A Gaussian Process-Based emulator for modeling pedestrian-level wind field

Abstract Wind tunnel tests and computational fluid dynamics (CFD) simulations remain the main modeling techniques in wind engineering despite being expensive, time-consuming, and requiring special facilities and expert knowledge. There is a clear need for a fast, accurate, but, at the same time, computationally economical substitute. This study proposes a Gaussian Process-based (GP-based) emulator to predict the pedestrian-level wind environment near a lift-up building – an isolated, unconventionally configured building. The proposed GP-based emulator transcends the limitations of previous emulators as it can handle many inputs (8) and output parameters (384) and a large dataset (150 CFD simulations). To increase computational efficiency, the current study proposes a data reduction method based on Principle Component Analysis (PCA) and a technique to estimate hyper-parameters based on optimization. The latter can efficiently compute 250 hyper-parameters and requires no prior knowledge of their probability distributions. The emulator is faster, by a factor of 107 than CFD simulations in predicting wind speeds, and its accuracy is substantiated using both qualitative and quantitative analyses, which reveal that the emulator's predictions of all-prominent flow features near a building have no systematic bias, are highly accurate, and have great reproductivity.

[1]  Yoshihide Tominaga,et al.  AIJ guidelines for practical applications of CFD to pedestrian wind environment around buildings , 2008 .

[2]  Jakob Spiegelberg,et al.  Can we use PCA to detect small signals in noisy data? , 2017, Ultramicroscopy.

[3]  Theodore Stathopoulos,et al.  Modelling wind-induced interference effects using backpropagation neural networks , 1997 .

[4]  Stanislaw Osowski,et al.  Forecasting of the daily meteorological pollution using wavelets and support vector machine , 2007, Eng. Appl. Artif. Intell..

[5]  Miltiadis Alamaniotis,et al.  Monthly load forecasting using kernel based gaussian process regression , 2014 .

[6]  A. U. Weerasuriya,et al.  Optimizing Lift-up Design to Maximize Pedestrian Wind and Thermal Comfort in ‘Hot-Calm’ and ‘Cold-Windy’ Climates , 2020 .

[7]  Junichi Mori,et al.  A Gaussian mixture copula model based localized Gaussian process regression approach for long-term wind speed prediction , 2013 .

[8]  Ricardo Nicolau Nassar Koury,et al.  Prediction of wind speed and wind direction using artificial neural network, support vector regression and adaptive neuro-fuzzy inference system , 2018 .

[9]  Ted Stathopoulos,et al.  Modeling the effect of topography on wind flow using a combined numerical-neural network approach , 2007 .

[10]  Ke Sun,et al.  Developing a meta-model for sensitivity analyses and prediction of building performance for passively designed high-rise residential buildings , 2017 .

[11]  D. Sorensen,et al.  Approximation of large-scale dynamical systems: an overview , 2004 .

[12]  Jianzhou Wang,et al.  Short-term wind speed prediction using empirical wavelet transform and Gaussian process regression , 2015 .

[13]  Philippe Renard,et al.  Distance-based Kriging relying on proxy simulations for inverse conditioning , 2013 .

[14]  Zhu Xue,et al.  Prediction of wind loads on high-rise building using a BP neural network combined with POD , 2017 .

[15]  Ashraf A. El Damatty,et al.  Enhancing wind performance of tall buildings using corner aerodynamic optimization , 2017 .

[16]  Calvin A. Whealton,et al.  Metamodeling for Uncertainty Quantification of a Flood Wave Model for Concrete Dam Breaks , 2020, Energies.

[17]  Yukio Tamura,et al.  Characteristics of pedestrian-level wind around super-tall buildings with various configurations , 2017 .

[18]  Gang Hu,et al.  Predicting wind pressures around circular cylinders using machine learning techniques , 2019, ArXiv.

[19]  Marcel Bottema.  Towards rules of thumb for wind comfort and air quality , 1999 .

