摘要
选择Na+,K+,Ca+,Mg2+,Cl-,SO42-和HCO3-这7种离子的含量作为判别因素,建立了基于分形滤波的矿井突水水源的自组织特征映射(SOFM)神经网络识别模型.实例结果表明,基于分形滤波的SOFM网络模型具有收敛快、精度高的优点,能很好地满足矿井突水水源的识别需要.
Selecting the ion contents of the seven kinds ofNa^+,K^+,Ca^+,Mg^2+,Cl^-,SO4^2-and HCO3^- as discriminating factors, a self-organizing feature map (SOFM) neural network identification model based on fractal filtering of mine water inrush is built. The re- suits show that the SOFM network model based on fractal filter has the advantages of fast convergence and high accuracy, can well meet the identification needs of the mine water inrush
出处
《华北水利水电学院学报》
2012年第5期99-102,共4页
North China Institute of Water Conservancy and Hydroelectric Power
基金
数学天元基金项目(11026196)