{"references": ["[1] Philip P. Walsh, Paul Fletcher . Gas Turbine Performance , Second edition [M]. UK: Blackwell Science Ltd, 2004", "[2] Chew, J. W. Developments in turbo machinery internal air systems\", Proceedings of the Institution of Mechanical Engineers, Part C [J]. Journal of Mechanical Engineering Science, 2009, Vol. 223, No. 1, p. 189 - 234.", "[3] Ethan Stearns, Dave Cloud, et al. Probabilistic thermal analysis of gas turbine internal hardware[R], ASME Paper, 2006-GT-90881, 2006", "[4] Vince Sidwell, David Darmofal, et al. probabilistic analysis of a turbine cooling air supply system: the effect on airfoil oxidation life[R], ASME Paper, 2003-GT-38119, 2003", "[5] David Cloud, Ethan Stearns, et al. Probabilistic analysis of a turbofan secondary flow system[R]. ASME Paper 2004-GT-53197, 2004", "[6] Adam Cooke, Peter Childs, Christopher Long, et al.Investigation into the effect of uncertainty in thermal properties on turbo machinery disc heat transfer using both a monte carlo simulation Technique and a taylor series uncertainty propagation method [R], ASME Paper, 2007-GT-27573, 2007", "[7] St efan Brack, Yannick Muller, et al. Probabilistic analysis of the secondary air system of a low-pressure turbine [R], 2014-GT-26184, 2014", "[8] Thomas Bischoff, Matthias Voigt, and Ed Chehab, et al. Probabilistic analysis of stationary gas turbine secondary air Systems [R], 2006-GT-90261, 2006"]}

The present investigation aims at performing a probabilistic and sensitive analysis of a secondary air system for a two-spool engine. Through this analysis, the robust of secondary air system could be gained and the most sensitive geometric parameters are found. Furthermore, the more restrict mechanical tolerance could be demanded by the designer of secondary air system. An in-house code called ASUS was developed for probabilistic and sensitive analysis, which is based on Monte Carlo Simulation method (MCS). The analysis model was based on the deterministic secondary air system network for a two-spool engine with take-off condition with all geometric parameters, including seals, gaps, holes and tubes. These parameters were normally distributed based on the real geometric tolerances to form the probabilistic inputs. The boundary condition was treated in two ways; one is constant while the other is normally distributed with an assumed change range ±5%. Probabilistic and sensitive analyses were carried out on these two cases. Through the analyses, in both cases the rim seal after 1st rotor is sensitive to the mechanical tolerance and boundary change. Therefore, the special attention should be paid to this flow path.