当前位置: X-MOL 学术Meteorol. Appl. › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
Deep‐learning model used to predict thunderstorms within 400 km 2 of south Texas domains
Meteorological Applications ( IF 2.3 ) Pub Date : 2020-03-01 , DOI: 10.1002/met.1905
Hamid Kamangir 1, 2 , Waylon Collins 3 , Philippe Tissot 2 , Scott A. King 1
Affiliation  

Correspondence Hamid Kamangir, Department of Computing Sciences, Texas A&M University-Corpus Christi, TX, USA. Email: hkamangir@islander.tamucc.edu Abstract A deep-learning neural network (DLNN) model was developed to predict thunderstorm occurrence within 400 km South Texas domains for up to 15 hr (±2 hr accuracy) in advance. The input features were chosen primarily from numerical weather prediction model output parameters/variables; cloud-toground lightning served as the target. The deep-learning technique used was the stacked denoising autoencoder (SDAE) in order to create a higher order representation of the features. Logistic regression was then applied to the SDAE output to train the predictive model. An iterative technique was used to determine the optimal SDAE architecture. The performance of the optimized DLNN classifiers exceeded that of the corresponding shallow neural network models, a classifier via a combination of principal component analysis and logistic regression, and operational weather forecasters, based on the same data set.

中文翻译:

用于预测德克萨斯州南部 400 km 2 范围内雷暴的深度学习模型

通讯员 Hamid Kamangir,德克萨斯农工大学-科珀斯克里斯蒂,美国,计算科学系。电子邮件:hkamangir@islander.tamucc.edu 摘要 开发了一种深度学习神经网络 (DLNN) 模型,可提前预测南德克萨斯地区 400 公里范围内雷暴的发生时间长达 15 小时(±2 小时精度)。输入特征主要是从数值天气预报模型输出参数/变量中选择的;云对地闪电作为目标。所使用的深度学习技术是堆叠去噪自编码器 (SDAE),以创建特征的高阶表示。然后将逻辑回归应用于 SDAE 输出以训练预测模型。使用迭代技术来确定最佳 SDAE 架构。
更新日期:2020-03-01
down
wechat
bug