Thunderstorm predictions using artificial neural networks

Simple item page
dc.contributor.authorTissot, Philippe
dc.creator.orcidhttps://orcid.org/0000-0002-2954-2378en_US
dc.creator.orcidhttps://orcid.org/0000-0002-2954-2378
dc.date.accessioned2022-09-20T15:31:57Z
dc.date.available2022-09-20T15:31:57Z
dc.date.issued2016-10-19
dc.description.abstractArtificial neural network (ANN) model classifiers were developed to generate ≤ 15 h predictions of thunderstorms within three 400-km2 domains. The feed-forward, multilayer perceptron and single hidden layer network topology, scaled conjugate gradient learning algorithm, and the sigmoid (linear) transfer function in the hidden (output) layer were used. The optimal number of neurons in the hidden layer was determined iteratively based on training set performance. Three sets of nine ANN models were developed: two sets based on predictors chosen from feature selection (FS) techniques and one set with all 36 predictors. The predictors were based on output from a numerical weather prediction (NWP) model. This study amends an earlier study and involves the increase in available training data by two orders of magnitude. ANN model performance was compared to corresponding performances of operational forecasters and multi-linear regression (MLR) models. Results revealed improvement relative to ANN models from the previous study. Comparative results between the three sets of classifiers, NDFD, and MLR models for this study were mixed—the best performers were a function of prediction hour, domain, and FS technique. Boosting the fraction of total positive target data (lightning strikes) in the training set did not improve generalization.en_US
dc.description.abstractArtificial neural network (ANN) model classifiers were developed to generate ≤ 15 h predictions of thunderstorms within three 400-km2 domains. The feed-forward, multilayer perceptron and single hidden layer network topology, scaled conjugate gradient learning algorithm, and the sigmoid (linear) transfer function in the hidden (output) layer were used. The optimal number of neurons in the hidden layer was determined iteratively based on training set performance. Three sets of nine ANN models were developed: two sets based on predictors chosen from feature selection (FS) techniques and one set with all 36 predictors. The predictors were based on output from a numerical weather prediction (NWP) model. This study amends an earlier study and involves the increase in available training data by two orders of magnitude. ANN model performance was compared to corresponding performances of operational forecasters and multi-linear regression (MLR) models. Results revealed improvement relative to ANN models from the previous study. Comparative results between the three sets of classifiers, NDFD, and MLR models for this study were mixed—the best performers were a function of prediction hour, domain, and FS technique. Boosting the fraction of total positive target data (lightning strikes) in the training set did not improve generalization.
dc.identifier.citationCollins, W. G. , & Tissot, P. (2016). Thunderstorm Predictions Using Artificial Neural Networks. In (Ed.), Artificial Neural Networks - Models and Applications. IntechOpen. https://doi.org/10.5772/63542en_US
dc.identifier.citationCollins, W. G. , & Tissot, P. (2016). Thunderstorm Predictions Using Artificial Neural Networks. In (Ed.), Artificial Neural Networks - Models and Applications. IntechOpen. https://doi.org/10.5772/63542
dc.identifier.doihttps://doi.org/10.5772/63542
dc.identifier.urihttps://hdl.handle.net/1969.6/94038
dc.language.isoen_USen_US
dc.language.isoen_US
dc.rightsAttribution 4.0 International*
dc.rightsAttribution 4.0 International
dc.rights.urihttp://creativecommons.org/licenses/by/4.0/*
dc.rights.urihttp://creativecommons.org/licenses/by/4.0/
dc.subjectthunderstorm predictionen_US
dc.subjectartificial neural networksen_US
dc.subjectcorrelation-based feature selectionen_US
dc.subjectminimum redundancy maximum relevanceen_US
dc.subjectmulti-linear regressionen_US
dc.subjectthunderstorm prediction
dc.subjectartificial neural networks
dc.subjectcorrelation-based feature selection
dc.subjectminimum redundancy maximum relevance
dc.subjectmulti-linear regression
dc.titleThunderstorm predictions using artificial neural networksen_US
dc.titleThunderstorm predictions using artificial neural networks
dc.typeArticleen_US
dc.typeArticle

Files

Original bundle

Now showing 1 - 1 of 1
Thumbnail Image
Name:
Thunderstorm Predictions Using Artificial Neural Networks.pdf
Size:
4.75 MB
Format:
Adobe Portable Document Format
Description:
Download

License bundle

Now showing 1 - 1 of 1
No Thumbnail Available
Name:
license.txt
Size:
1.72 KB
Format:
Item-specific license agreed upon to submission
Description:
Download

Collections

COE Faculty Works