A Probabilistic Approach to Fired Heater Tube Remaining Life Assessments

A case history of a Catalytic Reformer fired heater assessment is used to illustrate an alternative approach to quantify some of the uncertainties associated with fired heater tube remaining life estimates. Such uncertainties include temperature and pressure history, tube wall thickness, and existing creep damage. While it is well understood that fired heater tubes are susceptible to creep and corrosion damage, it can be challenging to quantify the margin of error associated with typical analyses. For example, in order to determine estimates of remaining life, it is not unusual for analysts to use overly conservative input parameters (e.g., minimum measured thicknesses, maximum temperatures and pressures). This often results in a “worst case scenario” or “pass/fail” answer, but it does not necessarily provide insight into the effect of uncertainty. In this paper, an alternative method is utilized to predict heater tube lifetimes. The uncertainty associated with the primary input parameters for the API-579 Part 10 Omega creep procedure is characterized statistically by treating them as random variables and random processes. The probability of tubes failing by a specified time is estimated by performing Monte Carlo simulations. This study is of a practical nature, illustrating how the methodology can be used to aid in decision making.