Predicting Maximum Wind Speed of Typhoons based on Convolutional Recurrent Neural Network via COMS Satellite Data

This study focuses on understanding the evolution of Hurricane Earl (2010) with respect to random perturbations in the storm’s initial strength, size, and asymmetry in wind distribution. We rely on the Unified Wave Interface-Coupled Model (UWIN-CM), a fully coupled atmosphere–wave–ocean system to generate a storm realization ensemble, and use polynomial chaos (PC) expansions to build surrogate models for time evolution of both the maximum wind speed and minimum sea level pressure in Earl. The resulting PC surrogate models provide statistical insights on probability distributions of model responses throughout the simulation time span. Statistical analysis of rapid intensification (RI) suggests that initial perturbations having intensified and counterclockwise-rotated winds are more likely to undergo RI. In addition, for the range of initial conditions considered RI seems mostly sensitive to azimuthally averaged maximum wind speed and asymmetry orientation, rather than storm size and asymmetry magnitude; this is consistent with global sensitivity analysis of PC surrogate models. Finally, we combine initial condition perturbations with a stochastic kinetic energy backscatter scheme (SKEBS) forcing in the UWIN-CM simulations and conclude that the storm tracks are substantially influenced by the SKEBS forcing perturbations, whereas the perturbations in initial conditions alone had only limited impact on the storm-track forecast.

Download Full-text

Joint Modeling of Severe Dust Storm Events in Arid and Hyper Arid Regions Based on Copula Theory: A Case Study in the Yazd Province, Iran

Climate ◽

10.3390/cli8050064 ◽

2020 ◽

Vol 8 (5) ◽

pp. 64 ◽

Cited By ~ 1

Author(s):

Tayyebeh Mesbahzadeh ◽

Maryam Mirakbari ◽

Mohsen Mohseni Saravi ◽

Farshad Soleimani Sardoo ◽

Nir Y. Krakauer

Keyword(s):

Wind Speed ◽

Natural Disasters ◽

Return Period ◽

Vertical Velocity ◽

Geopotential Height ◽

Dust Storms ◽

Maximum Wind Speed ◽

Near Surface ◽

Copula Theory ◽

Maximum Wind

Natural disasters such as dust storms are random phenomena created by complicated mechanisms involving many parameters. In this study, we used copula theory for bivariate modeling of dust storms. Copula theory is a suitable method for multivariate modeling of natural disasters. We identified 40 severe dust storms, as defined by the World Meteorological Organization, during 1982–2017 in Yazd province, central Iran. We used parameters at two spatial vertical levels (near-surface and upper atmosphere) that included surface maximum wind speed, and geopotential height and vertical velocity at 500, 850, and 1000 hPa. We compared two bivariate models based on the pairs of maximum wind speed–geopotential height and maximum wind speed–vertical velocity. We determined the bivariate return period using Student t and Gaussian copulas, which were considered as the most suitable functions for these variables. The results obtained for maximum wind speed–geopotential height indicated that the maximum return period was consistent with the observed frequency of severe dust storms. The bivariate modeling of dust storms based on maximum wind speed and geopotential height better described the conditions of severe dust storms than modeling based on maximum wind speed and vertical velocity. The finding of this study can be useful to improve risk management and mitigate the impacts of severe dust storms.

Download Full-text

Ensemble Recurrent Neural Network Based Probabilistic Wind Speed Forecasting Approach

Energies ◽

10.3390/en11081958 ◽

2018 ◽

Vol 11 (8) ◽

pp. 1958 ◽

Cited By ~ 20

Author(s):

Lilin Cheng ◽

Haixiang Zang ◽

Tao Ding ◽

Rong Sun ◽

Miaomiao Wang ◽

...

