Probabilistic methods for assessing performance of concrete structures in nuclear power plants

Many nuclear power plants in North America and Europe are in their third decade of operation, and their in-service condition assessment accordingly is of regulatory interest. The evaluation of a NPP facility with regard to its suitability for continued service must address the condition of mechanical, electrical and structural components. While structural components generally play a passive role in mitigating accidents from internal events, they play a significant role in plant safety when accidents are initiated by extreme environmental events. Failure of structural components and systems may cause other systems that are otherwise redundant to malfunction (common-cause failures). Moreover, in contrast to many mechanical and electrical components, structural repair or replacement often is impractical or unfeasible. Recent probabilistic safety analyses of NPP facilities have shown that structural performance is key in limiting core damage or offsite risk to acceptable levels. Many safety-related structures in NPPs are reinforced (or prestressed) concrete. Aging effects in reinforced concrete structures may cause their structural properties to change over time. Some of these effects and changes are relatively benign; others may cause strength or stiffness to deteriorate in time. Evaluation of an existing structure for continued service must provide quantitative evidence that its strength and stiffness are sufficient to withstand future extreme events within a proposed service period with an acceptable level of reliability. To achieve the desired reliability goals, a reinforced concrete structure may have to be inspected and maintained periodically and, eventually, repaired or upgraded. The loads on the structure from operating and environmental events, as well as the structural capacity to withstand these loads are uncertain in nature; moreover, significant differences between original and current design codes may exist. Sources of uncertainty that must be taken into account in condition assessment of an existing reinforced concrete structure include: (1) lack of in-service measurements of strength and loads; (2) variations in service loads; (3) limitations in available models for quantifying time-dependent changes in concrete and steel; (4) limitations in nondestructive evaluation (NDE) technologies in difficult field conditions; and (5) shortcomings in existing methods for repair. Advances in structural reliability theory over the past decade now have made it possible to analyze most of these uncertainties more rationally within the framework of probability theory. An aging management system should provide a set of criteria, rules and quantitative tools that can be used to evaluate reinforced concrete components and structures during the projected service life. Optimal intervals of inspection and repair can be determined, based on minimum life cycle expected costs or other requirements. When limited resources are available, it may be most effective to restrict the condition assessment efforts on a few safety-critical components. Moreover, guidelines for selecting and executing the experimental (in situ or laboratory) tests, in relation to their efficiency, must be given, as well as criteria for evaluating the results of the inspection in relation to their statistical consistency. These guidelines must consider that no nondestructive evaluation method can detect a given defect with certainty; moreover, that in some cases the values of a quantity only indirectly related to the quantities of engineering interest are inferred.