We have used our expertise in acoustics, software development and machine learning to develop an advanced processing pipeline that can analyse the data from existing oil and gas well logging tools.
In cased holes, such as oil and gas wells, it is often critical to inspect the condition of the casing and cement around the casing. Without sufficient bonding quality, oil and gas could escape causing severe environmental damage.
We combine knowledge of the underlying physics with machine learning to provide an optimal evaluation of the bonding quality using pitch-catch, pulse-echo and CBL data. Additionally, we have developed techniques to calculate the total length of the sealing barrier; the top of cement; distinguish different types of bonding and divide the well into different sections based on the characteristics.
From this, Equanostic can provide an answer to whether a cementing job meets requirements in the form of a detailed, auto-generated report. While the analysis pipeline is fully automated, we have internal tools that allow manual inspection of the data and results. This helps us to ensure the automated pipeline makes the correct evaluation and allows us to easily dig deeper whenever necessary.
The analysis is designed to run using data from qualified well logging technology as input. This type of logging is already performed across large parts of Norway and Europe, allowing our techniques to be applied quickly to existing data sets as well as data from new logging runs. The pipeline has been tested on DLIS data from real logging runs of wells in Europe with good results.