Deep learning-based investigation of wind pressures on tall building under interference effects

Hu, Gang; Liu, Lingbo; Tao, Dacheng; Song, Junhua; Tse, Kam Tim; Kwok, K.C.S.

doi:10.1016/j.jweia.2020.104138

Cited by 97 publications

(12 citation statements)

References 63 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The building facades are often damaged under strong winds (Hu et al, 2017, 2019; Liang et al, 2020). Though some new methods emerge (Hu et al, 2020; Lin et al, 2021; Kim et al, 2021), wind tunnel testing is still the most favorable way to obtain the wind pressures on buildings (Sy et al, 2019; Behera et al, 2020; He et al, 2019).…”

Section: Introductionmentioning

confidence: 99%

Experimental investigations of interference effects on wind pressures of tall buildings

Peng

Liu

et al. 2021

Advances in Structural Engineering

Self Cite

View full text Add to dashboard Cite

The interference effects between two tall buildings on building surface wind pressures are studied by using wind tunnel experiments. Different breadth ratios ( B r = B interfering/ Bprincipal), depth ratios ( D r = D interfering/ Dprincipal), and relative locations of the two buildings are tested. The results show that the interference effects on mean wind pressures on the windward face and the right face are more considerable than those on the other two faces. The windward face mainly shows the shielding effect, while the right face exhibits the amplification effect. The surface interference factors ( SIF) of windward and right faces under different breadth ratios and depth ratios have clear correlation with the reference values ( B r = D r = 1.0), and a series of correlation formulas are obtained. The SIF values of the windward face in tandem arrangement and the maximum and minimum interference factors of the right face in parallel arrangement are formulated. The most prominent local peak pressure under different cases mainly concentrates in the side edge and upper edge of the windward face, the left upper corner, left lower corner, and the upper-middle part of the right face.

show abstract

Section: Introductionmentioning

confidence: 99%

Experimental investigations of interference effects on wind pressures of tall buildings

Peng

Liu

et al. 2021

Advances in Structural Engineering

Self Cite

View full text Add to dashboard Cite

show abstract

“…GAIN was implemented on five real-world datasets obtained from the University of California (Oakland, CA, USA), Irvine repository to quantitatively evaluate the imputation performance. It was observed that GAIN models significantly outperformed the other state-of-the-art data imputation methods [ 68 , 69 , 70 ]. A comparison about the models used in the literature for data imputation and prediction is presented in Table 1 .…”

Section: Related Workmentioning

confidence: 99%

Wind-Induced Pressure Prediction on Tall Buildings Using Generative Adversarial Imputation Network

Kim

Yuvaraj

Preethaa

et al. 2021

Sensors

Self Cite

View full text Add to dashboard Cite

Wind tunnel testing techniques are the main research tools for evaluating the wind loadings of buildings. They are significant in designing structurally safe and comfortable buildings. The wind tunnel pressure measurement technique using pressure sensors is significant for assessing the cladding pressures of buildings. However, some pressure sensors usually fail and cause loss of data, which are difficult to restore. In the literature, numerous techniques are implemented for imputing the single instance data values and data imputation for multiple instantaneous time intervals with accurate predictions needs to be addressed. Thus, the data imputation capacity of machine learning models is used to predict the missing wind pressure data for tall buildings in this study. A generative adversarial imputation network (GAIN) is proposed to predict the pressure coefficients at various instantaneous time intervals on tall buildings. The proposed model is validated by comparing the performance of GAIN with that of the K-nearest neighbor and multiple imputations by chained equation models. The experimental results show that the GAIN model provides the best fit, achieving more accurate predictions with the minimum average variance and minimum average standard deviation. The average mean-squared error for all four sides of the building was the minimum (0.016), and the average R-squared error was the maximum (0.961). The proposed model can ensure the health and prolonged existence of a structure based on wind environment.

show abstract

“…As a multi-field interdisciplinary subject, ML involves probability theory, statistics, approximation theory, convex analysis, algorithm complexity theory and other subjects. Algorithms and models in ML enable computer systems to learn from data and establish mathematical models to make predictions with minimal human intervention (Hu et al, 2020; Huang et al, 2017; Kotsiantis, 2007). In general, ML techniques are mainly comprised by supervised ML, unsupervised ML and reinforced ML (Sun et al, 2021).…”

Section: Introductionmentioning

confidence: 99%

Machine learning based algorithms for wind pressure prediction of high-rise buildings

Huang

et al. 2022

Advances in Structural Engineering

View full text Add to dashboard Cite

In recent years, machine learning (ML) techniques have been used in various fields of engineering practice. In order to evaluate the feasibility of machine learning algorithms for prediction of wind-induced effects on high-rise buildings, four ML algorithms including ridge regression, decision tree, random forest and gradient boosting regression tree are adopted in this study to predict wind pressures on Commonwealth Advisory Aeronautical Research Council standard tall building. The gradient boosting regression tree model is proved to be well performed in predicting both mean wind pressures and fluctuating wind pressures. Compared to expensive wind tunnel tests and time-consuming computational fluid dynamic simulations, it is expected that the gradient boosting regression tree model is an efficient and economical alternative for predicting wind pressures on high-rise buildings.

show abstract

Deep learning-based investigation of wind pressures on tall building under interference effects

Cited by 97 publications

References 63 publications

Experimental investigations of interference effects on wind pressures of tall buildings

Experimental investigations of interference effects on wind pressures of tall buildings

Wind-Induced Pressure Prediction on Tall Buildings Using Generative Adversarial Imputation Network

Machine learning based algorithms for wind pressure prediction of high-rise buildings

Contact Info

Product

Resources

About