多尺度特征融合的岩性识别模型

打开文本图片集

中图分类号：TE122 文献标志码：A 文章编号：1672-9315（2025）05-0904-12

DOI：10. 13800/j. cnki. xakjdxxb. 2025.0507

A multi-scale feature fusion model for lithology identification

JIA Pengtao1， CHENG Yuchao¹， JIANG Yongjie²，LI Na1

1.CollgeofAtificialIntelligence&ComputerScence，Xi'anUniversityofcienceandTechnologyXi’an7O4China; 2.Shaanxi Coal Industry Group Huangling Jianzhuang Mining Co.，Ltd.， Yan’an7273oo， China）

Abstract： To address the issues in the existing lithology recognition methods，such as insufficient feature representation，limited model generalization ability and low recognition accuracy，a novel lithology recognition model based on multi-scale feature fusion was proposed， named MSFR-Net.A Feature Extraction Module（FEM） was designed from the raw logging curves. It combined Bidirectional Gated Recurrent Units（BiGRU） and an even-odd sequence interaction mechanism to perform deep mining of multi-scale features from the logging curves. A Random Convolution Module was constructed， which utilized a dynamic convolution kernel parameter optimization strategy to efectively capture spatial correlations between strata features.Based on a decision module consisting of six base classifiers，a multi-classifier collaborative decision mechanism was employed to improve the model’ s classification performance and complete the lithology recognition task.The results show that MSFR-Net performs beter in such key metrics asaccuracy，precision，recall，and F1-score in tests based on real logging data，compared to ten commonly used lithology recognition models，such as SVM and GRU.In the D-well experiment， MSFR-Net achieves a prediction accuracy of 95.1% for major lithology categories and 72.7% for minor lithology categories，an improvement of 13.27% over SVM and 12.61% over LSTM.MSFR-Net， through the synergistic optimization of multi-scale feature fusion and ensemble learning strategies，effectively enhances the extraction of key geological features and the model’ s generalization performance， providing a new technical approach for lithology recognition.

Key Words： well log curves; lithology identification; feature fusion; Rocket network; multi-clasifier

0 引言

岩性智能识别是储层表征与油气资源评价的核心技术之一。（剩余17647字）