论文部分内容阅读
自组织Ｋｏｈｏｎｅｎ网络是一种学习速度很快的神经网络，可以用于分类、聚类、解释等问题。本文依据奥灰岩地震波运动学和动力学特征，提取时间域最大互相关系数、分形关联维、频率域主频、频带宽度和主频带能量共５个参数，利用自组织（Ｓｅｌｆ－Ｏｒｇａｎｉｚｉｎｇ）Ｋｏｈｏｎｅｎ人工神经网络横向预测含水裂隙发育带。试算结果表明，方法可行，可望成为预测奥灰岩岩溶裂隙发育带的一种有效方法。
Self-organizing Kohonen network is a fast learning neural network that can be used for classification, clustering, interpretation and other issues. Based on the kinematic and kinetic characteristics of the limestone seismic wave, this paper extracts five parameters of maximum cross-correlation coefficient, fractal correlation dimension, frequency domain frequency, frequency bandwidth and main band energy, using self-organizing (Self-Organizing) Kohonen Artificial Neural Network Lateral Prediction of Water - bearing Fissure Development Zone. The test results show that the method is feasible and it is expected to be an effective method to predict the karst fracture zone in the Ordovician limestone.