Neural-network-based acquisition method of seismic first arrival wave travel time

Abstract

The invention discloses a neural-network-based acquisition method of seismic first arrival wave travel time. The method comprises: original seismic waveform data are obtained, the obtained data are processed into a data set including an original waveform and a corresponding marking point, and the data set is divided into a training data set and a testing data set; according to a first arrival wavetravel time data obtaining process, a structure of a neural network is determined by taking simulation of a manual pick-up result as a target and combining features of the original seismic waveform data; the neural network is trained by using the training data set as an input of the neural network and the corresponding marking point as an output of the neural network, the trained neural network is tested by using the testing data set, and if a precision requirement is met, a trained neural network is obtained; and the marking point of the original seismic waveform data is obtained automatically based on the trained neural network and thus seismic first arrival wave travel time is obtained. Therefore, the accuracy of automatic acquisition of the seismic first arrival wave travel time is improved and the manual correction workload is reduced.

Description

technical field [0001] The invention relates to the technical field of geophysical exploration, in particular to a neural network-based method for acquiring travel time of first arrival waves of earthquakes. Background technique [0002] There are two main methods for obtaining the travel time of the first arrival of a traditional earthquake: 1) The energy ratio method determines the travel time of the first arrival by calculating the energy ratio between each sampling point of the seismic data and combining the waveform and energy characteristics of the first arrival. 2) Image edge detection method. By using the differential operator to extract the edge of the digital image from the collected seismic data, the travel time of the first arrival wave is determined according to the peak position. [0003] However, the above two traditional methods are based on a definite algorithm plus manual correction to obtain the picked first arrival travel time data. Because the accuracy ...

AI Technical Summary

Problems solved by technology

[0003] However, the above two traditional methods are based on a definite algorithm plus manual correction to obtain the first arrival travel time data picked up. Because the accuracy of the algorithm is limited, the workload of manual correction is relatively large, and researchers usually need to manually extract shot by shot.

Images