A Particle Swarm Optimization-Based Multi-level Processing Parameters Optimization Method for Controlling Microstructures of an Aged Superalloy During Isothermal Forging

Abstract

To obtain the designed target microstructures of an aged superalloy during isothermal forging, a multi-level processing parameters optimization method is developed based on particle swarm optimization (PSO) algorithm. In the developed method, the accurate material models are used to characterize the microstructural evolution. Based on the designed target microstructures, the global and local optimality criterions are constructed to alternately optimize global and local multi-level processing parameters by the PSO algorithm with a linear decreasing inertia weight strategy. The optimized initial volume fraction of δ phase (δVF), deformation temperature and strain rate are 12.95%, 1000 °C and 0.001 s−1, respectively. According to these optimized parameters, the recrystallization volume fraction, average grain size and δVF are 100%, 11.2 µm and 2.1%, respectively, which well agree with the designed targets. Additionally, the processing parameters optimized by the developed method and traditional processing maps are compared. It is found that the developed method is more effective to control microstructures for the studied superalloy.