Microseismic p-wave first-arrival picking method and system based on capsule neural network

Abstract

The invention relates to the technical field of microseismic data processing, in particular to a method and system for picking up the first arrival of microseismic P-waves based on a capsule neural network; the method includes preparing an original data set; making a data training set; selecting the original data set The first arrival point of a part of the sample signal is labeled as the labeled part, and the other part of the sample signal is regarded as the unlabeled part; the data training set is input into the combined training model to predict and evaluate the characteristics of the microseismic signal; The microseismic signal features are used for target detection, and the first arrival point of the microseismic signal is obtained; the system includes a data acquisition module, a data training set making module, a data training module and an output module; the embodiment of the present invention uses a capsule neural network and a Combined with semi-supervised learning, the RPN network is used to detect the microseismic signal, realize the first arrival point picking of the microseismic signal, improve the accuracy of the feature extraction of the microseismic signal and the accurate picking of the P wave first arrival point.

Description

technical field [0001] The invention relates to the technical field of microseismic data processing, in particular to a method and system for picking up the first arrival of microseismic P-waves based on a capsule neural network. Background technique [0002] Effective detection of microseismic signals is of great significance to stable and high production in oilfield development. Usually, the effective signal energy of microseismic is weak, the signal-to-noise ratio is low, or even completely submerged in the noise. Although there are many conventional seismic data processing methods, if they are directly applied to microseismic data, satisfactory results are often not obtained, which will directly affect the quality and accuracy of microseismic monitoring. Therefore, finding a suitable method to identify weak and effective signals in microseismic data is the key to the processing and interpretation of microseismic data. [0003] In 2018, the method of picking up the firs...

AI Technical Summary

Problems solved by technology

[0004] The disadvantage of the existing technology is that although this method can improve the signal recognition rate to a certain extent, the accuracy rate of extracting microseismic signal features is not high, resulting in only 73.5% of the accuracy rate of P wave first arrival point picking

Images