Skip to content

Complexity Modeling

Tubular Corrosion Model

Client
Eni
Country
Italy
Year
2021
Industry
Oil & Gas
Service
Reliability and Safety

For Eni, kwantis developed a tubular corrosion model for Carbon Capture and Storage (CCS) operations. It turns the material degradation caused by corrosion into a failure probability that feeds a Fault Tree Analysis (FTA).

The challenge

In CCS operations, corrosion from produced or injected fluids is the main parameter affecting tubular reliability. To include it in a risk assessment, the material degradation it causes has to be evaluated, interpreted and expressed as a failure probability.

Our approach

The model first identifies each material/fluid combination, so that corrosion is estimated only where the tubing is susceptible to it. Fluids that are compatible with the tubular material are excluded, to keep the estimate precise.

Corrosion is quantified as a corrosion rate in mm/y by tracking the thickness reduction, to safeguard tubular integrity and barrier efficacy. The process includes:

  • Corrosion rate: calculated for sweet and sour corrosion. An industry-standard corrosion model covers mainly uniform sweet corrosion, and sour environment corrosion rates come from a data-driven rule of thumb.
  • Failure rate: the corrosion rate and the thickness reduction allowance are converted into a failure rate in failures/hour.
  • Probability of Failure on Demand: computed from the Mean Time to Repair, the monitoring frequency and the tubular material failure rate.

The outcome

Corrosion now enters the risk assessment of CCS operations as a failure probability within the FTA, and only where tubing materials are susceptible to the fluids they handle.

Related case studies

How can we help you?

Tell us about your project, asset or data challenge. Our consultants and engineers will get back to you.

Get in touch

Quick links

↑ ↓ to navigate ↵ to open All results

Send us a message

We will get back to you shortly.

Protected by reCAPTCHA: the Google privacy policy and terms apply.