New experimental technique to determine coal self-ignition duration

Abstract

An artificial neural network (ANN) model was adopted to simulate the relationship between self-ignition duration and sulfur content, ash content, oxygen consumption rate, carbon monoxide as well as carbon dioxide generation rate of coal at different temperatures of self heating process. The data from spontaneous combustion experiments were used for ANN training to obtain the connection strength between nerve cells. An oil-bath programmed temperature experiment device was designed and the experimental condition and the size of the test tube were determined for testing the oxygen consumption and the gases generation rate of coal during self-heating process. The sulfur content, the ash content and the data from the oil-bath experiment were taken as ANN inputs to calculate the experiment self-ignition duration of coal. Compared with spontaneous combustion experiment, less than 1% of coal sample and 10% of time are required with an error of less than 3 days to test self-ignition duration of coal.

Keywords

coal / self-ignition duration / programmed heating experiment / artificial neural network