Employing statistical model emulation as a surrogate for CFD

Predictive models for assessing the passive solar and daylight potential of neighborhood designs: A comparative proof-of-concept study

“…GPs have the advantage over MLR in that they are based on a fully probabilistic framework, and thus, that uncertainties can be extracted for every prediction (Moonen and Allegrini, 2015). Furthermore, the Bayesian calibration framework automatically identifies the relevant input parameters.…”

“…For a stepwise treatment of the construction of a metamodel based on GPs and its validation, the interested reader is referred to Moonen and Allegrini (2015). For a fundamental treatment of the subject, additionally accounting for systematic bias between inputs and outputs, we refer to Higdon et al (2008) and Kennedy and O'Hagan (2001).…”

“…Alternatively, a single GP-metamodel could be constructed having a vectorial output (Moonen and Allegrini, 2015).…”

“…We follow the mathematical formulation described in Moonen and Allegrini (2015), as the predecessor of this approach has shown to yield promising results in the field of building energy simulation (Evins et al, 2014). In this approach, the relation between the simulated responses y(x) and the output of the metamodel f (x) can be expressed as:…”

Geospatial uncertainty modeling using Stacked Gaussian Processes

“…They have found various applications in the environmental modeling community, where they are used as data-driven models capable to predict various quantities of interest with quantified uncertainties such as ultra fine particles (Reggente et al (2014)), mean temperatures over North Atlantic Ocean (Higdon (1998)), wind speed (Hu and Wang (2015)), and monthly streamflow (Sun et al (2014)), just to name a few. When the training data for GPs comes from simulators rather than field measurements, then GPs become computational efficient surrogate models or emulators of highfidelity models (Kennedy et al (2002); O'Hagan (2006); Conti and O'Hagan (2010)), with various applications in environmental modeling such as fire emissions (Katurji et al (2015)), ocean and climate circulation (Tokmakian et al (2012)), urban drainage (Machac et al (2016)), and computational fluid dynamics (Moonen and Allegrini (2015)).…”