A Method for Quantitative Prediction of Coal Thickness Using Empirical Mode Decomposition and Support Vector Machine

Abstract

The invention discloses a method for quantitatively predicting coal thickness by using empirical mode decomposition and support vector machine. Firstly, the reflection coefficient and synthetic seismic records are calculated according to the longitudinal wave velocity and density of logging data, and noise is added to the synthetic seismic records; The mode decomposition method is to perform empirical mode decomposition on the synthetic seismic record after adding noise; then calculate the correlation coefficient between each eigenmode function obtained after the empirical mode decomposition and the synthetic seismic record without adding noise, and determine the basic data; The seismic attributes are extracted from the basic data, and normalized preprocessing is carried out on each seismic attribute; then the seismic attributes for quantitative prediction are selected by using the gray correlation degree method; finally, the support vector machine is used for learning and training, and the non-destructive analysis of the whole work area is carried out. Linear quantitative prediction of coal thickness. This method only needs 3D seismic data and drilling data, and can realize the non-linear quantitative prediction of coal thickness in the whole work area, which can provide strong geological guarantee for coal mines.

Description

technical field [0001] The invention relates to a method for quantitatively predicting coal thickness by using empirical mode decomposition and support vector machine, and predicts the coal thickness in a coal mining area with three-dimensional seismic exploration data. Background technique [0002] In the construction and production process of modern large-scale mines, the thickness of coal seam is an indispensable data for the calculation of coal reserves and the reasonable layout of roadways. According to relevant statistical results, if the actual coal thickness is 10% to 20% thinner than the original design coal thickness, then the coal output will drop by 35% to 40%. The coal seam is a thin layer in seismic exploration, and the quantitative prediction of the thickness of the thin layer has always been one of the recognized problems. At present, the commonly used method of drilling interpolation to predict coal thickness cannot guarantee the accuracy of coal thickness ...

AI Technical Summary

Problems solved by technology

At present, the commonly used method of drilling interpolation to predict coal thickness cannot guarantee the accuracy of coal thickness prediction far away from the drilling location; seismic wave amplitude or frequency domain parameters to predict coal thickness (such as tuning method, spectral moment method, etc.) are affected by the signal noise of seismic data. The influence of ratio and fidelity is large, and the multi-solution is strong. Generally, only the lateral variation trend of coal thickness can be predicted, and it is difficult to obtain more accurate quantitative prediction results.

Images