Abstract
The conventional corrosion management process consists of defining the expected process conditions, identifying potential corrosion threats, and estimating their likely rate, then using that information to develop mitigation plans and inspection schedules. The Virtual Corrosion Engineer (VCE) project aims to improve this process by utilizing online monitoring data to automate the running of the best available corrosion models and provide a continuously updated dashboard in real time. This paper provides an overview of the VCE, together with a brief discussion of the underlying models for two exemplar damage mechanisms, High-Temperature Hydrogen Attack (HTHA) and Under Deposit Corrosion (UDC) in steam generators.
Highlights
In the Virtual Corrosion Engineer (VCE) process, work is performed up front to automate the connection between monitoring data and corrosion models so that assessment results are updated continuously, without human intervention
This paper provides a brief discussion of the undering models for for twotwo exemplar damage mechanisms, High-Temperature
The magnetite particles are generated by corrosion elsewhere in the system and transported into the steam generator, where they deposit onto the heat transfer surfaces [12,13,14,15,16]
Summary
A key concept behind this project is to change from a system based largely on the initial selection of the most likely DMs in the expected conditions to a continuously updated assessment of all potential DMs in the actual conditions as measured by process monitoring instrumentation. In the VCE process, work is performed up front to automate the connection between monitoring data and corrosion models so that assessment results are updated continuously, without human intervention Because of this automation, there is no need to preselect only the most likely DMs; computers today are more than capable of simultaneously running. 97.5th percentile method identifies a point that is quite some distance away from any of the actual operating conditions for this equipment and cannot reasonably be considered representative It would not, for example, be possible to recommend operational changes based on the location of this point.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
