English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Tools annual

Global: Zendy Free Plan

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Reconstruction of flow field from real sparse data by a physics-orientedapproach is a current challenge for fluid scientists in the AI community. Theproblem includes feature recognition and implementation of AI algorithms thatlink data to a physical feature space in order to produce reconstructed data.The present article applies machine learning approach to study contribution ofdifferent flow-based features with practical fluid mechanics applications forreconstruction of the missing data of turbomachinery PIV measurements. Supportvector regression (SVR) and multi-layer perceptron (MLP) are selected as tworobust regressors capable of modelling non-linear fluid flow phenomena. Theproposed flow-based features are optimally scaled and filtered to extract thebest configuration. In addition to conventional data-based validation of theregressors, a metric is proposed that reflects mass conservation law as animportant requirement for a physical flow reproduction. For a velocity fieldincluding 25% of clustered missing data, the reconstruction accuracy achievedby SVR in terms of R2-score is as high as 0.993 for the in-plane velocityvectors in comparison with that obtained by MLP which is up to 0.981. In termsof mass conservation metric, the SVR model by R2-score up to 0.96 isconsiderably more accurate than the MLP estimator. For extremely sparse datawith a gappiness of 75%, vector and contour plots from SVR and MLP wereconsistent with those of the original field.

Flow based features and validation metric for machine learning reconstruction of PIV data