针对地震单炮的地球物理特征,将单炮划分为三个区域,利用各个区域的信号特征作为神经网络的输入神经元,通过网络的组织学习能力,来预测废道的类型,之后通过相关性等道间属性来对预测的废道进行二次判别。该方法提高了废道识别的准确率,有效的降低了网络的学习时间,具有更明显的地球物理特征。实际资料应用证明该方法具有良好的适应性。 According to the geophysical characteristics of a single seismic shot, the single shot is divided into three regions, the signal characteristics of each region are used as the input neurons of the neural network, and the type of the waste track is predicted through the organizational learning ability of the network. After that, Inter-channel properties to predict the waste of secondary discrimination. The method improves the accuracy of waste lane identification, effectively reduces the learning time of the network and has more obvious geophysical characteristics. The actual data application proves that the method has good adaptability.