Iterative inversion of phase-Doppler-anemometry size distributions from sprays of optically inhomogeneous liquids

Abstract

Using phase Doppler anemometry (PDA) to investigate sprays of optically inhomogeneous liquids leads to blurred measured size distributions. The blurring function is formed by performance of PDA measurements on single-size droplets generated by a piezoelectric droplet generator. To obtain the undistorted droplet-size distributions, a constrained iterative inversion algorithm is applied. The number of iteration steps to achieve the best possible restoration is determined by the use of synthetically generated data that has noise properties similar to the measured histograms. The obtained size distributions are checked by comparison with undistorted measurement results of an atomized optical homogeneous liquid.