近年来，引力波成为了天文学研究的一个热门话题，引力波探测器的发展也取得了重大进展。诸如致密双星系统，恒星的坍缩和脉冲星等来源都是可能的探测对象。虽然理论上这些来源都会产生引力波，但是由于超高精度的要求，引力波探测并不是一件容易的事情。降低大型引力波干涉仪的噪声，是目前世界上有待解决的重大课题。本篇论文介绍了引力波的概念及其发展历史，同时也描述了相关的引力波源和噪声源，对主要的四个噪声：地震噪声，热噪声，引力梯度噪声，量子噪声进行了详细的分析。同时对引力波探测器的结构和原理进行了描述，它的主体是带有法布里珀罗腔的迈克尔逊干涉仪，并配有功率循环镜和信号循环镜。文中对这些引力波探测器的相关结构与技术都做了简单的介绍，在此基础上针对一种更新型的改进版干涉仪—速度计做了更为详尽的描述。速度计能在低频段有效的降低辐射压力噪声，通过搭建萨格纳克速度计能更直观有效的对引力波干涉仪有一个深入的认识。相比于迈克尔逊干涉仪，分束器的扰动会在速度计中产生更大的影响。我们使用专门设计的萨格纳克干涉仪对这个扰动带来的结果进行了一个测量，结合仿真模拟等方法，我们确定了扰动的影响与频率无关，但与幅度成正比。通过这些，之后我们便能设计反馈系统，“锁”住这个干涉仪。

In recent years, gravitational waves have become a hot topic in astronomical research, and the development of gravitational wave detectors has also made significant progress. Sources such as dense binary systems, star collapses, and pulsars are possible targets. Although in theory these sources will produce gravitational waves, gravitational wave detection is not an easy task due to the requirement of high precision. Reducing the noise of large gravitational wave interferometers is a major issue that needs to be solved in the world. This paper introduces the concept of gravitational wave and its development history. It also describes the related gravitational wave source and noise source. It analyzes the main four noises: seismic noise, thermal noise, gravitational gradient noise and quantum noise. At the same time, the structure and principle of the gravitational wave detector are described. Its main body is a Michelson interferometer with Fabry-Perot cavities, and is equipped with power circulation mirrors and signal circulation mirrors. In this paper, the related structure and technology of these gravitational wave detectors are briefly introduced. On this basis, a newer version of the improved interferometer, which is called speedmeter is described in more detail. The speedmeter can effectively reduce the radiation pressure noise in the low frequency band. By constructing the Sagnac speedmeter, it can have a more intuitive and effective understanding of the gravitational wave interferometer. Misalignment of the beam splitter will have a greater influence on the speedmeter than the Michelson interferometer. We used a specially designed Sagnac interferometer to measure the results of this misalignment. Combined with simulation and other methods, we determined that the effect of the misalignment is independent of frequency, but proportional to the amplitude. Through these, we can design a feedback system to "lock" the interferometer.