Abstract
Induction welding is a promising technique for assembling carbon fabric-reinforced thermoplastic composite aerospace components, owing to its potential for automation and ability to meet high production rate demands. This joining technique also reduces stress concentrations at the interfaces between the components in comparison to traditional joining techniques.
This work focuses on developing an accurate and computationally efficient simulation method for predicting temperature distribution within carbon fabric-reinforced thermoplastic composite laminates during induction heating. The method targets both process and structure design for induction-welded assemblies, necessitating a macroscopic laminate-level approach with homogenised material properties.
A key challenge identified was obtaining consistent, anisotropic electrical conductivity values for the inductively heated material. After the introduction to the topic of this work, the development of a reliable method for measuring anisotropic electrical conductivity is presented, which forms the foundation for the key findings of this study. Subsequently, a model representation on the composite laminate level was developed. Next, this laminate level homogenised anisotropic electrical conductivity is integrated into induction heating simulations, which are suited for the induction heating analysis of realistic composite structures.
In the final part of this work, it is shown that, for quasi-isotropic laminates, which are commonly used in practice, the electrical conductivity description can be simplified to an isotropic representation based on measured electrical conductivity values. This simplification significantly reduces computational time while maintaining accuracy.
The major result of this work is an approach to predicting heat generation during induction heating of arbitrary carbon fabric-reinforced TPCs. The proposed method involves first measuring the anisotropic electrical conductivity of the composite using the six-probe method. This measured electrical conductivity can then be incorporated into induction heating simulations by representing it as a function of the current direction. For quasi-isotropic laminates, the electrical conductivity description can be simplified to an isotropic representation based on measured electrical conductivity values. This simplification significantly reduces computational time while maintaining accuracy.
To further advance the process modelling of induction welding in practical applications, it is recommended to develop simulations in greater detail to include more physical phenomena and to acquire more accurate thermal material property data. In particular, meso-scale (fibre bundle) modelling would offer better insights into the underlying mechanisms that determine macro-scale conductivity. Actual welding includes melting, possible deconsolidation and reconsolidation, and it has a pronounced effect on the temperature and structure development during welding, so requiring further model developments are necessary. Finally, reliable simulations require accurate data on the electrical conductivity, as well as specific heat capacity, so additional experimental efforts are required.
This work focuses on developing an accurate and computationally efficient simulation method for predicting temperature distribution within carbon fabric-reinforced thermoplastic composite laminates during induction heating. The method targets both process and structure design for induction-welded assemblies, necessitating a macroscopic laminate-level approach with homogenised material properties.
A key challenge identified was obtaining consistent, anisotropic electrical conductivity values for the inductively heated material. After the introduction to the topic of this work, the development of a reliable method for measuring anisotropic electrical conductivity is presented, which forms the foundation for the key findings of this study. Subsequently, a model representation on the composite laminate level was developed. Next, this laminate level homogenised anisotropic electrical conductivity is integrated into induction heating simulations, which are suited for the induction heating analysis of realistic composite structures.
In the final part of this work, it is shown that, for quasi-isotropic laminates, which are commonly used in practice, the electrical conductivity description can be simplified to an isotropic representation based on measured electrical conductivity values. This simplification significantly reduces computational time while maintaining accuracy.
The major result of this work is an approach to predicting heat generation during induction heating of arbitrary carbon fabric-reinforced TPCs. The proposed method involves first measuring the anisotropic electrical conductivity of the composite using the six-probe method. This measured electrical conductivity can then be incorporated into induction heating simulations by representing it as a function of the current direction. For quasi-isotropic laminates, the electrical conductivity description can be simplified to an isotropic representation based on measured electrical conductivity values. This simplification significantly reduces computational time while maintaining accuracy.
To further advance the process modelling of induction welding in practical applications, it is recommended to develop simulations in greater detail to include more physical phenomena and to acquire more accurate thermal material property data. In particular, meso-scale (fibre bundle) modelling would offer better insights into the underlying mechanisms that determine macro-scale conductivity. Actual welding includes melting, possible deconsolidation and reconsolidation, and it has a pronounced effect on the temperature and structure development during welding, so requiring further model developments are necessary. Finally, reliable simulations require accurate data on the electrical conductivity, as well as specific heat capacity, so additional experimental efforts are required.
Original language | English |
---|---|
Qualification | Doctor of Philosophy |
Awarding Institution |
|
Supervisors/Advisors |
|
Award date | 25 Jun 2024 |
Place of Publication | Enschede |
Publisher | |
Print ISBNs | 978-90-365-6148-8 |
Electronic ISBNs | 978-90-365-6149-5 |
DOIs | |
Publication status | Published - 25 Jun 2024 |