In 1991, the American Society of Mechanical Engineers (ASME), published a research report (CRTD-Vol. 20-1, 1991) on the use of risk-based practices for the development of inspection guidelines. This document presented an overall risk-based inspection approach in four sequential stages: (1) System boundary definition; (2) Qualitative risk assessment; (3) Failure modes, effects and criticality analysis; and (4) Inspection program development.
In 1994, ASME extended this approach to specifically include fossil fuel fired electric power generation station applications in subsequent research report (CRTD Vol. 20-3,1994). It addresses the in-service inspection (ISI) of components of fossil fuelled power plants. TWI undertook a review of this document, along with several other recently published industry-specific RBI guidance documents, as part of a major group sponsored project (GSP) in 2001. (The deliverables of this GSP, including the publications' review, an industry survey on current risk-based management practices, case studies and a standalone best practice guide, are available from TWI).
Risk-based inspection and maintenance of fossil fuelled power plants considers all components that contribute to plant unavailability, although the primary focus is on components within the pressure system. The ASME document presents:
- A method for qualitative risk assessment;
- An overview of quantitative risk analysis;
- An overview of inspection program development; and
- Multi-component economic maintenance optimisation.
The description of the methodology for qualitative risk assessment in CRTD Vol. 20-3 was brief, and is generally in line with that presented in CRTD Vol. 20-1. For quantitative risk analysis, three alternative probability of failure (PoF) approaches were presented based on: (1) generic failure data analysis; (2) fault-tree (FTA) or event-tree analysis (ETA); and (3) structural risk and reliability analysis (SRRA) modelling. To determine, quantitatively, consequence of failure (CoF), the criterion used was plant unavailability resulting in an expression of risk as frequency of average MWh lost per forced outage due to failure of the component. For the subsequent development of inspection programs, the use of SRRA was recommended.
CRTD Vol. 20-3 also presented a fully-quantitative methodology for the economic optimisation of maintenance activities in multi-component systems. This approach requires multi-disciplinary skills in decision theory, probabilistic damage modelling and engineering financial analysis. The analysis is complex and in practice will require the design of complex spreadsheet, or the use of reliability analysis software.
Please contact us for more information.