Single Camera, Three-Dimensional Particle Tracking Velocimetry.

Abstract

This thesis presents the development of single-camera, three-dimensional particle-tracking velocimetry (SC3D-PTV), a method for measuring 3D air flow inside an optically-accessible combustion engine. The efficiency and pollutant formation of an internal combustion engine are strongly influenced by the air flow inside the engine cylinder, so understanding the flow is critical for improving engine operation. However, because 3D, volumetric results are needed to fully quantify these flows, and limited optical access is available, existing flow measurement techniques are ill-suited to engine measurements. To address this need, the SC3D-PTV method can perform high-speed, high-resolution 3D flow measurements with limited optical access, promote greater understanding of in-cylinder flow, and lead to improved engine operation. The optical element used for SC3D-PTV is similar to a stereo-microscope. A single large lens and two smaller lenses are used to create two parallel imaging sub-systems within a single housing. The two imaging sub-systems view the same measurement volume from different angles, but share a focal plane without perspective distortion. The positions of the object within the two images indicate the 3D position of the object, and 3D velocities are measured by taking images at successive points in time. A novel PTV algorithm relying on the similarity of the particle images corresponding to a single, physical particle produces 3-component, volumetric velocity fields without the reconstruction of an instantaneous 3D particle field. Validation of the SC3D-PTV method was obtained by analyzing a single experimental data set from a simple flow with both the SC3D-PTV algorithm and a stereoscopic PIV algorithm and comparing the results. After validation, the SC3D-PTV technique was applied to the air flow inside a motored engine. The three-component, 3D results provided by SC3D-PTV were shown to provide details of the flow that were lost when performing the planar, two- and three-component measurements commonly used to study engine flows. Because SC3D-PTV can fully quantify the 3D flow structures found within engines, even with very limited optical access, the development of SC3D-PTV significantly advances the study of engine flows and offers the chance to gain insights that would be impossible with existing measurement techniques.