Study On Data Processing Technology Based On Strong Motion Records Of Lushan Earthquake

Strong motion records processing is the basic work of the research on earthquake engineering and engineering earthquake,which could provide scientific data for seismic fortification,earthquake early warning and rapid intensity and strong motion observation of important engineering.But because of instruments and environment,the noise in strong motion records is inevitable.So it is very important to study the filtering technique in strong motion records processing.In this paper,we studied the Butterworth(IIR)causal filtering technique and window function(FIR)causal filtering technique,compared with the Butterworth(IIR)acausal filter technique.And based on these filtering techniques,three kinds of integral method were analyzed.In addition,strong motion records are non-stationary signals.It is focus in recent years to deal with it by using time-frequency analysis technology.In this paper,a new time-frequency analysis method called ESMD,which was proposed in 2013,was introduced into the field of strong motion records processing.In this paper,it was compared with EMD method to obtain useful information from strong motion records.This paper focuses on the above issues,the main research is following.1.The processing methods of strong motion records at home and abroad was introduced.This paper defined the principle of the filter corner frequency must be determined according to the Nyquist sampling theorem,the instrument and the strong motion record information.Searching frequency method to determine the corner frequency was proposed,and its deviation was analyzed based on the strong motion records of Lushan earthquake.Then,the corner frequency of 204 strong motion records of Lushan earthquake were analyzed.The records were classified by the site conditions and epicenter distance.Then,we summarized the characteristics of the passband,and analyzed the influence of ground motion parameters on the range of passband.2.Butterworth(IIR)causal filtering technique was studied based on strong motion records of Lushan earthquake.The flow chart of the system and the method of determining the performance parameters were given.The differences between filtering through orders and filtering through transition band were analyzed.The effects of all performance parameters on the filtering results were counted.And it was compared with the Butterworth(IIR)acausal filtering method.Then,we compared the results of three methods of integration(Trapezoidal Rule,FFT,Kinoshita's Lattice Filter).At last,the reasonable suggestion was given.3.The window function(FIR)causal filtering technique was studied based on strong motion records of Lushan earthquake.The paper focused on 7 kinds of window functions.The effects of the parameters such as window length,maximum ripple and side lobe attenuation on the filtering effect and ground motion parameters were studied.And it was compared with the Butterworth(IIR)acausal filtering method and the Butterworth(IIR)causal filtering method.Three integration methods were also analyzed by using with different window function(FIR)causal filtering.At last,the reasonable suggestion was given.4.A new time-frequency analysis technique method called ESMD was applied in strong motion data processing.Its principle was analyzed.And based on strong motion records of Lushan earthquake,it was compared with the EMD method.We analyzed the characteristics of each method.By using the method of filtering and time-frequency analysis,strong motion records denoising and base-line correction has been realized.