Tunnel engineering electromagnetic wave multi-domain feature intelligent extraction method, device and storage device

Abstract

The application provides a tunnel engineering electromagnetic wave multi-domain feature intelligent extraction method, equipment and storage equipment, first, the detection results of the geological radar are used to extract the time domain features of the electromagnetic wave, that is, the average relative amplitude; then the detection results of the geological radar are used to extract the time-frequency domain features of the electromagnetic wave, that is, the average maximum specific frequency; finally, the relative amplitude matrix of the electromagnetic wave is used to draw a gray scale diagram, and the spatial domain features of the electromagnetic wave, that is, the maximum common axis length, are extracted therefrom. Finally, for each tunnel face, the multi-domain features of the mileage point and the three mileage points adjacent to the mileage point are extracted, and finally, the electromagnetic wave multi-domain feature library is constructed, and the multi-domain features are used for the advanced geological prediction of the rock mass in front of the tunnel face. The application has the beneficial effects that: higher modeling accuracy and stronger generalization ability can be obtained, and a good foundation is laid for the multi-source fusion advanced geological prediction research of the tunnel engineering.

Images