000004503 001__ 4503
000004503 005__ 20141118192701.0
000004503 0177_ $$2doi$$a10.3850/978-981-07-2219-7_P178

000004503 0247_ $$210.3850/978-981-07-2219-7_P178
$$adoi
000004503 04107 $$aeng
000004503 046__ $$k2012-05-23
000004503 100__ $$aRoy, Pronab
000004503 24500 $$aProbabilistic Characterization of B<sub>1</sub> for Nuclear Pipe Bends Subjected to Internal Pressure with In-Plane Closing Moment

000004503 24630 $$n5.$$pProceedings of the 5th Asian-Pacific Symposium on Structural Reliability and its Applications
000004503 260__ $$bResearch Publishing, No:83 Genting Lane, #08-01, Genting Building, 349568 SINGAPORE
000004503 506__ $$arestricted
000004503 520__ $$2eng$$aFor a nuclear pipe bend or elbow, the primary stress index B<sub>1</sub> is calculated in terms of the pipe's bend factor (h) as per ASME B&PV Code, Section III (NB-, NC-, ND-3600): 
 B<sub>1</sub>=0.4h-0.1 
 where, h = tR/r<sup>2</sup><sub>m</sub>, t = nominal thickness, R = nominal bend radius, and r<sub>m</sub> = mean pipe radius. The plastic collapse behaviour of pipe bends and elbows under internal pressure and moment is characterized by geometric nonlinearity, and it changes depending on the orientation of flexure. The nonlinear behaviour is affected by the pipe geometry, defined by its nominal outer diameter D<sub>o</sub>, t, and R. These effects can be included, only in an approximate sense through the stress indices B<sub>1</sub> and B<sub>2</sub>, while using a simplified analysis method. Eq.(1) cannot represent all pipe bends with varying D<sub>o</sub>, t, and R and different flexure conditions. For the reliability-based design, it is necessary to obtain probabilistic models of all relevent parameters ('random varaibles'). The primary objective of this paper is the probabilistic characterization of B<sub>1</sub> for pipe bends subjected to internal pressure with in-plane closing moment. The scope of this study is limited to Class 2 and 3, stainless steel (Type 304), thin shell pipes with a 90° long radius bend. The probabilistic characterization is obtained using a 'crude' Monte Carlo simulation with repeated random samplings of the basic random varaiables: D<sub>o</sub>, t, h. The stress index B<sub>1</sub> is calculated as
 B<sub>1</sub>=2t/PD<sub>o</sub>(S<sub>y</sub>-B<sub>2</sub>M<sub>c</sub>/Z<sub>p</sub>) 
 where, Z<sub>P</sub> = plastic section modulus and P = design internal pressure for considering Service Level B and target reliability 3 at room temperature. The collapse moment (M<sub>c</sub>) is evalutated from the moment versus end rotation plot for each pipe specimen by nonlinear finite element analysis. From the linear regression analyses the best fit line is obtained for large number of samplings as 
 B<sub>1</sub>=-0.03380 + 0.1054h + e 
 where, e is a random error with mean zero and standard deviation 0.07232. The values of B<sub>1</sub> are very low, which indicates that the effect of internal pressure in the collapse behavior of pipe bends is negligible. This new expression for B<sub>1</sub> can be used for reliability-based code calibration of nuclear pipe bends and elbows.

000004503 540__ $$aText je chráněný podle autorského zákona č. 121/2000 Sb.
000004503 653__ $$aStress index, Pipe bends, Monte carlo simulation, Finite element analysis, Linear regression.

000004503 7112_ $$a5th Asian-Pacific Symposium on Structural Reliability and its Applications$$cSingapore (SG)$$d2012-05-23 / 2012-05-25$$gAPSSRA2012
000004503 720__ $$aRoy, Pronab$$iGhosh, Siddhartha
000004503 8560_ $$ffischerc@itam.cas.cz
000004503 8564_ $$s150893$$uhttps://invenio.itam.cas.cz/record/4503/files/P178.pdf$$yOriginal version of the author's contribution as presented on CD, .
000004503 962__ $$r4180
000004503 980__ $$aPAPER