### 5th Workshop on Structural Analysis of Lightweight Structures

### Program

#### Morning Session

Time: 9:45 - 12:30

**Design criteria for wind-loaded structures beyond Eurocode**

*Daniel Straub (TU Munich)*

Eurocode provides the basis for the design of structures. It ensures a generally accepted level of safety throughout its domain of applicability. However, there are many instances of structures whose design is not fully covered by Eurocode; one example are lightweight structures subject to wind for which the simple load formulas given in Eurocode are not appropriate. When designing such structures, it is nevertheless desired that the overall safety of the structure is in accordance with Eurocode. In this talk, we address appropriate means to ensure this.

The lecture consists of two parts. First, we present a methodology for determining the relevant design parameters based on a limited set of response time-series from numerical simulations (or experiments). We show how the safety margin becomes a function of the amount of data available. This approach is demonstrated through application to a membrane structure analysed with fluid-structure interaction analysis.

The second part is looking at the issue that simple formulas given in Eurocode for establishing wind loads are based on conservative assumptions and parameter choices, i.e. the models contain “hidden safety”. As a consequence, even if advanced numerical models are more accurate than the Eurocode formulations, they can lead to reduced safety because they eliminate the conservative assumptions. To compensate, it can be necessary to employ additional safety factors. We make recommendations on these safety factors based on an analysis of a building portfolio.

**3D-LightTrans – Large scale manufacturing technology for high-performance lightweight 3D multifunctional composites**

*Marianne Hörlesberger (Austrian Institute of Technology, Wien)*

Textile reinforced polymer composites (TRPC) hold the promise for enhanced products featuring superior properties, such as light weight and high strength, with comparatively low material costs. This promising potential is nevertheless hindered by the lack of appropriate processing technologies to enable low-cost manufacturing of mass products with sufficient quality. The goal of the 3D-LightTrans project was to create a highly flexible manufacturing chain for low cost production of integral large scale 3D textile reinforced polymer composite parts and its integration in the complete supply chain. 3D-LightTrans manufacturing chains is based on hybrid yarn incorporating thermoplastic matrix material, processed to deep draped multilayer textiles and multifunctional 3D-textile constructions, which is fixed to dry pre-forms and finally, processed into composites by thermoforming. A model is presented for full automation (in nowadays mostly manually performed) complex handling operations.

**Stochastic Fourier integral operators for parameter estimation and**

damage detection

damage detection

*Martin Schwarz (University of Innsbruck)*

This talk addresses a novel for wave propagation in material with spatially varying random properties. The target model is the linear elastic wave equation in solids. We construct a deterministic Fourier integral operator to solve the problem. For modelling the uncertainty we

use a new approach:

The randomness is included into the solution operator rather than in the coefficients of the model. The solution can be represented by stochastic Fourier integral operators (sFIO), which

have nice computational properties for (numerical) simulations. Our model predicts the fully time dependent dynamic response of the structure, and its stochastic properties. Calibrating the sFIO to measurements can be used for both parameter estimation and damage detection.

**Multiscale model for braid-reinforced polymers. I: Elastic properties and application to coil springs**

*Marc Luger (University of Innsbruck)*

Purposive custom-tailoring of laminated textile-reinforced polymers demands reliable simulative frameworks to predict the individual reinforcement layers’ structural response, yielding potential of optimization of structural components’ mechanical performance. Multiscale modeling is standardly employed to obtain the effective (homogenized) properties of composites, whereas both analytical and numerical upscaling methods are commonly applied. Whilst analytical methods of homogenization usually cause low computational expenses, the geometrical representation of the material is rather simplified and little information is gained concerning the local stress and strain situation. As regards numerical methods, FE-based unit cells are standardly employed to predict the behavior of composite materials.

Within the presented multiscale method, a modeling approach for braid-reinforced polymers employing 3D continuum models to overcome the aforementioned shortcomings is proposed. Hereby, attempting to reduce the number of parameters, idealized geometric properties are employed for the observation scales’ numerical representations. Starting from the definition of observation scales – such as the yarn, the braid, and the component scale – upscaling procedures are employed to obtain effective elastic properties at each scale, whereas the properties of the constituents, i.e. fiber and matrix, serve as input for multiscale modeling:

● At yarn scale, suitable geometric fiber arrays are employed within these FE-models to obtain effective elastic yarn properties. Besides the elastic properties of the aforementioned constituent materials, the fiber-volume-ratio is required.

● At braid scale, FE-based unit-cells are employed to obtain the effective elastic braid properties, applying the effective yarn properties, and matrix properties describing the resin pockets. The influence of geometric braid properties such as braid angles, braid patterns, and yarn undulations is reflected within the geometric arrangement of the FE model.

● Finally, at component scale, the effective braid properties are employed within structural simulations to assess the component’s mechanical performance, rendering prospects for purposive optimization.

In order to assess the predictive capabilities of the proposed multiscale model, the experimentally determined structural response of coil springs, consisting of glass-fiber-reinforced epoxy, is compared to numerical results obtained from the proposed multiscale approach. Effective braid properties are hereby applied to a suitable FE representation of the spring at component scale, accounting for the local geometric braid characteristics, such as e.g. curvature-induced variation of the braid angle.

Acknowledgement

This research was funded by the K-Regio project “Innovative Tube Design – Entwicklung und Optimierung von neuen High-Tech Faserverbundstrukturen für den industriellen Einsatz auf Basis modell- und simulationsbasierter Methoden” in cooperation with Thöni Industriebetriebe GmbH, superTEX composites GmbH, Intales GmbH. Financial support by the Tyrolean Government, the European Regional Development Fund (ERDF) and the University of Innsbruck is gratefully acknowledged.

