel Density Estimation

Energies ◽  
2020 ◽  
Vol 13 (22) ◽  
pp. 6125
Author(s):  
Lei Zhang ◽  
Lun Xie ◽  
Qinkai Han ◽  
Zhiliang Wang ◽  
Chen Huang
Keyword(s):  
Wind Speed ◽  
Quantile Regression ◽  
Probability Density ◽  
Density Estimation ◽  
Kernel Density Estimation ◽  
Confidence Level ◽  
Kernel Density ◽  
Conditional Quantiles ◽  
Wind Speeds ◽  
Density Forecasting

Based on quantile regression (QR) and kernel density estimation (KDE), a framework for probability density forecasting of short-term wind speed is proposed in this study. The empirical mode decomposition (EMD) technique is implemented to reduce the noise of raw wind speed series. Both linear QR (LQR) and nonlinear QR (NQR, including quantile regression neural network (QRNN), quantile regression random forest (QRRF), and quantile regression support vector machine (QRSVM)) models are, respectively, utilized to study the de-noised wind speed series. An ensemble of conditional quantiles is obtained and then used for point and interval predictions of wind speed accordingly. After various experiments and comparisons on the real wind speed data at four wind observation stations of China, it is found that the EMD-LQR-KDE and EMD-QRNN-KDE generally have the best performance and robustness in both point and interval predictions. By taking conditional quantiles obtained by the EMD-QRNN-KDE model as the input, probability density functions (PDFs) of wind speed at different times are obtained by the KDE method, whose bandwidth is optimally determined according to the normal reference criterion. It is found that most actual wind speeds lie near the peak of predicted PDF curves, indicating that the probabilistic density prediction by EMD-QRNN-KDE is believable. Compared with the PDF curves of the 90% confidence level, the PDF curves of the 80% confidence level usually have narrower wind speed ranges and higher peak values. The PDF curves also vary with time. At some times, they might be biased, bimodal, or even multi-modal distributions. Based on the EMD-QRNN-KDE model, one can not only obtain the specific PDF curves of future wind speeds, but also understand the dynamic variation of density distributions with time. Compared with the traditional point and interval prediction models, the proposed QR-KDE models could acquire more information about the randomness and uncertainty of the actual wind speed, and thus provide more powerful support for the decision-making work.

Multivariate and Multimodal Wind Distribution Model Based on Kernel Density Estimation

2011 ◽  
Author(s):  
Jie Zhang ◽  
Souma Chowdhury ◽  
Achille Messac ◽  
Luciano Castillo
Keyword(s):  
Wind Speed ◽  
Density Estimation ◽  
Kernel Density Estimation ◽  
Wind Direction ◽  
Wind Farm ◽  
Kernel Density ◽  
Distribution Model ◽  
Wind Condition ◽  
Air Density ◽  
Wind Distribution

This paper presents a new method to accurately characterize and predict the annual variation of wind conditions. Estimation of the distribution of wind conditions is necessary (i) to quantify the available energy (power density) at a site, and (ii) to design optimal wind farm configurations. We develop a smooth multivariate wind distribution model that captures the coupled variation of wind speed, wind direction, and air density. The wind distribution model developed in this paper also avoids the limiting assumption of unimodality of the distribution. This method, which we call the Multivariate and Multimodal Wind distribution (MMWD) model, is an evolution from existing wind distribution modeling techniques. Multivariate kernel density estimation, a standard non-parametric approach to estimate the probability density function of random variables, is adopted for this purpose. The MMWD technique is successfully applied to model (i) the distribution of wind speed (univariate); (ii) the distribution of wind speed and wind direction (bivariate); and (iii) the distribution of wind speed, wind direction, and air density (multivariate). The latter is a novel contribution of this paper, while the former offers opportunities for validation. Ten-year recorded wind data, obtained from the North Dakota Agricultural Weather Network (NDAWN), is used in this paper. We found the coupled distribution to be multimodal. A strong correlation among the wind condition parameters was also observed.

Sign in / Sign up

Export Citation Format

Share Document