Bedrock depth prediction method and system based on generative adversarial network and environmental element data

Abstract

The invention discloses a bedrock depth prediction method and system based on a generative adversarial network and environmental element data. The method of the invention includes pre-designing and generating a generative adversarial network including a generator G and a discriminator D and completing the training; A set of environmental element data as the center is input into the trained generator G, and the bedrock depth data y corresponding to the predicted position is obtained through the forward inference of the generator G g . Based on the strong fitting ability of the generative confrontation network, the invention can effectively utilize more environmental information and obtain a more accurate prediction model with better generalization ability when it is applied to bedrock depth prediction. The model structure of the generative adversarial network gives full play to the fitting potential of the generative adversarial network, and realizes the bedrock depth prediction with extremely high accuracy. It has the advantages of simple data sampling, fast prediction speed and high prediction accuracy.

Description

technical field [0001] The invention belongs to the field of geographic data processing, and in particular relates to a bedrock depth prediction method and system based on generative confrontation network and environmental element data. Background technique [0002] Bed rock refers to the destruction of minerals originally formed under high temperature and high pressure after weathering occurs, and some new minerals that are relatively stable under normal temperature and pressure are formed, which constitute the continental crust surface weathering layer, and the complete surface under the weathering layer is formed. The rock is called bedrock, and the bedrock outcropping is called outcrop. Bedrock is the hard rock layer in the surface of the land. Generally, they are mostly covered by soil layers, and the burial depth varies, ranging from a few meters to tens of meters, and as many as hundreds of meters. It is composed of one or several kinds of rocks among sedimentary ro...

AI Technical Summary

Problems solved by technology

Current prediction methods based on simple machine learning or statistics have low prediction accuracy, rely on complex feature engineering, and have limited generalization capabilities

Images