Abstract:Well logging is a critical tool for reservoir evaluation and fluid identification. However, due to the expanding borehole, instrument failure, economic constraints, etc., some types of well logs are occasionally missing or unreliable. Existing logging curve reconstruction methods based on empirical formulas and fully connected deep neural networks (FCDNN) can only consider point-to-point mapping relationships. Recurrently structured neural networks can consider a multipoint correlation, but it is difficult to compute in parallel. To take into account the correlation between log sequences and achieve computational parallelism, we propose a novel deep learning framework for missing well log reconstruction based on a state-of-the-art transformer architecture (MWLT). The MWLT utilizes a self-attentive mechanism instead of a circular recursive structure to model the global dependencies of the inputs and outputs. To employ different usage requirements, we design the MWLT in three scales: small, base, and large, by adjusting the parameters in the network. 8609 samples from 209 wells in the Sichuan Basin, China, are utilized for training and validation, and two additional blind wells are used for testing. The data augmentation strategy with random starting points is implemented to increase the robustness of the model. The results show that our proposed MWLT achieves a significant improvement in accuracy over the conventional Gardner’s equation and FCDNN, both on the validation dataset and blind test wells. The MWLT-Large and MWLT-Base have lower prediction errors than MWLT-Small but require more training time. Two wells in the Songliao Basin, China, are employed to evaluate the cross-regional generalized performance of the proposed method. The generalizability test results demonstrate that density logs reconstructed by MWLT remain the best match to the truth compared to other methods. The parallelizable MWLT automatically learns the global dependence of the subsurface reservoir, enabling an efficient and advanced missing well log reconstruction performance.

Logging Curve Restoration Based on Correlation Information Mining

Well logging curve reconstruction based on kernel ridge regression

Application of Neural-network Technology in Reconstructing Logging Curves

An Intelligent Depth Correction Method for Logging Curves Based on Pearson Correlation Coefficient and DTW

Restoring Method for Missing Data of Spatial Structural Stress Monitoring Based on Correlation

Well Logs Reconstruction Based on Deep Learning Technology

Stratigraphic Subdivision-Based Logging Curves Generation Using Neural Random Forests

Well Logging Reconstruction Based on Bidirectional GRU

Using Data Mining to Identify Carbonate Lithology

Efficient selective attention LSTM for well log curve synthesis

Visual Analytics of Stratigraphic Correlation for Multi-Attribute Well-Logging Data Exploration

Empirically Informed Convolutional Neural Network Model for Logging Curve Calibration

Comparative Study and Application of Logging Standardization Methods

Well Logging Reconstruction Based on a Temporal Convolutional Network and Bidirectional Gated Recurrent Unit Network with Attention Mechanism Optimized by Improved Sand Cat Swarm Optimization

Missing Well Log Reconstruction Using A Sequence Self-attention Deep Learning Framework

Studies on Application of Logging & well Logging Lithologic Data in Reservoir Inversion

Application of conventional logging interpretation fracture method based on neural network in offshore Oilfield L

Well-Logging Prediction Based on Hybrid Neural Network Model

Missing Data Interpolation in Well Logs Based on Generative Adversarial Network and Improved Krill Herd Algorithm

Deep-learning missing well-log prediction via long short-term memory network with attention-period mechanism

Automated gamma-ray log pattern alignment and depth matching by machine learning