满足数据同一性的强震动记录去噪滤波后处理输出方法

Strong motion records are widely applied in engineering at present. However, there are few studies in the world focusing on the data output after the records processed by high-pass acausal filtering, and there is no literature reporting this issue in China yet. In this study, a case study of typical record is illustrated on the physics of baseline drift of displacement time series resulted from integrating pad-stripped accelerogram directly. Then a new procedure of post-processing output for the pad-stripped strong motion records is proposed, which has advantages that it could correct the baseline of displacement time series and output the compatible data. The six-order polynomial is suggested to be an optimal function used for fitting the baseline of displacement time series by comparing the decision coefficients between the baselines fitted by 2-8 orders polynomial functions and that fitted by linear function. An implicit method tapering the end of time series is developed, which has advantages that it can achieve automatically tapering the end of velocity and displacement time series when making an integration of acceleration time series. 112 strong motion records collected in Lushan earthquake in 2013 are used to constitute a comparison of their displacement time series, PGD and d_rms values between the derived data from the output data of proposed procedure and the pad-stripped data. Moreover, some other acceleration-related commonly-used IM parameters are also involved in the comparison. The results show that the procedure proposed in this study is effective and accurate in post-processing output data. It could help to guide and normalize the data users to have an appropriate way for outputting strong motion records after filtering processing.