Eine Plattform für die Wissenschaft: Bauingenieurwesen, Architektur und Urbanistik
Eine Plattform für die Wissenschaft: Bauingenieurwesen, Architektur und Urbanistik
Knowledge-enhanced deep learning for simulation of tropical cyclone boundary-layer winds
https://orcid.org/0000-0003-4326-3655
https://orcid.org/0000-0002-9163-4716
Abstract Accurate and efficient modeling of the wind field is critical to effective mitigation of losses due to the tropical cyclone-related hazards. To this end, a knowledge-enhanced deep learning algorithm was developed in this study to simulate the wind field inside tropical cyclone boundary-layer. More specifically, the machine-readable knowledge in terms of both physics-based equations and/or semi-empirical formulas was leveraged to enhance the regularization mechanism during the training of deep networks for dynamics of tropical cyclone boundary-layer winds. To comprehensively appreciate the high effectiveness of knowledge-enhanced deep learning to capture the complex dynamics using small datasets, two nonlinear flow systems governed respectively by 1D and 2D Navier-Stokes equations were first revisited. Then, a knowledge-enhanced deep network was developed to simulate tropical cyclone boundary-layer winds using the storm parameters (e.g., spatial coordinates, storm size and intensity) as inputs. The reduced 3D Navier-Stokes equations based on several state-of-the-art semi-empirical formulas were employed in the construction of deep networks. Due to the effective utilization of the prior knowledge on the tropical cyclone boundary-layer winds, only a relatively small number of training datasets (either from field measurements or high-fidelity numerical simulations) are needed. With the trained knowledge-enhanced deep network, it has been demonstrated that the boundary-layer winds associated with various tropical cyclones can be accurately and efficiently predicted.
Highlights A knowledge-enhanced deep learning was developed to simulate tropical cyclone boundary-layer winds. Knowledge in terms of physics-based equations and/or semi-empirical formulas was used to enhance regularization mechanism. Small number of training datasets are needed due to effective utilization of the prior knowledge on tropical cyclone winds.
Knowledge-enhanced deep learning for simulation of tropical cyclone boundary-layer winds
https://orcid.org/0000-0003-4326-3655
https://orcid.org/0000-0002-9163-4716
Abstract Accurate and efficient modeling of the wind field is critical to effective mitigation of losses due to the tropical cyclone-related hazards. To this end, a knowledge-enhanced deep learning algorithm was developed in this study to simulate the wind field inside tropical cyclone boundary-layer. More specifically, the machine-readable knowledge in terms of both physics-based equations and/or semi-empirical formulas was leveraged to enhance the regularization mechanism during the training of deep networks for dynamics of tropical cyclone boundary-layer winds. To comprehensively appreciate the high effectiveness of knowledge-enhanced deep learning to capture the complex dynamics using small datasets, two nonlinear flow systems governed respectively by 1D and 2D Navier-Stokes equations were first revisited. Then, a knowledge-enhanced deep network was developed to simulate tropical cyclone boundary-layer winds using the storm parameters (e.g., spatial coordinates, storm size and intensity) as inputs. The reduced 3D Navier-Stokes equations based on several state-of-the-art semi-empirical formulas were employed in the construction of deep networks. Due to the effective utilization of the prior knowledge on the tropical cyclone boundary-layer winds, only a relatively small number of training datasets (either from field measurements or high-fidelity numerical simulations) are needed. With the trained knowledge-enhanced deep network, it has been demonstrated that the boundary-layer winds associated with various tropical cyclones can be accurately and efficiently predicted.
Highlights A knowledge-enhanced deep learning was developed to simulate tropical cyclone boundary-layer winds. Knowledge in terms of physics-based equations and/or semi-empirical formulas was used to enhance regularization mechanism. Small number of training datasets are needed due to effective utilization of the prior knowledge on tropical cyclone winds.
Knowledge-enhanced deep learning for simulation of tropical cyclone boundary-layer winds
https://orcid.org/0000-0003-4326-3655
https://orcid.org/0000-0002-9163-4716
Abstract Accurate and efficient modeling of the wind field is critical to effective mitigation of losses due to the tropical cyclone-related hazards. To this end, a knowledge-enhanced deep learning algorithm was developed in this study to simulate the wind field inside tropical cyclone boundary-layer. More specifically, the machine-readable knowledge in terms of both physics-based equations and/or semi-empirical formulas was leveraged to enhance the regularization mechanism during the training of deep networks for dynamics of tropical cyclone boundary-layer winds. To comprehensively appreciate the high effectiveness of knowledge-enhanced deep learning to capture the complex dynamics using small datasets, two nonlinear flow systems governed respectively by 1D and 2D Navier-Stokes equations were first revisited. Then, a knowledge-enhanced deep network was developed to simulate tropical cyclone boundary-layer winds using the storm parameters (e.g., spatial coordinates, storm size and intensity) as inputs. The reduced 3D Navier-Stokes equations based on several state-of-the-art semi-empirical formulas were employed in the construction of deep networks. Due to the effective utilization of the prior knowledge on the tropical cyclone boundary-layer winds, only a relatively small number of training datasets (either from field measurements or high-fidelity numerical simulations) are needed. With the trained knowledge-enhanced deep network, it has been demonstrated that the boundary-layer winds associated with various tropical cyclones can be accurately and efficiently predicted.
Highlights A knowledge-enhanced deep learning was developed to simulate tropical cyclone boundary-layer winds. Knowledge in terms of physics-based equations and/or semi-empirical formulas was used to enhance regularization mechanism. Small number of training datasets are needed due to effective utilization of the prior knowledge on tropical cyclone winds.
Knowledge-enhanced deep learning for simulation of tropical cyclone boundary-layer winds
Snaiki, Reda (Autor:in) / Wu, Teng (Autor:in)
23.08.2019
Aufsatz (Zeitschrift)
Elektronische Ressource
Englisch
Numerical modelling of extreme tropical cyclone winds
| British Library Conference Proceedings | 1999
Gust factors for tropical cyclone, monsoon and thunderstorm winds
| Elsevier | 2015
Gust factors for tropical cyclone, monsoon and thunderstorm winds
| Online Contents | 2015
Tropical cyclone intensity forecasting using model knowledge guided deep learning model
| DOAJ | 2024
A re-examination of the characteristics of tropical cyclone winds
| British Library Conference Proceedings | 1999