Abstract

Computer-aided engineering (CAE) simulation is increasingly used in metals industry to speed up manufacturing and production, and material data have been an integral part ever since such simulations were made possible. Recent advancement in computing power has been driving the simulation capabilities forward, from only being able to deal with simple heat transfer in the early days to the current coupled analysis of a multitude of physical phenomena. Such phenomena include heat transfer (thermal field), deformation mechanics (stress/strain field) and phase transformations (microstructural field). An accurate coupling of these phenomena is essential to achieve reliable simulation, which demands the availability of a wide range of material data, from physical and thermophysical properties to rheological properties as well as phase transformation kinetics. The traditional way of obtaining such data is through experimentation, which is not only expensive and time-consuming, but even impossible in cases where accurate simulation requires the property of each phase involved rather than that of the alloy. To provide reliable and cost-effective material data for process simulation, computer-based models must be developed so that such properties can be readily calculated, which is the so-called materials modelling. Processing simulation essentially consists of two types of modelling. One is the materials modelling ? the modelling of composition-processing-microstructure-property relationships; the other is the CAE simulation typically based on finite-element or finite-difference analysis and alike. For historical reasons, the development of these two types of modelling techniques falls into two separate research areas, resulting in two types of computer software, each performing fairly well in its own field but no links exist between them. While the advanced computing power has made the integration of materials modelling into processing simulation possible, the demand for higher accuracy in simulation results has made it necessary. Modelling materials properties and behaviour has been the focus of our research in the past two decades. It incorporates a spectrum of material models covering thermodynamics (the CALculation of PHAse Diagrams, or CALPHAD approach), phase transformation kinetics, and microstructure-property relationships. Most of the material data required for processing simulation can now be readily calculated and then easily transferred to many commercial CAE tools for the simulation of casting, welding, forming, heat treatment and additive manufacturing of various industrial alloys such as Fe-, Ni-, Ti-, Al- and Mg-based alloys. This paper first reviews the development of material models over recent years, followed by recent case studies on processing simulation of some industrial alloys. The simulation practice at present leaves materials design outside the optimisation loop of product design and manufacturing, which reduces the potential design space and may result in suboptimal end products. The necessity of material data in simulation means an alloy has to be physically prepared and have its properties measured before any simulation of its processing becomes possible. The case studies presented here therefore have demonstrated great potential, as most of the material data essential for processing simulation can now be reliably calculated for the first time. It has become a real possibility to merge materials design and processing optimisation into one complete design space, and we are moving one-step closer towards true virtual design.