[20]  G. Mahinthakumar,et al.  Enhanced Simulation-Optimization Approach Using Surrogate Modeling for Solving Inverse Problems , 2012 .

[21]  Bert Blocken,et al.  LES over RANS in building simulation for outdoor and indoor applications: A foregone conclusion? , 2018, Building Simulation.

[22]  Marco Montemurro,et al.  Multi-scale identification of the elastic properties variability for composite materials through a hybrid optimisation strategy , 2019, Composites Part B: Engineering.

[23]  Zhang Lin,et al.  Evaluation of pedestrian wind comfort near ‘lift-up’ buildings with different aspect ratios and central core modifications , 2017, Building and Environment.

[24]  Bruno Sudret,et al.  Meta-model-based importance sampling for reliability sensitivity analysis , 2014 .

[25]  Juan M. Gimenez,et al.  Prediction of wind pressure coefficients on building surfaces using artificial neural networks , 2018 .

[26]  Louis J. Durlofsky,et al.  A new carbon capture proxy model for optimizing the design and time-varying operation of a coal-natural gas power station , 2016 .

[27]  Jung-Il Choi,et al.  Uncertainty quantification of upstream wind effects on single-sided ventilation in a building using generalized polynomial chaos method , 2017 .

[28]  Peter D. Scott,et al.  Uncertainty propagation in puff-based dispersion models using polynomial chaos , 2009, 2009 12th International Conference on Information Fusion.

[29]  Luca Caracoglia,et al.  Artificial Neural Network model to predict the flutter velocity of suspension bridges , 2020 .

[30]  B. Launder,et al.  The numerical computation of turbulent flows , 1990 .

[31]  Aleksandrs Korjakins,et al.  Surrogate Models for Optimum Design of Stiffened Composite Shells , 2004 .

[32]  Vincent J.L. Gan,et al.  A holistic framework to utilize natural ventilation to optimize energy performance of residential high-rise buildings , 2019, Building and Environment.

[33]  Panagiotis G. Asteris,et al.  Self-compacting concrete strength prediction using surrogate models , 2017, Neural Computing and Applications.

[34]  Kheireddine Redouane,et al.  Adaptive surrogate modeling with evolutionary algorithm for well placement optimization in fractured reservoirs , 2019, Appl. Soft Comput..

[35]  Feng Xu,et al.  Prediction of hourly PM 2.5 using a space-time support vector regression model , 2018 .

[36]  Lorenz T. Biegler,et al.  Multi-scale optimization for process systems engineering , 2014, Comput. Chem. Eng..

[37]  A. U. Weerasuriya,et al.  Pedestrian-level wind environment near a super-tall building with unconventional configurations in a regular urban area , 2020 .

[38]  A. U. Weerasuriya,et al.  Pedestrian-level wind conditions in the space underneath lift-up buildings , 2018, Journal of Wind Engineering and Industrial Aerodynamics.

[39]  Tor Arne Johansen,et al.  Real-Time Production Optimization of Oil and Gas Production Systems: A Technology Survey , 2007 .

[40]  Bithin Datta,et al.  Comparative Evaluation of Genetic Programming and Neural Network as Potential Surrogate Models for Coastal Aquifer Management , 2011 .

[41]  A. OHagan,et al.  Bayesian analysis of computer code outputs: A tutorial , 2006, Reliab. Eng. Syst. Saf..

[42]  Gianluca Iaccarino,et al.  Quantifying inflow and RANS turbulence model form uncertainties for wind engineering flows , 2015 .

[43]  Kenny C. S Kwok,et al.  Wind-induced responses of a tall building with a double-skin façade system , 2017 .

[44]  P. Bradshaw,et al.  Momentum transfer in boundary layers , 1977 .

[45]  P. Vanrolleghem,et al.  Fast, simultaneous simulation of the integrated urban wastewater system using mechanistic surrogate models. , 2001, Water science and technology : a journal of the International Association on Water Pollution Research.

[46]  Hyo Seon Park,et al.  Convolutional neural network‐based wind‐induced response estimation model for tall buildings , 2019, Comput. Aided Civ. Infrastructure Eng..

