A platform for research: civil engineering, architecture and urbanism
A platform for research: civil engineering, architecture and urbanism
Interpolation of wind-induced pressure time series with an artificial neural network
This paper presents an approach for interpolating wind-induced pressure time series on a model low-rise building. The approach involves using artificial neural networks (ANN) that is capable of capturing the complex variations of the pressure time series and then predicting them over a long time. The ANN is trained with the time series data from adjacent taps and historical data and optimized at a single location within the corner vortex for a single (cornering) wind direction. The good performance and robustness of the proposed neural network is demonstrated by the prediction of pressure time series data at other roof locations within the corner vortex (except in the corner itself) and for slightly altered wind directions and terrains. Comparison of the results with those obtained via linear interpolation (LI) clearly indicates that the ANN approach overcomes the problem of spatial filtering associated with LI when low-resolution data is used. The main downside of the technique is the higher level of complexity and computational effort.
Interpolation of wind-induced pressure time series with an artificial neural network
This paper presents an approach for interpolating wind-induced pressure time series on a model low-rise building. The approach involves using artificial neural networks (ANN) that is capable of capturing the complex variations of the pressure time series and then predicting them over a long time. The ANN is trained with the time series data from adjacent taps and historical data and optimized at a single location within the corner vortex for a single (cornering) wind direction. The good performance and robustness of the proposed neural network is demonstrated by the prediction of pressure time series data at other roof locations within the corner vortex (except in the corner itself) and for slightly altered wind directions and terrains. Comparison of the results with those obtained via linear interpolation (LI) clearly indicates that the ANN approach overcomes the problem of spatial filtering associated with LI when low-resolution data is used. The main downside of the technique is the higher level of complexity and computational effort.
Interpolation of wind-induced pressure time series with an artificial neural network
Chen, Y. (author) / Kopp, G.A. (author) / Surry, D. (author)
Journal of Wind Engineering and Industrial Aerodynamics ; 90 ; 589-615
2002
27 Seiten, 31 Quellen
Article (Journal)
English
Interpolation of wind-induced pressure time series with an artificial neural network
| Online Contents | 2002
Forecasting of Wind Induced Pressure on Setback Building Using Artificial Neural Network
| BASE | 2020
Prediction of Wind-Induced Pressure on Pentagon Plan Shape Building using Artificial Neural Network
| Springer Verlag | 2022
Artificial Neural Network-Based Prediction of Wind Pressure Coefficients on Buildings
| Springer Verlag | 2021
Neural Network and Autoregressive Modeling of Wind-Induced Pressure
| British Library Conference Proceedings | 2000