遗传算法的BP网络模型进行瓦斯涌出量预测

1.西安科技大学 地质与环境学院,陕西 西安 710054; 2.煤炭科学研究总院 西安研究院,陕西 西安 710054

遗传算法; BP网络; 采煤工作面; 瓦斯涌出量; 预测

Prediction of gas emission quantity of mining faces based on genetic BP neutral network optimal model
WANG Sheng-quan1,LIU Bo-gen2,ZHANG Zhao-zhao1,FAN Qi1,FENG Hai1

(1.College of Geology and Environment,Xi'an University of Science and Technology, Xi'an 710054, China; 2. Xi'an Branch, Coal Science Research Institute, Xi'an 710054, China)

prediction; gas emission quantity; mining faces; BP neutral network; genetic algorithm

备注

从提高采煤工作面瓦斯涌出量预测的速度和精度入手,将遗传算法与神经网络2种非线性最优化算法的优势加以融合,提出了一种利用遗传算法同时优化BP网络的连接权和拓扑结构的网络模型,并以韩城下峪口煤矿为例,进行了实际应用。结果 表明:改进后的BP网络模型预测精度较高,具有良好的应用前景。

Starting with the improving of the prediction speed and precise of gas emission quantity at mining faces, the authors put forward a BP neutral network model for optimizing both the network link weight and topologic structure with a genetic algorithm. Taking Hancheng Xiayukou coal mine as an example, a practical application method is given. The results shows the model is of a higher precision and application prospect.