[47]  Qiusheng Li,et al.  Numerical simulation on fluid-structure interaction of wind around super-tall building at high reynolds number conditions , 2013 .

[48]  Cristina H. Amon,et al.  Bayesian Surrogates Applied to Conceptual Stages of the Engineering Design Process , 2003 .

[49]  Frank Sehnke,et al.  Wind turbine power curve modeling based on Gaussian Processes and Artificial Neural Networks , 2018, Renewable Energy.

[50]  Bryan A. Tolson,et al.  Review of surrogate modeling in water resources , 2012 .

[51]  Carl E. Rasmussen,et al.  Understanding Probabilistic Sparse Gaussian Process Approximations , 2016, NIPS.

[52]  Bernardete Ribeiro,et al.  Active learning metamodeling for policy analysis: Application to an emergency medical service simulator , 2019, Simul. Model. Pract. Theory.

[53]  Piyush M. Tagade,et al.  A Gaussian process emulator approach for rapid contaminant characterization with an integrated multizone-CFD model , 2013, 1305.2565.

[54]  Slawomir Koziel,et al.  Multi-fidelity design optimization of transonic airfoils using physics-based surrogate modeling and shape-preserving response prediction , 2010, J. Comput. Sci..

[55]  Xiaosheng Qin,et al.  Uncertainty analysis of flood inundation modelling using GLUE with surrogate models in stochastic sampling , 2015 .

[56]  R. S. Sonparote,et al.  Dynamic wind response of tall buildings using artificial neural network , 2019, The Structural Design of Tall and Special Buildings.

[57]  Jörg Franke,et al.  The COST 732 Best Practice Guideline for CFD simulation of flows in the urban environment: a summary , 2011 .

[58]  Jie Yu,et al.  Short-term wind speed prediction using an unscented Kalman filter based state-space support vector regression approach , 2014 .

[59]  Agathoklis Giaralis,et al.  Wind-Induced Vibration Mitigation in Tall Buildings Using the Tuned Mass-Damper-Inerter , 2017 .

[60]  Juan M. Gimenez,et al.  Optimization of RANS turbulence models using genetic algorithms to improve the prediction of wind pressure coefficients on low-rise buildings , 2019, Journal of Wind Engineering and Industrial Aerodynamics.

[61]  Xiaofeng Meng,et al.  Wind Power Forecasts Using Gaussian Processes and Numerical Weather Prediction , 2014, IEEE Transactions on Power Systems.

[62]  Jamshid Ghaboussi,et al.  Estimation of Aeroelastic Parameters of Bridge Decks Using Neural Networks , 2004 .

[63]  Dacheng Tao,et al.  Deep learning-based investigation of wind pressures on tall building under interference effects , 2020 .

[64]  Fergus R. Fricke,et al.  The interference index and its prediction using a neural network analysis of wind-tunnel data , 1999 .

[65]  Mark van der Wilk.  Sparse Gaussian process approximations and applications , 2019 .

[66]  Ivo Couckuyt,et al.  Elastic characterization of membranes with a complex shape using point indentation measurements and inverse modelling , 2010 .

[67]  James Irving,et al.  On uncertainty quantification in hydrogeology and hydrogeophysics , 2017 .

[68]  C. Gorlé,et al.  Improving urban flow predictions through data assimilation , 2018 .

[69]  C. Gorlé,et al.  Quantifying inflow uncertainties for CFD simulations of the flow in downtown Oklahoma City , 2014 .

[70]  Stefano Ermon,et al.  Sparse Gaussian Processes for Bayesian Optimization , 2016, UAI.

[71]  Osamah Basheer Shukur,et al.  Daily wind speed forecasting through hybrid KF-ANN model based on ARIMA , 2015 .

[72]  Ashutosh Sharma,et al.  A review on the study of urban wind at the pedestrian level around buildings , 2018, Journal of Building Engineering.

