Deep and offshore drilling operations face complex geological formations, uncertain formation pressures, and narrow safety density windows, making them susceptible to lost circulation risks. To address these challenges, this paper introduces an innovative, intelligent lost circulation monitoring model that incorporates geological lithology information. This model not only utilizes real-time drilling parameters, but also encodes geological information such as rock type as inputs to the model. By combining these key lithological features, the model can comprehensively assess wellbore stability and reduce the lost circulation risks. In this paper, the Conditional Tabular Generative adversarial network (CTGAN) model is used to enhance the data of small-sample risk data, which can effectively expand the data distribution space and improve the performance of the model. This paper conducts a comparative analysis of intelligent monitoring results using artificial neural networks (ANNs), long short-term memory (LSTM), and temporal convolutional networks (TCNs). The results show that the TCN achieves an identification accuracy of 93.7%. Furthermore, the analysis reveals that the inclusion of lithology information significantly enhances the model’s performance, resulting in a 7.1% increase in accuracy. The false alarm rate of the model can be reduced by 10.2%, considering the fluctuation of the logging curve caused by the on/off condition of the pump. This indicates that the introduction of lithology information and the condition of the pump on−off provide advantages in monitoring and identifying lost circulation risks, enabling a more precise assessment of wellbore stability and a reduction in lost circulation incidents. The method of lost circulation monitoring proposed in this paper provides an important safety guarantee for the oil drilling industry.

As unconventional hydrocarbon resources gain increasing importance, the risk of lost circulation during drilling operations has also grown significantly. Accurate and reliable risk diagnosis methods are essential to ensure safety and operational efficiency in complex drilling environments. This study proposes a novel lost circulation risk monitoring framework based on time-series shapelet transformation, integrated with Generative Adversarial Network (GAN)-based data augmentation and real-time model updating strategies. GANs are employed to synthesize diverse, high-quality samples, enriching the training dataset and improving the model’s ability to capture rare or latent lost circulation signals. Shapelets are then extracted from the time series using a supervised shapelet transform that searches for discriminative subsequences maximizing the separation between normal and lost-circulation samples. Each time series is subsequently represented by its minimum distances to the learned shapelets, so that critical local temporal patterns indicative of early lost circulation can be explicitly captured. To further enhance adaptability during field applications, a real-time model updating mechanism is incorporated. The system incrementally refines the classifier using newly incoming data, where high-confidence predictions are selectively added for online updating. This strategy enables the model to adjust to evolving operating conditions, improves robustness, and provides earlier and more reliable risk warnings. We implemented and evaluated Support Vector Machine (SVM), k-Nearest Neighbors (kNNs), Logistic Regression, and Artificial Neural Networks (ANNs) on the transformed datasets. Experimental results demonstrate that the proposed method improves prediction accuracy by 6.5%, measured as the accuracy gain of the SVM classifier after applying the shapelet transformation (from 84.7% to 91.2%) compared with using raw, untransformed time-series features. Among all models, SVM achieves the best performance, with an accuracy of 91.2%, recall of 90.5%, and precision of 92.3%. Moreover, the integration of real-time updating further boosts accuracy and responsiveness, confirming the effectiveness of the proposed monitoring framework in dynamic drilling environments. The proposed method offers a practical and scalable solution for intelligent lost circulation monitoring in drilling operations, providing a solid theoretical foundation and technical reference for data-driven safety systems in dynamic environments.