Cementing oil wells is an essential stage in guaranteeing the well’s safety and productivity. However, failures in cementing can cause fluid leaks, environmental contamination and loss of production.
The invention is a computational method for detecting and evaluating cementing failures in oil well casings. To do this, the method uses machine learning and highly reliable numerical simulations to interpret acoustic profiling signals.
The method first uses numerical simulations to create a representative data set of cementing faults. It then uses machine learning to train a predictive model that can identify and estimate the severity of the faults.
Patent title:
Computational method for detecting and estimating cementing faults in oil well casings by acquiring acoustic profiling signals through the production column based on machine learning and high-fidelity simulations
Deposit Number:
BR 10 2021 018581 3
Pontifical Catholic University of Rio de Janeiro – PUC-Rio
Rua Marquês de São Vicente, 225, Gávea
Rio de Janeiro, RJ – Brasil
Zip code: 22451-900
Postal box: 38097
Phone:
+55 (21) 3527-2155