[73]  Majid Jamil,et al.  A comparative analysis of ANN and chaotic approach-based wind speed prediction in India , 2019, Neural Computing and Applications.

[74]  P. Moonen,et al.  Employing statistical model emulation as a surrogate for CFD , 2015, Environ. Model. Softw..

[75]  Joaquim R. R. A. Martins,et al.  A Surrogate-Based Multi-Scale Model for Mass Transport and Electrochemical Kinetics in Lithium-Ion Battery Electrodes , 2014 .

[76]  David Surry,et al.  Prediction of pressure coefficients on roofs of low buildings using artificial neural networks , 2003 .

[77]  Ahsan Kareem,et al.  Modeling hysteretic nonlinear behavior of bridge aerodynamics via cellular automata nested neural network , 2011 .

[78]  Girma Bitsuamlak,et al.  Multiobjective Aerodynamic Optimization of Tall Building Openings for Wind-Induced Load Reduction , 2018, Journal of Structural Engineering.

[79]  Haikun Wei,et al.  A Gaussian process regression based hybrid approach for short-term wind speed prediction , 2016 .

[80]  Dongxiao Niu,et al.  Short-term wind speed forecasting using wavelet transform and support vector machines optimized by genetic algorithm , 2014 .

[81]  Andrea Castelletti,et al.  Data-driven dynamic emulation modelling for the optimal management of environmental systems , 2012, Environ. Model. Softw..

[82]  Fuyou Xu,et al.  Interpolation of wind pressures using Gaussian process regression , 2019, Journal of Wind Engineering and Industrial Aerodynamics.

[83]  B. Sudret,et al.  An adaptive algorithm to build up sparse polynomial chaos expansions for stochastic finite element analysis , 2010 .

[84]  Guido Morgenthal,et al.  Prediction of aeroelastic response of bridge decks using artificial neural networks , 2020 .

[85]  Kenny C. S Kwok,et al.  Adopting ‘lift-up’ building design to improve the surrounding pedestrian-level wind environment , 2017, Building and Environment.

[86]  Antony M. Overstall,et al.  Multivariate emulation of computer simulators: model selection and diagnostics with application to a humanitarian relief model , 2015, Journal of the Royal Statistical Society. Series C, Applied statistics.

[87]  T. Stathopoulos.  Introduction to Wind Engineering, Wind Structure, Wind-Building Interaction , 2007 .

[88]  Nilay Shah,et al.  Machine-learning methods for integrated renewable power generation: A comparative study of artificial neural networks, support vector regression, and Gaussian Process Regression , 2019, Renewable and Sustainable Energy Reviews.

[89]  Bje Bert Blocken,et al.  Pedestrian wind comfort around buildings : comparison of wind comfort criteria based on whole-flow field data for a complex case study , 2013 .

[90]  CsatóLehel,et al.  Sparse on-line Gaussian processes , 2002 .

[91]  Márton Balczó,et al.  Simulation of flow in an idealised city using various CFD codes , 2011 .

[92]  Laura Uusitalo,et al.  An overview of methods to evaluate uncertainty of deterministic models in decision support , 2015, Environ. Model. Softw..

[93]  J. Murillo-Escobar,et al.  Forecasting concentrations of air pollutants using support vector regression improved with particle swarm optimization: Case study in Aburrá Valley, Colombia , 2019, Urban Climate.

[94]  M. Awang,et al.  System Identification Based Proxy Model of a Reservoir under Water Injection , 2017 .

[95]  T. Simpson,et al.  Use of Kriging Models to Approximate Deterministic Computer Models , 2005 .

[96]  Wahid Ghaly,et al.  Aerodynamic optimization of turbomachinery blades using evolutionary methods and ANN-based surrogate models , 2008 .

[97]  T. Stathopoulos,et al.  Pedestrian wind environment around tall buildings , 2016 .

[98]  Jack P. C. Kleijnen,et al.  Kriging Metamodeling in Simulation: A Review , 2007, Eur. J. Oper. Res..