Stochastic finite-element modelling and optimization for net-shape forging of three-dimensional aero-engine blades

Abstract

The stochastic nature of the forging process is a key influencing factor for achievable dimensional and shape accuracy of forged aerofoil blades for aero-engine applications. In this article, a finite-element-based stochastic modelling and optimization method is presented for net-shape forging of three-dimensional (3D) aerofoil blades. A combined Monte Carlo simulation and response surface method as well as the first-order second-moment reliability method are developed to quantify the stochastic characteristics of forging errors due to uncertainties of process parameters and forging conditions. A two-step optimization approach is presented to minimize systematic errors using a direct compensation method and to reduce random variations through a variance control procedure. A 3D blade forging case study is presented to evaluate the stochastic characteristics of the dimensional errors of forged aerofoil blades in comparison with measurement data and to demonstrate much improved forging accuracy using the proposed stochastic optimization method. The results from the case study also suggest that the stochastic-based optimization method may be easily applied to other metal-forming processes.