Analysis of the filtering effect of the stochastic estimation and accuracy improvement by sensor location optimization

Abstract

The reconstruction of the flow behind a backward facing step at a Reynolds number of 60,000 using Linear Stochastic Estimation (LSE) and modified is investigated. In particular, the turbulent spatial integral length scales reconstructed for several sensor configurations are studied. The reconstruction of the Proper Orthogonal Decomposition (POD) modes is also performed in order to show the limitations of the LSE reconstruction for complex flows, for which taking into account only a few POD modes is not enough to represent the flow dynamics. The importance of the sensors location on the reconstruction is also emphasized and the opportunity to use a sensor location optimization algorithm investigated.

Linear stochastic estimation of the coherent structures in internal combustion engine flow

“…al. 22 found improvement when the sensor locations were close to the extrema of POD modes which agrees with work by Cohen et. al.…”

“…Sensor locations are chosen at random as literature discussed suggests that sensor location can impact the accuracy of resulting estimations. The purpose of this work is not to optimise sensor location selection (for that see 22,24 ) but instead to provide a guide for the number of sensors required for an acceptable accuracy of estimation. Therefore, locations are randomised and the investigation is repeated a total of 10 times and the mean is taken to exclude the effects of sensor location.…”

“…al. 22 describe the limitations in estimating smaller scales of turbulence using wall based pressure measurements. However, the study goes on to demonstrate that improvements may be made by considering sensor location optimisation according to an algorithm developed by Muradore et.al.…”

“…The works discussed 3,4,10,21,22,22,23,25 suggest that the LSE technique is suitable to estimate the cyclically variable coherent structures within the IC engine, using a number of point based velocity measurements as sensors. By using velocity components as sensors, one can expect a significant spatial correlation and therefore satisfactory measure of accuracy in the estimations generated with minimum sensor points.…”

Spurious PIV Vector Correction Using Linear Stochastic Estimation

“…This in turn allows synchronised estimation of velocity from only the sensor measurements. The limitation of using this sensor type is described, particularly the importance of the proximity and distribution of these sensors [22].…”