粒子群算法在煤与瓦斯突出危险预测中的应用

Application of Particle Swarm Optimization in Risk Prediction of Coal and Gas Outburst

煤与瓦斯突出危险是影响煤矿生产安全的一个重大问题,为了解决危险预测的问题,将反映煤与瓦斯突出的六个指标:垂深、倾角、巷道类型、煤层厚度、地质构造和作业方式作为输入层参数,使用BP神经网络与粒子群算法结合建立模型,导入数据到Matlab中进行模拟仿真,将预测结果与实际情况相对比。结果表明:粒子群算法结合神经网络对预测煤与瓦斯突出危险是有效的,相较于传统预测方法,其预测的速度、精度都有所提升,可以将该算法应用到突出危险预测当中。

Coal and gas outburst is one of major issues threatening the safety of coal mine production.In order to realize risk prediction,we chose six indicatorsof coal and gas outburst as input parameters,including vertical depth,dip angle,roadway type,coal seamthickness,geological structure and operation mode.A model was established based on BP neural network and particle swarm optimization(PSO).The data were imported into Matlab for simulation and then the predicted results were compared to the actual situation.The results showed that the combination of particle swarm optimization and neural network was effective in the prediction.Compared with the traditional methods,its prediction speed and accuracy have been improved and the algorithm could be used in the prediction of the coal and gas outburst.