Lost circulation remains a major challenge during well construction, often leading to non-productive time, increased material consumption, and additional treatment costs. In field practice, the selection of lost circulation materials (LCMs) is still largely based on empirical rules or laboratory testing; however, these approaches are not always suitable for rapid decision-making under variable downhole conditions. This study presents a physics-guided surrogate modeling framework for predicting fracture sealing performance and supporting injection strategy selection. The approach combines laboratory observations with coupled computational fluid dynamics and discrete element method (CFD-DEM) simulations to represent both measured behavior and a broader range of mechanically consistent sealing scenarios. The final dataset included 300 cases, comprising 45 physical experiments and 255 CFD-DEM-generated synthetic cases. A hybrid machine learning architecture based on Temporal Convolutional Networks and Artificial Neural Networks was developed to predict sealing pressure under different material and fluid conditions. The model achieved an R² of 0.89 and a mean absolute percentage error of 6.4%, while showing 94% agreement with laboratory-based recommendations for injection strategy. The proposed framework can therefore serve as a rapid engineering support tool for preliminary formulation screening and a more computationally efficient digital workflow for fracture sealing design in drilling operations.

The achievements in the field of genetics and biotechnology in recent decades have radically changed the direction of biological science. Genetic engineering allows, by purposeful intervention in the genetic material of living organisms, to change their hereditary characteristics and form new biological properties. Thanks to these technologies, the genetic structure of organisms has been studied in more depth, and the foundation for new diagnostic, therapeutic and biotechnological approaches at the molecular level has been laid. The scientific knowledge obtained as a result of the application of genetic engineering has made a significant contribution to the development of biology, especially molecular genetics and biotechnology. GMO technologies (Genetically Modified Organisms) are one of the most striking results of this development. They have opened up important opportunities in areas such as increasing agricultural productivity, breeding plants resistant to diseases and climate change, eliminating hereditary diseases in animal husbandry, and producing biologically active substances such as vaccines and insulin in biomedicine. However, the widespread application of GMO technologies has also raised certain biological, ecological and ethical problems. The mixing of genetically modified species with natural populations as a result of gene flow can lead to a decrease in biodiversity, disruption of ecosystem stability and the creation of new genetic combinations. In addition, the impact of long-term consumption of GMO products on human health has not yet been fully scientifically proven.Along with scientific and technological progress, ethical issues also play an important role in this area. Problems such as the limits of genetic intervention, human genome editing, potential impacts on nature and public trust in biotechnological development require a balanced and responsible approach. The main goal of the scientific community is to ensure that genetic technologies serve human well-being and the sustainable development of the ecosystem.As a result, genetic engineering and GMO technologies will remain one of the main directions of future biological science. International control mechanisms, scientific regulations and public awareness programs should be strengthened for the sustainable, ethical and environmentally safe application of these technologies. Thus, genetic engineering can act as one of the main guarantors not only of scientific progress, but also of the future well-being of humanity.

Purpose. The purpose of this paper is to develop and validate a Python-based workflow for processing continuous hook-load data and applying it to lithology-driven friction-factor calibration in torque-and-drag analysis during casing-running operations in deep, high-pressure wells. Methods. The proposed workflow is based on the automated processing of continuous mud-logging hook-load data recorded during casing deployment. The procedure includes data cleaning, threshold-based segmentation, rolling filtering, outlier removal, and extraction of representative tripping-in and tripping-out load trends. The processed trends are then compared with field measurements and used to calibrate interval-specific friction factors in torque and drag simulations for two offshore wells. Findings. The results show that the developed workflow converts noisy continuous hook-load records into stable load trends suitable for calibration. In both analyzed wells, a single constant friction factor was insufficient to reproduce the full mechanical response of the casing run. Better agreement was achieved when friction factors were calibrated for individual depth intervals. The calibrated profiles also showed that changes in friction behavior were generally consistent with lithological transitions and heterogeneous wellbore conditions. Originality. The study’s originality lies in combining continuous mud-logging hook-load data, automated Python-based processing, and interval-specific torque- and drag-calibration into a single workflow for lithology-sensitive interpretation of casing-running behavior. Practical implications. The proposed approach reduces reliance on manual point selection, improves the consistency of post-run torque-and-drag calibration, and can serve as a practical engineering tool for analyzing casing-running performance in geologically complex wells.

The South Caspian Basin presents a globally challenging offshore environment due to extreme high-pressure, high-temperature (HPHT) conditions and narrow drilling windows. While Artificial Intelligence (AI) is vital for safety monitoring, conventional “Red AI” architectures suffer from high latency and energy intensity, creating a dangerous “inference lag” during pressure excursions. This paper proposes a “Green AI” framework tailored for “Digital Transformation 2.0,” prioritizing algorithmic efficiency for real-time edge deployment on offshore platforms. The methodology integrates a robust signal conditioning pipeline, including Savitzky-Golay filtering and IQR-based outlier detection, to refine stochastic drilling telemetry. The core optimization utilizes a regression-oriented Knowledge Distillation pipeline, where a lightweight “Student” model (1D-CNN + GRU) emulates a high-capacity “Teacher” model by minimizing a joint loss function, followed by magnitude-based pruning and 8-bit post-training quantization. Validated using Azerbaijani offshore drilling data, the framework achieves a 7.7-fold acceleration in inference speed and a 70% reduction in power consumption, while maintaining a highly competitive F1-score of 0.961 (reflecting a marginal 1.6% degradation from the baseline). From an industrial safety perspective, this optimized engine delivers a critical 55-s early warning advantage for gas influx detection, directly mitigating risks of blowouts and ecological damage. Furthermore, by lowering the hardware energy footprint to 3.8 Watts, the system ensures autonomous monitoring on low-power edge devices, enhancing infrastructure resilience and reducing costly Non-Productive Time (NPT). This research demonstrates that sustainable, efficiency-driven AI is essential for the economic and environmental viability of modern offshore operations in the South Caspian energy sector.