Investigation on the multiscale multiphysics based approach to modelling and simulation in abrasive flow machining of aerofoil structures and its application perspectives

Abstract

Abrasive Flow Machining technology is attracting more and more attentions and expanding into more areas by the industry and research community, particularly in the context of increasing demands for post-processing of complex aerofoil structures and additively manufactured components. This thesis presents an analytical scientific approach for investigating the material removal and surface generation in Abrasive Flow Machining in relation to the affecting factors from the material properties of fluid media, abrasive grains, operation conditions and workpiece material with the aid of multiscale multiphysics modelling and simulation. The multiscale multiphysics approach combined with micro-cutting mechanics is further presented to modelling and analysis of the surface roughness and topography profile generation in the AFM process. The analysis is developed and implemented by using the COMSOL multiphysics computational environment integrated with MATLAB programming as needed. The work described is fundamental but essential as a part of the efforts for developing the simulation-based virtual AFM system. The improved Preston equation is developed in this PhD research, which aims to enhance the scientific understanding of the AFM process and its industrial application. The improved Preston equation can be used to aid the engineers to optimise the process for desired surface roughness and edge tolerance characteristics on complex geometries in an intuitive and scientific manner. The methodology of deriving the equation underpins the fundamental cutting mechanics of abrasive flow machining assuming all abrasive particles within the media are spherical as manufacturers defined. Further to the derivation, four factorial experimental trials and computational fluid dynamics (CFD) simulations are implemented to generate the flow features of media on a coupon to evaluate and validate the equation for its competency and accuracy on prediction of material removals. The modified Preston equation can significantly contribute to the optimisation of the AFM process, which will advantage the integrated machine-process design to predict the virtual surface roughness and material removal rates. In this doctoral research, micro cutting mechanics analysis and Monte Carlo (MC) algorithms are integrated to support the virtual AFM system on the generation of surface texture and topography in the AFM process, through abrasive micro machining with thousands of grains with the complex machining trajectories and advanced analysis in a multiscale multiphysics manner. The method and basic design to determine the viscosity functions are studied and explored using a capillary rheometer and the geometric properties of the solid abrasive particles in the fluidic media. Those particles properties and media viscosity are the bare minimum required for developing multiscale multiphysics simulations on the AFM process. The method is used to evaluate the relative importance of the elastic effects of the polymer filled with particles. AFM trials and industrial case studies on aerofoil structures are conducted to validate the modelling and simulations developed against industrial requirements. There are good agreements between the results of simulations and the trials, further supported by industrial manufacturing data. The effects of process variables and tooling characteristics on material removal and surface roughness generation are investigated by analysing the results of AFM trials and simulations. AFM trials on IBR (Integrally Bladed Rotor) segments are carried out to validate the modelling and simulation mainly focusing on the profile accuracy control of leading/trailing edges of the IBR blades.