Weak Signal Extraction Method Research Of Microseismic Data

Because the environment of borehole microseismic monitoring is more complicated,the microseismic data is low SNR and sometimes the useful microseismic signal is covered by noise completely,which makes the work of first arrival pickup and source location and fracture interpretation great disturbance.Therefore,it has great significant to make the research of weak signal extraction method.This paper firstly introduces the observation way of microseismic monitoring and the principle of monitoring technology.And seismic focal mechanism question has been carried on the brief discussion.The noise characteristics of the borehole microseismic signal are also analyzed in detail.As we known,the borehole microseismic useful signal is a time-limited pulse.And the microseismic signal of characteristics,embodied in the "change" on the word,are mainly the variety of energy and scope of frequency band.Also,it often appears the noise interference of strong energy.And then due to the weak energy microseismic events and the characteristics of low SNR data,we proposed entropy and mutual information method.Through studying the features of random variables and correlation degree,we proposed a first arrival pickup based on energy and mutual information.At the beginning,we use energy ratio algorithm to pick a rough arrival time,and then we utilize the mutual information algorithm to accurately pick up arrival time.According to the test of model and the analysis of real data in comparison with conventional method,we effectively verify the accuracy and feasibility of the method.And it achieves the automatic first arrival pickup accurately and rapidly.And due to the low SNR of microseismic data,it is very difficult to clearly identify the P-wave and S-wave.According to the characteristics of randomness and non-stationary of microseismic signals and the advantages of high resolution of time-frequency and signal reconstruction on synchrosqueezing transform(SST),we research the method to extract theweak signal of microseismic data based on SST.First,we utilize the SST to conduct the adaptive threshold denoising.Then,the SST coefficients are extracted near the center frequency of effective signal by integrating.Finally,we carry out the SST reconstruction using the extracted effective SST coefficients to implement the weak signal extraction.The test results of the synthesis non-stationary signal models with the different noise intensity and the actual borehole microseismic data show that this method presented in this paper has better noise immunity and higher signal extraction accuracy.Finally,because microseismic data has characteristics of randomness,non-stationary and time-frequency coupling and the empirical mode decomposition(EMD)has the problem of mode mixing,we proposed the microseismic data denoising method based on EMD mutual information entropy and synchrosqueezing transform.First,we utilize the EMD to decompose the microseismic data to obtain the component of IMF that is arranged from high frequency to low frequency.Second,the mutual information entropy of each adjacent IMF component can be calculated to identify the boundary of the high frequency and low frequency part.At last,we apply the SST to remove the aliasing noise in the high frequency part.And then the filtered high frequency part is reconstructed with low frequency part to achieve the microseismic data denoising.According to the synthesis of different noise intensity non-stationary signal model and the actual microseismic data and compared with removing the high frequency part filtering directly show that this method can remove the aliasing noise and extract the useful signal from the noise effectively,which improves the data of SNR.