Coal-rock dynamic disaster assessment model based on electromagnetic emission and least-squares support vector machine

Abstract

A new non-linear assessment model for coal-rock dynamic disasters is reported in this paper. Coal-rock dynamic disaster assessment parameters are difficult to obtain because assessment samples are usually rare. The least-squares, a non-linear and few sample assessment methods can satisfy requirements on the availability of few samples. Based on the electromagnetic emission (EME) testing results and multiple classification theory of support vector machine, an EME least-squares support vector machine assessment model is built. In a case study, the results show that the EME least-squares support vector machine assessment model for coal and gas outburst is reliable and accurate, and can be applied to the assessment of coal-rock dynamic disasters in coal mines.