关键词: 主元分析技术, 南海, 海底沉积物, 声速经验方程

Abstract: In order to obtain a forecasting equation of sound velocity for seafloor sediments, the authors first apply the technology of principal component analysis and analyze existing empirical equation and its calculation errors. A principle component model is then built up using the measurement data of sediments from the continental slope and shelf in southern South China Sea. The authors also studied in theory how to exclude correlated parameters from a number of parameters that can affect the sound velocity and how to optimize a few parameters that are obviously related to sound velocity without cross-correlation. Using the optimal physical parameters, such as porosity n, median grain size Md and plastic limit Wp, a three-parameter forecasting equation of sound velocity is established for the continental shelf and slope in southern South China Sea. The equation is Cp =1774–5.0944 +12.499 Md +0.9985Wp. The relative forecasting error ranges only from –4.48% to 3.77%.

Key words: principle component analysis technology, South China Sea, seafloor sediment, empirical sound velocity equation