Gas insulated switchgear (GIS) is widely used in power transmission and distribution systems due to its advantages of land saving, high operation reliability, strong environmental adaptability, simple installation and transportation, and low maintenance cost. Abnormal vibration induced by mechanical defects such as poor contact, loosen bolt and insulator crack. Precise vibration detection, signal processing and fault identification are essential for reliability operation of GIS equipment. This work summary the progress and prospect of vibration-based GIS mechanical defect diagnosis techniques. Contacted acceleration sensors or contactless laser Doppler vibrometer are adopted to collect GIS vibration data, and analyze the raw vibration data in the time domain and frequency domain to extract the statistical magnitude, signal process algorithms can obtain more clearly representation of vibration waveform, and artificial intelligence could effective extract high-dimensional features thus realized high accuracy mechanical fault classification.