English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

In this research paper, we study the capability of artificial neural networkmodels to emulate storm surge based on the storm track/size/intensity history,leveraging a database of synthetic storm simulations. Traditionally,Computational Fluid Dynamics solvers are employed to numerically solve thestorm surge governing equations that are Partial Differential Equations and aregenerally very costly to simulate. This study presents a neural network modelthat can predict storm surge, informed by a database of synthetic stormsimulations. This model can serve as a fast and affordable emulator for thevery expensive CFD solvers. The neural network model is trained with the stormtrack parameters used to drive the CFD solvers, and the output of the model isthe time-series evolution of the predicted storm surge across multiple nodeswithin the spatial domain of interest. Once the model is trained, it can bedeployed for further predictions based on new storm track inputs. The developedneural network model is a time-series model, a Long short-term memory, avariation of Recurrent Neural Network, which is enriched with ConvolutionalNeural Networks. The convolutional neural network is employed to capture thecorrelation of data spatially. Therefore, the temporal and spatial correlationsof data are captured by the combination of the mentioned models, the ConvLSTMmodel. As the problem is a sequence to sequence time-series problem, anencoder-decoder ConvLSTM model is designed. Some other techniques in theprocess of model training are also employed to enrich the model performance.The results show the proposed convolutional recurrent neural networkoutperforms the Gaussian Process implementation for the examined syntheticstorm database.

An advanced spatio-temporal convolutional recurrent neural network for storm surge predictions