基于随机森林结合地球物理测井资料的煤体结构识别方法及应用

在更加复杂的地质因素影响下,常规测井方法识别煤体结构准确度低,为精确识别煤体结构,研究了煤体结构测井曲线响应机理以及随机森林决策树个数的优选,从而建立煤体结构与测井曲线的随机森林分类模型进行煤体结构识别。结果表明:决策树个数为500时,随机森林分类模型效果最佳;通过袋外误差和模型对测试集样本的预测结果可知,随机森林分类模型的结果稳定且泛化性强,并且适合处理非均衡数据,预测精度较高。可见随机森林算法能有效识别煤体结构,为煤层气开发提供帮助。

Coal structure identification method based on random forest combined with geophysical logging data and its application

Under the influence of more complex geological factors, conventional logging methods have low accuracy in identifying coal structure. In order to accurately identify coal structure, the response mechanism of coal structure logging curve and the optimization of the number of random forest decision trees are used to investigate The random forest classification model of coal structure and logging curve is used to identify coal structure. The results show that: when the number of decision trees is 500, the random forest classification model has the best effect; through the out-of-bag error and the model''s prediction results on the test set samples, the results of the random forest classification model are stable and generalizable, and extremely suitable Processing unbalanced data, with high prediction accuracy. It can be seen that the random forest algorithm can effectively identify the coal structure and provide help for the development of coalbed methane.