Abstract
Formation lithology identification is of great importance for reservoir characterization and petroleum exploration. Previous methods are based on cutting logging and well-logging data and have a significant time lag. In recent years, many machine learning methods have been applied to lithology identification by utilizing well-logging data, which may be affected by drilling fluid. Drilling string vibration data is a high-density ancillary data, and it has the advantages of low-latency, which can be acquired in real-time. Drilling string vibration data is more accessible and available compared to well-logging data in ultra-deep well drilling. Machine learning algorithms enable us to develop new lithology identification models based on these vibration data. In this study, a vibration dataset is used as the signal source, and the original vibration signal is filtered by Butterworth (BHPF). Vibration time–frequency characteristics were extracted into time–frequency images with the application of short-time Fourier transform (STFT). This paper develops lithology classification models using new data sources based on a convolutional neural network (CNN) combined with Mobilenet and ResNet. This model is used for complex formation lithology, including fine gravel sandstone, fine sandstone, and mudstone. This study also carries out related model accuracy verification and model prediction results interpretation. In order to improve the trustworthiness of decision-making results, the gradient-weighted class-activated thermal localization map is applied to interpret the results of the model. The final verification test shows that the single-sample decision time of the model is 10 ms, the test macro precision rate is 90.0%, and the macro recall rate is 89.3%. The lithology identification model based on vibration data is more efficient and accessible than others. In conclusion, the CNN model using drill string vibration supplies a superior method of lithology identification. This study provides low-latency lithology classification methods to ensure safe and fast drilling.
Highlights
Lithology classification of underground formation is of great importance in the field of oil and gas exploration engineering as lithology represents the reservoir petrophysical characteristics [1]
Lithology Identification Model Establishment Process and Framework where Mj is the set of elements that this layer needs to map; Fj l is the eigenvalue of the position j of the l th layer; kij l is the weight of the convolution kernel at the l th position ij; ∗ is convolution calculation, and it can be analogized to the weighted summation of the eigenvalues in the range of the mapping set. bj l is the extra bias added by position j at level l th layer
A new method of lithology identification relied on drill string vibration data is
Summary
Lithology classification of underground formation is of great importance in the field of oil and gas exploration engineering as lithology represents the reservoir petrophysical characteristics [1]. Imamverdiyev applied a new 1D-convolutional neural network (CNN) model by using the logging data as the input dataset in lithology identification. It is compared with four machine learning methods, and it shows a better performance in prediction than other methods [28]. Drill string vibrations are applied to make lithology identification through a convolutional neural network algorithm. Drill string datasets are applied to achieve a novel real-time lithology identification model through a convolutional neural network algorithm. The formation of a lithology identification model based on a convolutional neural network using drill string vibration data is a novel method for lithofacies prediction
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have