When Fully Deterministic is Not Possible in Structural Integrity

The structural integrity related design codes and standards in use today are often the result of much iteration over time to reflect the improved knowledge and understanding from research & development or more typically lessons learned from operational experience and application of the codes and standards themselves. The use of design by analysis, rather than design by rule, first appeared in Section III of the ASME Boiler and Pressure Vessel Code in 1963. Invariably the early versions of these codes and standards relied heavily on engineering judgement to generate designs that were deemed fit-for-purpose with the knowledge and data available.

For example, the introduction of fatigue design for pressure vessels as a requirement to Section III of the ASME Code was based upon fatigue testing in air of small scale polished test specimens. The fatigue design curves generated for application to large scale components included transference factors to account for size effects, data scatter and environment. The analysis methods used to predict stresses were conservative via simplifications, and before the widespread use of finite element methods, which when combined with the fatigue design curve was considered to provide a deterministic prediction of fatigue usage factor for a component design life.

However, over time, the understanding of stainless steel fatigue and fatigue crack growth in a hot wet PWR environment and its dependence on strain rate or stress intensity factor rate has improved significantly such that the previously specified acceptance criteria is being challenged. When combined with increasingly complex finite element modelling techniques to eat into the previous simplified analysis conservatisms then the concept of what was considered ‘deterministic’ is now questioned. Instead, a more meaningful criteria of Target Reliability is discussed along with the different difficulties and problems that this brings.

In order to address issues like these for companies facing challenges with acceptance criteria for critical engineering applications, FESI will be holding a one-day workshop dedicated to when fully deterministic is not possible in structural integrity. Delegates will be invited to participate in a discussion to identify challenges and express their view on the methods to ensure high integrity in safety-critical structures expected to bear loads. At the end of the workshop FESI would summarise the gaps in knowledge and possible ways that these could be addressed.

Photograph of the Nightingale Hall © Rolls Royce Leisure Derby.