Remote flow sensing of complex systems: steps towards spatio-temporal~prediction of flow patterns

Prediction of the spatial and temporal phenomena of wind flow patterns~through urban areas is investigated. Typically sparse measurements~are used in wind forecasting models for updating and prediction via a~method~called variational data assimilation. To improve upon this method, an~experimental investigation combining various measurement tools (Hot Wire~Anemometry,~Stereoscopic Particle Image Velocimetry SPIV), static pressure~measurements and Laser Doppler Velocimetry(LDV)) is carried out to study~the~airflow around wall mounted obstacles in a turbulent boundary layer.~The method of Proper Orthogonal Decomposition (POD) is used to~decompose the flow field into a finite set of POD coefficients which~vary only in time associated with a corresponding set of POD basis~functions which vary only in space. Direct measurement models utilizing~the~measurements from SPIV and LDV, along with indirect measurement models~using~sparse measurements from microphones are investigated and may ultimately~be~combined with state-space models to obtain more robust dynamical models.