Model of coal gas permeability prediction based on PSO-LSSVM

Three main influential factors of coal permeability were summarized, which were effective stress, temperature and gas pressure. The least square support vector machine was applied to predict permeability. The three factors and compressive strength were used as the input, the permeability as target output. Particle swarm optimization algorithm was used to optimize the parameters of least square support vector machine to improve prediction precision. PSO-LSSVM was compared with BP neural network and SVM by a test. The comparative experiment results show that PSO-LSSVM can be used to predict permeability with high accuracy.