EAGER: Maximizing Spatial Resolution and Accuracy of PIV with Optical Flow Velocimetry

目前研究流体流动最常用的方法是用粒子图像测速法测量速度场。这已经是三十多年来的标准方法，并已被证明是一种有效和可靠的技术。然而，最近的研究表明，通过使用不同的方法处理图像，可以使用相同的设备进行测量，但精度和细节要高得多。该项目的目的是开发和验证一种新的算法，以从粒子图像中获得速度测量，以取代目前的技术状态。新算法将显著提高流体流量测量系统在学术界、工业界和国家实验室的应用能力。完成的算法和验证将分发给广泛的潜在用户。本研究的重点是开发一种新的算法，用于从激光片照射的示踪粒子对图像中进行速度测量。新算法将直接取代最先进的粒子图像测速处理软件中使用的基于相互关联的方法，但该方法不需要任何新设备，如相机或激光器，因为两种算法输入到软件中的采集图像是相同的。众所周知，即使是最先进的基于相关的算法在从图像计算速度场时也会遭受空间分辨率的损失，这主要是由于在处理过程中使用了有限大小的查询窗口。这种新方法将利用高等数学和流体力学的控制定律，形成一种健壮的、受物理启发的方法，以解决当前算法的缺点。这种新算法将使研究人员能够测量湍流中的速度，其精度、分辨率和动态范围都比迄今为止使用传统实验硬件时要高得多。项目结束时的愿景是，在进行研究以确定最佳分配模式之后，将经过彻底测试的最终版本算法传播给尽可能多的流体动力学实验人员。该奖项反映了美国国家科学基金会的法定使命，并通过使用基金会的知识价值和更广泛的影响审查标准进行评估，被认为值得支持。

Fluid flows are most often studied today by measuring the velocity fields using particle image velocimetry. This has been the standard approach for over thirty years and has proven to be an effective and reliable technique. However, recent work has shown that it is possible to produce measurements with the same equipment but much higher accuracy and detail by using a different method for processing the images. The aim of this project is to develop and validate a new algorithm to obtain velocity measurements from particle images to replace the current state of the art. The new algorithm will significantly advance the capabilities of fluid flow measurement systems employed in academia, industry, and national laboratories. The finished algorithms and validations will be disseminated to a wide range of potential users.This research focuses on the development of a novel algorithm for velocimetry from pairs of images of tracer particles illuminated by a laser sheet. The new algorithm will substitute directly for the cross-correlation based approaches used in state-of-the-art particle image velocimetry processing software, but the approach does not require any new equipment, such as cameras or lasers, since the acquired images input to the software are the same for both algorithms. It is well known that even the most advanced correlation-based algorithms suffer from a loss in spatial resolution when computing the velocity field from images due primarily to the use of finite-sized interrogation windows employed during processing. The new method will utilize advanced mathematics coupled with the governing laws of fluid mechanics to form a robust, physics-inspired method that will address the shortcomings of current algorithms. The novel algorithm will allow researchers to measure velocities in turbulent flows with significantly greater accuracy, resolution, and dynamic range than has been possible to date, while using conventional experimental hardware. The vision for the end of the project is to disseminate the thoroughly tested, final version of the algorithm to the widest group of fluid dynamic experimenters as possible, after conducting a study to determine the optimal mode of distribution.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.