流速是确定水流运动特性的基本参量，在水力学相关的研究中占有重要地位，是理论分析和实验验证的重要指标。开发适于水力学研究的流速测量方法与技术是流体力学研究的重要课题。近年来，随着科技水平提高，流速测量方法得到了较快的发展。电磁技术、超声波技术、激光技术、图像处理技术等先进技术不断被应用到流速测量领域，实现了从一维到三维，从稳态到瞬态，从局部单点到全流场流速测量的跨越。LSPIV方法是近年发展起来的能够实现瞬时、三维、全流场流速测量的技术，且随着时代发展，该方法呈现出十分明显的实时化、便携化发展的趋势，在科学研究和实验中应用广泛。但传统的LSPIV装置比较复杂昂贵，前期布置花费时间长，携带不便，测量过程麻烦，野外测量实用性较低。论文旨在开发适用于野外的，能快速实现流速测量的系统。首先，论文基于LSPIV技术的基本原理开发了适用于移动端（智能手机，iPhone 6s）的流速测量系统LSPIV-T。系统借助标识物实现水体表面点速度测量，测量频率为5HZ，具有实时、廉价、便携等优势。系统使用了修正的归一化RGB颜色模型进行标识物识别，通过共线公式建立像平面直角坐标系与空间直角坐标系的关系；并调用陀螺仪、加速度计、GPS等内置传感器获取计算所需外参数，通过相机标定实验获取计算所需内参数。LSPIV-T系统在人机交互机制的基础上实现了模型的自动更新。其次，本文设计了不同工况来验证程序的准确性。在能够有效进行标识物识别的前提下，程序测量序列的平均值与真实值的相对偏差不超过3%。在此基础上，论文率定了系统的适用范围。经过实验，LSPIV-T系统的量程在0.015~1.55m/s，有效测距范围0.4~2.0m，可接受的俯仰角和翻转角不超过15°，可接受任意角度的水平旋转角。论文还对LSPIV-T系统进行了误差分析，确定高程差的测量误差是程序误差的主要来源。最后，论文在野外对LSPIV-T系统进行了实验，并采用声波多普勒测速仪（ADCP）作对照实验。实验在万泉河清华大学段举行，结果显示LSPIV-T系统能够实现野外环境下的水体表面流速测量，与ADCP相比测量误差不超过15%。
Velocity is the basic parameter to determine the characteristics of water flow, which is an important indicator for theoretical analysis and experimental verification. It plays an important role in hydraulic research. Inventing new techniques suitable for flow velocity measurement is an important subject of hydrodynamic research. Recently, with the improvement of technology, flow velocity measurement method has been developed rapidly. Lots of new technologies, such as ultrasonic technology, laser technology, digital image processing technology and so on, have been applied to the field of velocimetry. It makes the measurement from one dimension to three, steady to transient, single point to full flow field come true. Large-scale particle image velocimetry(LSPIV), which can achieve three-dimensional and full-flow velocity measurement, is a new technology developed in recent years. As time going on, the method presents a very obvious real-time, portable development trend. It was used in scientific research and experimentation widely. However, the traditional LSPIV device is more complex, and the process of pre-layout takes a long time. Besides, LSPIV system is very expensive and inconvenient to carry. It lacks practicality in the field measurement process.The paper aims to develop a field system that can achieve flow velocity measurements quickly. Firstly, the flow velocity measurement system was developed, LSPIV-T, which is suitable for smartphone (iPhone 6s) based on the basic principle of LSPIV. The system measures the surface flow velocity by means of a tracer floating on the surface. It is a real-time, inexpensive and portable system. The video is recorded at the frequency of 5HZ. The system uses the normalized RGB color model to identify the tracer, and establishes the relationship between the imaging plane coordinate system and the space Cartesian coordinate system by collinear formula. To complete the velocimetry, the system calls the gyroscope, accelerometer, GPS and other sensors to obtain the required external parameters, uses the camera calibration experiment to obtain the internal parameters.Secondly, we designed different operating conditions to verify the accuracy and scope of the program. The relative deviation of the measurement errors is no more than 3% under the condition that the system can identify the tracer effectively. Experimental results demonstrate that the measurement range of LSPIV-T system is 0.015~1.55m/s, the range of effective recording distance is 0.4~2.0m, and the acceptable pitch angle and flip angle should be not more than 15°. Error analysis shows that the measurement error of the elevation difference is the main source of the program.Finally, the LSPIV-T system was tested in a real river, along with the surface velocity radar (SVR). The experiment was held at Wanquan River. The LSPIV-T system can achieve the measurement of velocity of the water surface, the measurement error is not more than 15% compared with the SVR.
