Particle image velocimetry method and device based on global optimization

Abstract

The embodiment of the present application relates to a particle image velocimetry method and device based on global optimization, including: obtaining a predetermined query window size, step size, and optimization termination condition; Particle image velocimetry PIV flow field; based on the window size, the initial velocity field in the initial PIV flow field is optimized until the optimization termination condition is reached, and the optimized velocity field is obtained, wherein the PIV flow field An initial velocity field and a correlation coefficient matrix are included; the optimized velocity field is fitted based on the correlation coefficient matrix to obtain the final velocity field of the particle image pair. Thus, the accuracy and resolution of the velocity field can be improved by globally optimizing the velocity field and fitting.

Description

Technical field [0001] The present application relates to the field of experimental fluid mechanics, and more particularly to a global optimized particle image speed method and apparatus. Background technique [0002] Particle Image Velocime, PIV technology can obtain two dimensional or three-dimensional velocity distribution of flow fields, is an important means of quantitative measurement complex flow field in experimental fluid mechanics. This technology is a non-contact measurement. In the experiment, a small smaller tracer particles capable of follow fluid motion are spread in the flow field, which are evenly distributed in the flow field and illuminate by the laser sheet light or the brightness. The camera receives the tracer particle scattering light to form a particle image, and the acquisition time interval of continuous two images (image pairs) is Δt. In order to capture the movement of the particles, it is generally less than ΔT. Image pair information containing fluid...

AI Technical Summary

Problems solved by technology

However, limited by the experimental conditions and the algorithm itself, the flow field obtained by the cross-correlation algorithm still faces two major problems: noise and unsatisfactory resolution

When the size of the query window is large, the obtained velocity field is relatively smooth, but the spatial resolution of the velocity field is low due to the modulation of the window; when the size of the query window is small, although the velocity field with high spatial resolution can be obtained, the flow field contains More errors and bad points affect the further feature extraction of the flow field

Although PIV technology has been widely used in the study of fluid mechanics, it can be seen from the above analysis that the flow field obtained by PIV also has inherent shortcomings.

First, there are experimental errors and noises in the PIV flow field, which cannot fully satisfy the governing equation of fluid mechanics (N-S equation); second, the spatial resolution of the PIV flow field is limited by the size of the query window, and the small-scale flow structure cannot be fully resolved

For example, in a turbulent boundary layer, the velocity field calculated by PIV is not reliable near the wall due to the presence of velocity gradients and fewer tracer particles

Images