Term 252

Smart Autonomous Downhole Vision & Logging System (SADV-Log)

Project Type: Self-Initiated

Project Description

For petroleum, geotechnical, and carbon capturing and storage operations to make effective decisions, timely and accurate subsurface data is crucial. Even though they are very complex, conventional well logging systems are frequently costly, necessitate several deployment runs, and offer little direct visual confirmation of wellbore conditions. In addition to increasing rig time and overall costs, this leads to operational inefficiencies. The Smart Autonomous Downhole Vision & Logging System (SADV-Log), a laboratory-scale proof- of-concept system that combines real-time downhole visual inspection with fundamental geophysical sensing, is designed and developed in this project. Within a sealed mechanical housing, the system integrates a low-cost camera with gamma ray, resistivity, ultrasonic distance, and caliper sensors. To provide quick lithology and well integrity insights, data is sent via a wired communication interface to a surface-based Python application that performs automated preprocessing and machine-learning classification. The project's main goal is to show that combining sensor-based and visual data into a single, affordable diagnostic tool appropriate for controlled cased-hole and open-hole environments is feasible. The suggested system seeks to improve interpretability, shorten operating times, and offer a modular platform for training and scholarly research. The findings demonstrate the potential of AI-based interpretation and inexpensive sensor integration as a basis for upcoming intelligent downhole logging technologies.