Coal Quality Exploration Technology Based on an Incremental Multilayer Extreme Learning Machine and Remote Sensing Images

Abstract

This paper proposes a new coal quality exploration method that detects coal quality in coal mining areas and explores and monitors the distribution and change of coal through remote sensing images. First, we collected a large number of coal and noncoal samples such as sandstones, shales, and coal gangues. Second, we measured the actual spectral data of these samples using a spectrometer. For coal mines, we used the chemical analysis method to quantify coal’s fixed carbon and categorize the coal mines into three types based on the fixed carbon content present in coal. Third, we collected satellite remote sensing images of coal mining areas and established spectral data relations between the measured spectral data of the samples and the remote sensing images. Fourth, we proposed an incremental multilayer learning machine algorithm and used the algorithm combined with spectral data to build a coal quality classification model to identify coal quality in remote sensing images. Finally, the model accurately described the distribution map of coal quality. Compared with traditional coal exploration methods, this method has the advantages of high speed, high accuracy, and low price.