A method and system for denoising and reconstructing microseismic signals based on multi-scale decomposition

CN120195735BActive Publication Date: 2026-06-30CHINA UNIV OF MINING & TECH

Patent Information

Authority / Receiving Office: CN · China
Patent Type: Patents(China)
Current Assignee / Owner: CHINA UNIV OF MINING & TECH
Filing Date: 2025-03-20
Publication Date: 2026-06-30

AI Technical Summary

Technical Problem

Existing technologies struggle to achieve comprehensive noise reduction when processing microseismic signals due to multi-scale noise. Traditional methods suffer from limited processing capabilities and lack targeted decomposition, leading to a decline in signal quality.

Method used

A multi-scale decomposition method for microseismic signal denoising and reconstruction is adopted. Local features of the signal are extracted by Shapelet, and signal denoising and reconstruction are performed by combining temporal convolutional network and U-Net network. A multi-scale feature matrix is constructed, the importance of signal blocks is evaluated by attention mechanism, and the model training is optimized by loss function.

Benefits of technology

It achieves efficient denoising of microseismic signals, retains key feature information, improves the precision of signal processing and the accuracy of mine slope stability assessment, and generates more accurate and reliable denoised signal output.

✦ Generated by Eureka AI based on patent content.

Smart Images

Figure CN120195735B_ABST

Patent Text Reader

Abstract

This invention relates to the field of signal denoising and reconstruction technology, specifically to a method and system for denoising and reconstructing microseismic signals based on multi-scale decomposition. The method includes: S1, performing multi-scale shapelet decomposition on noisy microseismic signals based on different sequence lengths, measuring the matching degree between each signal block and all candidate shapelets using Euclidean distance, and constructing a multi-scale feature matrix M; S2, based on the multi-scale feature matrix M, performing multi-scale weighted importance measurement using a temporal convolutional network and attention mechanism to evaluate the importance of each signal block in the denoising process; S3, constructing a U-Net microseismic signal denoising model, designing a loss function by combining reconstruction error and dynamic time warping, and iteratively training the U-Net-based microseismic signal denoising model multiple times until the error meets preset requirements. This invention can effectively identify and remove noise, especially in data containing different levels of noise, and this method helps to achieve high-quality denoising of microseismic monitoring data.

Need to check novelty before this filing date? Find Prior Art