A dimensionality reduction method to select the most representative daylight illuminance distributions

@article{Kent2020ADR,
  title={A dimensionality reduction method to select the most representative daylight illuminance distributions},
  author={Michael G. Kent and S. Schiavon and John Alstan Jakubiec},
  journal={Journal of Building Performance Simulation},
  year={2020},
  volume={13},
  pages={122 - 135}
}
ABSTRACT One challenge when evaluating daylight distribution is dealing with the large amount of temporal and spatial data, visualizations and variability in illuminances that are assessed in buildings. Using a dimensionality reduction method based on principal component analysis, we identified the most representative annual daylight distributions. We modelled a rectangular room containing an analysis grid of 3200 illuminance sensor points and simulated 3285 different temporal daylight… 
Microclimate Analysis as a Design Driver of Architecture
TLDR
The investigation shows that most of the available tools for advanced climatic modelling have been developed for research purposes and are not optimized for architectural and urban design; consequently, they require adaptations and modifications to extend their functionality or to achieve interoperability with software commonly used by architects.

References

SHOWING 1-10 OF 88 REFERENCES
Useful daylight illuminances: A replacement for daylight factors
Useful daylight illuminance: a new paradigm for assessing daylight in buildings
TLDR
A new paradigm to assess daylight in buildings called ‘useful daylight illuminance’, or UDI, is introduced, founded on an annual time-series of absolute values for illuminances predicted under realistic skies generated from standard meteorological datasets.
Analytic PCA Construction for Theoretical Analysis of Lighting Variability in Images of a Lambertian Object
TLDR
This work analytically constructs the principal component analysis for images of a convex Lambertian object, explicitly taking attached shadows into account, and finds the principal eigenmodes and eigenvalues with respect to lighting variability, and extends these results to the single-viewpoint case.
Acquiring linear subspaces for face recognition under variable lighting
TLDR
This paper shows how to arrange physical lighting so that the acquired images of each object can be directly used as the basis vectors of a low-dimensional linear space and that this subspace is close to those acquired by the other methods.
High-dimensional data clustering
Editorial: Light distribution – a missing variable
A frequent theme of current lighting research is the optimisation of light spectrum so as to provide a desirable perception closely linked to vision such as spatial brightness or a perception
Points of Significance: Principal component analysis
TLDR
PCA reduces data by geometrically projecting them onto lower dimensions called principal components, with the goal of finding the best summary of the data using a limited number of PCs, whereas linear regression minimizes the distance between the response variable and its predicted value.
...
...