论文部分内容阅读
模式识别在定量研究岩相模式中是一种有效的方法，神经网络是模式识别的一种新方法，和其它模式识别方法不同之处在于它能模拟人脑并行处理信息的模式和神经系统的可塑性。经传统岩相分析，鄂中拗陷二叠系碳酸盐岩确认为碳酸盐岩台坪相，它包含五个亚相，即局限台洼、台坡Ｂ、台洼、台坡Ａ和台滩相。它们在岩石的颜色、矿物组分、古生物种属、结构等特征上都有不同的差异。应用ＢＰ神经网络，特别是与模糊模式识别的结合，对拗陷区的岩相识别是成功的，令人鼓舞，两种方法各自的正确判对率约为７５％，而综合两种方法的判对结果，则岩相全部判对
Pattern recognition is an effective method for quantitatively studying facies model. Neural network is a new method of pattern recognition. It differs from other pattern recognition methods in that it can simulate the pattern that human brain processes information in parallel and the neural system plasticity. According to the analysis of the traditional lithofacies, the Permian carbonate rocks in the middle and lower reaches of Hubei Province are confirmed as carbonate platform facies. They include five subfacies, namely, the boundary depression Tai, Tai Po B, Tai Wa, Tai Po A and Taiwan beach phase. They have different characteristics in terms of rock color, mineral composition, paleontological species and structure. The combination of BP neural network, especially with fuzzy pattern recognition, is successful for facies identification of depressed areas. It is encouraging that the correct rate of each of the two methods is about 75%. However, combining the two methods Judgment of the results, all sentenced to rock facies