You are here

An end-to-end simulation solution for autoclave manufacturing

News International-French

14 Apr 2011

In response to the current challenges it is facing, the aeronautical industry is re-examining the autoclave process. Here, we present a simulation-based system that provides decision-making support all the way from design to manufacturing of composite autoclave parts.

(Published on June 2008 – JEC Magazine #41)







Production rates for composite parts in the aeronautical industry are on the rise. Parts characteristics are also changing, invalidating the “good old recipes”; more and more composite parts are being produced to fulfil structural functions, and these composite parts tend to be larger, thicker and more complex than before. While the critical influence of the manufacturing process on mechanical performance is well known, industry appears to take relatively little account of manufacturing effects, relying more on the use of stochastic methods than on data generated by manufacturing simulation.


This paper describes a complete system that offers simulation tools for making decisions throughout the composite-product development process, in particular about material selection, process definition/optimization, and what manufacturing data to make available in order to improve the accuracy of mechanical analysis for a reliable design. The whole solution is depicted schematically in figure 1. The various steps are described below.


From design to analysis

The simulation system is entered via an automated process that was developed for integration between the popular composite design tools FiberSim and the mechanical analysis software Sysply. The process gives instant, error-free transfer of design data to the finite element analysis (FEA) software. This capability will be extended this year to other design tools like Dassault Systèmes’ CPD. The proposed process represents a significant advance, since generating such finite element models manually can require many days of work.



Defining and optimizing the forming process

In composite design systems, a feasibility analysis is usually conducted using geometric methods. This makes it possible to assess draping feasibility and to calculate the flat patterns of the plies. For complex geometries or advanced materials, it might be necessary to conduct a mechanical draping simulation using the PAM-FORM component of the specific autoclave solution. This is the case, for instance, for the diaphragm forming process. These draping simulation techniques serve to identify possible manufacturing problems in the early design stage, this being a major source of potential time and cost savings.


Gas flow, heat transfer and curing

Following the draping simulation, an optional step involving a full analysis of the gas flow inside the autoclave may be performed, taking into account the interactions of the gas flow with the tooling and the parts. A detailed heat transfer model that includes laminate curing is created to determine whether the defined process will lead to high-quality cured parts. Figure 2 shows the example of a mesh used in a typical autoclave simulation, and figure 3 shows an example of the results. Figure 4 shows the heat transfer around the laminate and the effects of the exothermic reaction. These autoclave simulations serve to optimize the autoclave design and sizing, as well as the autoclave loading and the forming cycles. Numerical simulation can reduce the high cost of trial-and-error testing and, especially, help to reduce the high operating costs of prototype testing.





Quality assessment

The current version does not include compaction simulation. The techniques developed for the LRI/RFI processes [1] will be extended to compaction and to the squeeze flow occurring in the autoclave process. The user will also be provided with an indicator of void content based on temperature and pressure changes. Being able to assess the part quality before manufacturing is seen as a major breakthrough that will lead to substantial savings.


Predicting shape distortion

Models from Sicomp [2] were implemented to predict the composite shape distortion based on accurate time history temperature input used as boundary conditions in thermal expansion and chemical shrinkage modelling. This data can be used to modify the tool geometry so the part will be correct after distortion. The French LCM-SMART [3] Project led by Hexcel will investigate the use of the numerical technology available in sheet metal stamping.


Simulation is commonly used in failure analysis. A recent example is the work conducted with PAM-SHOCK for birdstrike certification of the Boeing 787’s movable trailing edge made of composite [4]. The effects of draping on mechanical performance have been recognized for a long time and have been demonstrated, for instance, in [5]. The aim of current developments is to make this technology available to engineers. The next step will be to account for residual stresses, which will be investigated in the PRECARBI Project [6]. A systematic use of manufacturing data in these simulations should lead to reliable mechanical performance analysis.


A unification of the various graphical user interfaces and solver platforms is under way so as to offer a seamless simulation of the complete autoclave-part development process. It is believed that this autoclave simulation tool will revolutionize the composite-manufacturing design process by offering the engineering community an integrated solution capable of simulating composite parts from end to end, without compromising on accuracy.




  1. P. Celle, P. de Luca, S. Drapier, J.M. Bergheau, “Numerical modeling of Infusion processes (LRI and RFI)”, Proceedings of US SAMPE , May 2007.
  2. J. Magnus Svanberg, “Predictions of Manufacturing Induced Shape Distortions”, Doctoral Thesis, 2002:40, ISSN: 1402-544,ISRN: LTU-DT - 02/40 - SE.
  3. LCM-SMART, “Procédés LCM innovants pour pièces complexes en composites”, Projet FCE.
  4. Georgiadis, S., Gunnion, A.J., Thomson, R.S., Cartwright, B.K., “Bird-strike Simulation for Certification of the Boeing 787 Composite Moveable Trailing Edge”, Composite Structures (2008).
  5. L. Greve, A.K. Pickett, “Modelling damage and failure in carbon/epoxy non-crimp fabric composites including effects of fabric pre-shear”, Composites: Part A 37 (2006) 1983- 2001.
  6. PRECARBI EC project, “Materials, Process and CAE tools developments for Pre-impregnated Carbon Binder yarn preform composites”, Contract No.: 30848.