Evaluation of instationary flow fields using crosscorrelation in image sequences

In order to investigate the temporal behavior of dynamic flow processes, a movie version of particle image velocimetry (PIV) with a high framing capability has been developed. The experimental setup includes a copper vapor laser capable to produce pulse sequences with more than 100 pulses with a repetition rate of up to 30 kHz and a rotating drum camera to record the Mie scattering signals from the particles seeded to the flow. With this setup, image sequences with a number of consecutive images (40 - 70 depending on the image size) of a dynamic process can be taken on high resolution 36 mm film. To evaluate the 2-dimensional velocity vector fields, an algorithm based on the crosscorrelation of subareas in consecutive images has been developed and optimized in terms of computing time. The technique was applied to resolve the instationary flow field in a single cylinder spark ignition (SI) engine during the compression stroke.