重力异常特征提取方法研究

Method for Extracting Feature of Gravity Anomaly

将主元分析和独立分量分析相结合应用于重力异常信号的特征提取。采用主元分析对测量数据进行去相关和降维预处理,再利用独立分量分析算法提取重力信号的特征系数。结合重力异常变化与特征系数的关系,提出了应用特征量化参数表征重力异常信号变化程度的方法。基于实测的重力异常信号进行了仿真试验,试验结果表明,本文提出的特征量化参数能有效的反映重力异常的幅值变化程度。

Principle components analysis(PCA) and independent components analysis(ICA) are used to express the feature of the gravity anomaly signals.Respectively,PCA is used to remove the correlation and reduce the dimensionality of the measurement signals and ICA is used to obtain the feature coefficients of the gravity anomaly.According to the relationship between the feature coefficients and the gravity anomaly,the quantitative parameter of the gravity anomaly features is proposed to reflect the change of the gravity anomaly.The emulation experiments based on the measurement data test the method’s feasibility,and the results show that the quantitative parameter can effectively response to the change of the gravity anomaly.