Design And Implementation Of A Real-time Hyper-speed Vision System Oriented To Unsteady Flow Field Measurement

Unsteady flow, which has the characteristics of time-varying velocity and low measurement accuracy, is playing a more and more important role in the fields of missile, vessels, hydraulic engineering, etc. Among all the flow field measurement techniques, PIV(Particle Image Velocimetry), which has the features of non-contact, full field and high precision, has become one of the most widely used technology in unsteady flow field measurement field. This paper focuses on the measurement of unsteady flow and presents a real-time adaptive measurement mechanism, on the basis of which, a hyper-speed real-time vision system is designed and implemented. The main research work is as follows.(1) According to characteristics of unsteady flow, a real-time adaptive adjustment measurement mechanism is presented. In the real-time adaptive adjustment mechanism, velocity of flow field at present moment is calculated in real time, based on which, velocity of the next moment can be predicted and a proper time interval of next time be worked out.(2) A hyper-speed image acquisition system is designed and implemented based on frame-straddling technique. In the acquisition system, two scientific CMOS sensors is adopted as the acquisition device and a FPGA(Field Programmable Gate Array) is used as the host controller. The frame-straddling technique makes the image acquisition system get rid of speed limits of sensors, thus realizing image acquisition with lower time interval.(3) A hyper-speed image processing cell is designed and implemented based on distributed processor architecture. In the processing cell, a SRIO(Serial RapidIO) data communication network is established based on a switch chip. Through the SRIO network, multi-processors can realize high-speed communication and task allocation of complex algorithms.(4)The hyper-speed vision system composed of hyper-speed image acquisition cell and hyper-speed image processing cell can realize 20 Hz particle images acquisition and real-time processing. At the end the article, a flow field experiment is designed to verify the effectiveness of the real-time adaptive measurement mechanism and the hyper-speed real-time vision system.