Abstract:Total organic carbon (TOC) content is a key indicator for determining the hydrocarbon content of shale. The current model for calculating the TOC content of shale is relatively simplistic, the modeling process is cumbersome, and the parameters involved are influenced by subjective factors, which have certain shortcomings. To address this problem, a time-domain convolutional neural network (TCN) model for predicting total organic carbon content based on logging sequence information was established by starting from logging sequence information, conducting logging parameter sensitivity analysis experiments, prioritizing logging-sensitive parameters as model feature vectors, and constructing a TCN network. Meanwhile, to overcome the problem of an insufficient sample size, a five-fold cross-validation method was used to train the TCN model and obtain the weight matrix with the minimum error, and then a shale reservoir TOC content prediction model based on the TCN model was established. The model was applied to evaluate the TOC logging of the Lianggaoshan Formation in the Sichuan Basin, China, and the predicted results were compared with the traditional ΔlogR model. The results indicate that the TCN model predicts the TOC content more accurately than the traditional model, as demonstrated by laboratory tests. This leads to a better application effect. Additionally, the model fully explores the relationship between the logging curve and the total organic carbon content, resulting in improved accuracy of the shale TOC logging evaluation.

What problem does this paper attempt to address?

### Problems the Paper Aims to Solve This paper aims to address the issue of evaluating the total organic carbon (TOC) content in shale reservoirs. The current models used to calculate shale TOC content are relatively simple, the modeling process is cumbersome, and the parameters involved are greatly influenced by subjective factors, which presents certain limitations. To overcome these issues, the authors propose a new method based on Temporal Convolutional Networks (TCN) to predict the TOC content in shale reservoirs using logging sequence information. ### Specific Problems and Background 1. **Limitations of Existing Models**: - The current TOC content calculation models are relatively simple, and the modeling process is complex. - Model parameters are influenced by subjective factors, leading to insufficient accuracy. - Traditional methods such as the ∆logR method face difficulties in defining model coefficients and overly rely on maturity indices. 2. **Characteristics of Shale Reservoirs**: - Shale reservoirs have the characteristic of self-generation and self-storage, making accurate prediction of TOC content crucial for oilfield development and subsequent well placement. - Although laboratory geochemical analysis is considered the most reliable method for evaluating TOC content, its discreteness, low vertical resolution, high cost, and inability to continuously characterize the formation TOC content limit its application. - Logging data has high vertical resolution and can be used to establish quantitative calculation models based on logging curve response characteristics to determine the TOC content of the entire well section. 3. **Improvements in Existing Technologies**: - Some scholars have improved the ∆logR method, such as using variable coefficient models and combining natural gamma methods. - In recent years, machine learning-based models have been widely used in logging data mining, including quantitative evaluation of reservoir parameters, pore structure classification, and fluid property identification. - Deep learning algorithms such as Recurrent Neural Networks (RNN), Gated Recurrent Units (GRU), Long Short-Term Memory networks (LSTM), and Bidirectional Long Short-Term Memory networks (Bi-LSTM) have been widely applied to time series problems in the logging field, but these networks have issues such as gradient explosion, gradient vanishing, or high memory usage. ### Solution To overcome the above problems, the authors propose a new method based on Temporal Convolutional Networks (TCN): 1. **Construction of the TCN Model**: - Starting from logging sequence information, conduct logging parameter sensitivity analysis experiments, and prioritize selecting logging sensitive parameters as model feature vectors. - Construct the TCN network, use the five-fold cross-validation method to train the TCN model, and obtain the weight matrix with the minimum error. - Establish a shale reservoir TOC content prediction model based on the TCN model. 2. **Model Validation**: - Apply the model to the TOC logging evaluation of the Lianggaoshan Formation shale reservoir in the Sichuan Basin, China, and compare it with the traditional ∆logR model. - The results show that the TCN model is more accurate in predicting TOC content than traditional models, and laboratory tests also verify this. Through this method, the authors hope to improve the accuracy of TOC content prediction in shale reservoirs, thereby better guiding oilfield development and well placement.

New Method for Logging Evaluation of Total Organic Carbon Content in Shale Reservoirs Based on Time-Domain Convolutional Neural Network

Theory-driven neural network for total organic carbon content estimation from well logs

Novel method for total organic carbon content prediction based on non-equigap multivariable grey model

An Improved Method to Accurately Estimate TOC of Shale Reservoirs and Coal-Measures

Prediction of TOC Content in Organic-Rich Shale Using Machine Learning Algorithms: Comparative Study of Random Forest, Support Vector Machine, and XGBoost

Prediction of TOC in Lishui–Jiaojiang Sag Using Geochemical Analysis, Well Logs, and Machine Learning

TOC prediction and grading evaluation based on variable coefficient △logR method and its application for unconventional exploration targets in Songliao Basin

An integrated Convolutional Neural Network (CNN) prediction framework for in-situ shale oil content based on conventional logging data

An improved ΔlogR model for evaluating organic matter abundance

Evaluation of the Total Organic Carbon (TOC) Using Different Artificial Intelligence Techniques

Total organic carbon content estimation for mixed shale using Xgboost method and implication for shale oil exploration

Machine learning with model selection to predict TOC from mineralogical constituents: case study in the Sichuan Basin

Detection of Organic Carbon in Shale by Laser Induced Breakdown Spectroscopy and Raman Spectroscopy Combined with Partial Least Squares Methods

Machine Learning Based Integrated Approach to Estimate Total Organic Carbon in Shale Reservoirs – A Case Study from Duvernay Formation, Alberta Canada

Shale gas geological "sweet spot" parameter prediction method and its application based on convolutional neural network

A novel hybrid CNN–SVM method for lithology identification in shale reservoirs based on logging measurements

Total organic carbon content prediction based on support-vector-regression machine with particle swarm optimization

A novel mineral composition inversion method of deep shale gas reservoir in Western Chongqing

Method and practice of deep favorable shale reservoirs prediction based on machine learning

New-Generation Nuclear Magnetic Resonance T1-T2 Logging–Derived Formation Evaluation in Jurassic Shale Oil Reservoirs of China

Application of the Fractal Method to the Characterization of Organic Heterogeneities in Shales and Exploration Evaluation of Shale Oil