Hier finden Sie wissenschaftliche Publikationen aus den Fraunhofer-Instituten.

Failure of a High-Temperature Water Pipe in a Fossil Power Plant Pipework: A probabilistic Analysis

: Cioclov, D.; Wiedemann, H.

Mohan, R. ; American Society of Mechanical Engineers -ASME-, Pressure Vessels and Piping Division:
Assessment Methodologies for Preventing Failure. Deterministic and Probabilistic Aspects and Weld Residual Stress. Vol.1
New York, NY: ASME, 2000 (PVP 410)
ISBN: 0-7918-1891-8
Pressure Vessels and Piping Conference <2000, Seattle/Wash.>
Conference Paper
Fraunhofer IZFP ()
corrosion; fatigue; pipe; pipe inspection; probabilistic interpretation

A horizontal pipe in the water level control system of a power plant boiler failed in service at a temperature of T=300°C and a pressure of 172 bar (registered). The fractographic analysis revealed a four stage time-dependent failure scenarion: I) corrosion pit formation on the inside surface of the pipe; II) stable corrosion fatigue crack growth originating at the corrosion pits; III) cracks coalescence into a single crack; and IV), sudden pipe burst from the single coalesced crack. Probabilistic FAD analysis of stages III and IV evinced a high probability of failure under operating conditions. Probabilistic FAD algorithm (FADSIM software) accounts for the uncertainly encountered in the estimation of crack size, material fracture toughness, and strenth characteristics. Normal distributions has been assumed throughout. The analysis gives due account to thermo-mechanical aging and adjacent interaction effects at the onset of multiple cracks coalescence (Stage III). The probabilistic FAD analysis demonstrated that the pipe failure was a natural final event a long process of material damage where corrosion, mechanical and thermal fatigue all combined with global material aging were synergistically involved in the failure.
In a broader sense, the outlined failure analysis offers an example of the increasing insight that can be achieved when fracture mechanics principles combined with probabilistic methods are used for risk evaluation in pressure vessels and piping systems.