A modern and open-source composite failure analysis framework for aircraft design in Abaqus
Topic(s) : Material and Structural Behavior - Simulation & Testing
Co-authors :
Costa RUI (UNITED KINGDOM), Silvestre TAVEIRA PINHO (UNITED KINGDOM)Abstract :
The landscape of composite structures in the aeronautical industry is ever-changing due to several factors, among those are: environmental, performance, safety, and cost factors. This changing landscape makes composite structure design more challenging and demanding than ever before; composite design engineers need more advanced tools for developing larger and more mission critical composite structures. [1,2]
Research into the failure mechanisms of composite materials has shown great activity in the last decades; state-of-the-art failure models are becoming more accurate and capable of representing the response of these complex materials. [3]
There are, however, two major factors holding back the interaction between these state-of-the-art research models and the engineering design teams and applications. Most research failure models live as written publications or as in-house legacy codes – in both cases this makes the failure models hard to benchmark, compare and test against a comprehensive set of suitable test cases. Also, most research models attempt to model the phenomena as accurately as possible without a clear focus on how the computational efficiency would scale for very large problems. [3,4]
In contrast, the bigger composite design engineering teams use well-known commercial software, such as Abaqus, as the basis for their design and analysis studies. Also, these teams may work in scales of 10s of meters with coarse finite element models incapable of reproducing the stress states needed to accurately determine the failure characteristics of the materials at the microstructural level (see Figure 1). [1]
With the aim of addressing these modelling dichotomies we present a composite failure analysis framework for Abaqus. This framework interacts with Abaqus through user subroutines within a modern and open-source codebase. It allows researchers to focus on translating their models while utilising the extensive test suite and models already available in the framework; it also allows design engineers to immediately test a new workflow using different failure models and routines (see Figure 2).
This framework was developed building on the existing experience of the team on building composite failure models for large structures and coarse finite element models. It aims at reducing the researcher development overhead while retaining all the performance benefits of user subroutines. We present the framework using well-known and well-tested failure models against a comprehensive set of suitable finite element models across the scale and complexity spectra. We believe that this framework has the potential to promote composite failure model development, interoperability between research and engineering scales, and benchmark and testing of the state-of-art on composite failure analysis.
Research into the failure mechanisms of composite materials has shown great activity in the last decades; state-of-the-art failure models are becoming more accurate and capable of representing the response of these complex materials. [3]
There are, however, two major factors holding back the interaction between these state-of-the-art research models and the engineering design teams and applications. Most research failure models live as written publications or as in-house legacy codes – in both cases this makes the failure models hard to benchmark, compare and test against a comprehensive set of suitable test cases. Also, most research models attempt to model the phenomena as accurately as possible without a clear focus on how the computational efficiency would scale for very large problems. [3,4]
In contrast, the bigger composite design engineering teams use well-known commercial software, such as Abaqus, as the basis for their design and analysis studies. Also, these teams may work in scales of 10s of meters with coarse finite element models incapable of reproducing the stress states needed to accurately determine the failure characteristics of the materials at the microstructural level (see Figure 1). [1]
With the aim of addressing these modelling dichotomies we present a composite failure analysis framework for Abaqus. This framework interacts with Abaqus through user subroutines within a modern and open-source codebase. It allows researchers to focus on translating their models while utilising the extensive test suite and models already available in the framework; it also allows design engineers to immediately test a new workflow using different failure models and routines (see Figure 2).
This framework was developed building on the existing experience of the team on building composite failure models for large structures and coarse finite element models. It aims at reducing the researcher development overhead while retaining all the performance benefits of user subroutines. We present the framework using well-known and well-tested failure models against a comprehensive set of suitable finite element models across the scale and complexity spectra. We believe that this framework has the potential to promote composite failure model development, interoperability between research and engineering scales, and benchmark and testing of the state-of-art on composite failure analysis.