Keyword(s):

Neural Network ◽

Wind Speed ◽

Recurrent Neural Network ◽

Fuzzy Inference ◽

Ensemble Model ◽

Probabilistic Forecasting ◽

Inference System ◽

Wind Speed Forecasting ◽

Wind Power Integration ◽

Neuro Fuzzy

Wind energy is a commonly utilized renewable energy source, due to its merits of extensive distribution and rich reserves. However, as wind speed fluctuates violently and uncertainly at all times, wind power integration may affect the security and stability of power system. In this study, we propose an ensemble model for probabilistic wind speed forecasting. It consists of wavelet threshold denoising (WTD), recurrent neural network (RNN) and adaptive neuro fuzzy inference system (ANFIS). Firstly, WTD smooths the wind speed series in order to better capture its variation trend. Secondly, RNNs with different architectures are trained on the denoising datasets, operating as submodels for point wind speed forecasting. Thirdly, ANFIS is innovatively established as the top layer of the entire ensemble model to compute the final point prediction result, in order to take full advantages of a limited number of deeplearningbased submodels. Lastly, variances are obtained from submodels and then prediction intervals of probabilistic forecasting can be calculated, where the variances inventively consist of modeling and forecasting uncertainties. The proposed ensemble model is established and verified on less than one-hour-ahead ultra-short-term wind speed forecasting. We compare it with other soft computing models. The results indicate the feasibility and superiority of the proposed model in both point and probabilistic wind speed forecasting.

Download Full-text

The Use of Meteorology Data in Short-Term Prediction of Wind Speed for Wind Turbine Using Elman Recurrent Neural Network

2020 International Conference on Data Science, Artificial Intelligence, and Business Analytics (DATABIA) ◽

10.1109/databia50434.2020.9190628 ◽

2020 ◽

Author(s):

Riswan Dinzi ◽

Muhammad Yusuf ◽

Fahmi Fahmi

Keyword(s):

Neural Network ◽

Wind Speed ◽

Wind Turbine ◽

Recurrent Neural Network ◽

Short Term ◽

Term Prediction ◽

Short Term Prediction ◽

Elman Recurrent Neural Network

Download Full-text

Variability of Daily Maximum Wind Speed across China, 1975–2016: An Examination of Likely Causes

Journal of Climate ◽

10.1175/jcli-d-19-0603.1 ◽

2020 ◽

Vol 33 (7) ◽

pp. 2793-2816 ◽

Cited By ~ 1

Author(s):

Gangfeng Zhang ◽

Cesar Azorin-Molina ◽

Deliang Chen ◽

Jose A. Guijarro ◽

Feng Kong ◽

...

Keyword(s):

Wind Speed ◽

Large Scale ◽

Southern Oscillation ◽

Vertical Wind Shear ◽

Maximum Wind Speed ◽

The Arctic ◽

Risk Governance ◽

Daily Maximum ◽

Near Surface ◽

Maximum Wind

AbstractAssessing change in daily maximum wind speed and its likely causes is crucial for many applications such as wind power generation and wind disaster risk governance. Multidecadal variability of observed near-surface daily maximum wind speed (DMWS) from 778 stations over China is analyzed for 1975–2016. A robust homogenization protocol using the R package Climatol was applied to the DMWS observations. The homogenized dataset displayed a significant (p < 0.05) declining trend of −0.038 m s−1 decade−1 for all China annually, with decreases in winter (−0.355 m s−1 decade−1, p < 0.05) and autumn (−0.108 m s−1 decade−1; p < 0.05) and increases in summer (+0.272 m s−1 decade−1, p < 0.05) along with a weak recovery in spring (+0.032 m s−1 decade−1; p > 0.10); that is, DMWS declined during the cold semester (October–March) and increased during the warm semester (April–September). Correlation analysis of the Arctic Oscillation, the Southern Oscillation, and the west Pacific modes exhibited significant correlation with DMWS variability, unveiling their complementarity in modulating DMWS. Further, we explored potential physical processes relating to the atmospheric circulation changes and their impacts on DMWS and found that 1) overall weakened horizontal airflow [large-scale mean horizontal pressure gradient (from −0.24 to +0.02 hPa decade−1) and geostrophic wind speed (from −0.6 to +0.6 m s−1 decade−1)], 2) widely decreased atmospheric vertical momentum transport [atmospheric stratification thermal instability (from −3 to +1.5 decade−1) and vertical wind shear (from −0.4 to +0.2 m s−1 decade−1)], and 3) decreased extratropical cyclones frequency (from −0.3 to 0 month decade−1) are likely causes of DMWS change.

Download Full-text