**Multiscale model for braid-reinforced polymers. II: Extension to viscoelastic properties**

*Ulrich Hofer (University of Innsbruck)*

Multiscale modeling is standardly employed to obtain the effective (homogenized) physical properties of composites, whereas both analytical and numerical upscaling methods are commonly applied. As regards effective viscoelastic properties, analytical methods are often obtained by extending their elastic counterparts to viscoelasticity by exploiting the so-called correspondence principle. As regards numerical methods, FE-based unit cells are standardly employed to determine the viscoelastic behavior of composites.

Within the presented multiscale method, fractional viscoelastic models are used to describe the time-dependent behavior of the matrix, well-capturing the experimentally observed viscoelastic behavior with a comparatively low number of parameters. At three observation scales – defined similar as in [Luger et al. 2018] – upscaling methods are introduced to obtain effective viscoelastic properties at each scale, whereas the properties of the constituents, i.e. fiber and matrix, serve as input for multiscale modeling:

● At yarn scale, the Chamis-equations – standardly used for predicting the effective elastic properties of unidirectionally-reinforced composites – are extended towards viscoelasticity by employing the aforementioned correspondence principle. Hereby, the viscoelastic homogenization problem is transformed into the Laplace-Carson domain, where the corresponding elastic problem is solved. An inverse Laplace-Carson transformation gives closed-form expressions for the effective viscoelastic properties of the yarns [Hofer et al. 2018].

● At braid scale, the upscaling strategy described in [Luger et al. 2018] is extended towards viscoelastic behavior. By doing so, FE-based unit cells are employed to obtain the effective viscoelastic braid properties, applying the effective yarn properties as obtained by the extended Chamis-equations and considering the influence of geometric braid properties, such as the braid angle.

● At the component scale, the effective braid-properties are employed within structural simulations to obtain the viscoelastic response of braid-reinforced components.

In order to assess the quality of the multiscale model, compressional creep tests on braid-reinforced tubes are performed. The model results agree well with the experimentally obtained results. Moreover, the experimentally observed influence of the braid geometry onto the viscoelastic behavior is well reproduced by the proposed multiscale method.

Acknowledgement

This research was funded by the K-Regio project “Innovative Tube Design – Entwicklung und Optimierung von neuen High-Tech Faserverbundstrukturen für den industriellen Einsatz auf Basis modell- und simulationsbasierter Methoden” in cooperation with Thöni Industriebetriebe GmbH, superTEX composites GmbH, Intales GmbH. Financial support by the Tyrolean Government, the European Regional Development Fund (ERDF) and the University of Innsbruck is gratefully acknowledged.

References

[Luger et al. 2018] Luger, M., Traxl, R., Hofer, U., Hirzinger, B., Lackner, R. (2018). RUC-based multiscale model for braid-reinforced polymers: Application to coil springs. Composites Part B: Engineering. Under revision.

[Hofer et al. 2018] Hofer, U., Luger, M., Traxl, R., Lackner, R. (2018). Closed-form expressions for effective viscoelastic properties of fiber-reinforced composites considering fractional matrix behavior. Mechanics of Materials. In Print.

#### Afternoon Session

Time: 13:35 - 17:00

**The potential of Insitu Laser-AFP for thermoplastic composites for high performance structures**

*Sebastian Nowotny (Deutsches Zentrum für Luft- und Raumfahrt e.V.)*

This talk gives an overview of the current state for Laser Automated Tape Placement for thermoplastic composites at the DLR.

The process allows in-situ manufacturing of large-scale, complex structures. Using this tape placement facility, expensive and time-consuming manual processes such as preforming and vacuum bagging processes are not necessary. Furthermore, the process data acquisition and monitoring system allows the component to be analysed in-situ, reducing the costs of non-destructive testing. This facility has been used to produce structures from a number of thermoplastic materials, including PEEK, PEKK, and PPS, for both real-world applications and fundamental engineering material research. Ongoing projects at this facility include the in-situ integration of stiffening elements to aircraft skins, and research on additional welding and joining processes.

**An approach towards an optimal, robust design of lightweight structures**

*Martijn Schmeetz (INTALES GmbH)*

Within the space industry there is a need for low mass structures in order to reduce costs of materials but also costs of fuel required to get these structures into space. In addition, novel manufacturing methods, like additive manufacturing, allow for the creation of more complex geometries which in turn require a more complex design process. Optimising such structures in terms of mass, while at the same time fulfilling all requirements regarding functionality and allowable values such as stresses, strains and displacements, poses a huge challenge. Gradient-based optimization strategies, but also evolutionary algorithms are characterized by the curse of dimensionality which leads to a large number of required model evaluations or to the situation where no global optimum can be identified.

For this reason, a novel approach is proposed where the optimization problem is tackled by a heuristic adaptation procedure on element level. During the FE-analysis, the structural requirements like allowable stresses or strains are checked and – if necessary – the thickness and/or material orientation of the element is updated. Access to the element routines in the FE-analysis is therefore a requirement for this strategy for which reason a number of state-of-the-art element routines have been developed and adopted as user element routines to be linked with the commercial FEM software ABAQUS. The optimization process is performed in an iterative manner, meaning that the external loading is applied in several load portions until the full force magnitudes are active. The main advantage of this approach is the applicability to large and complex FE-models, with the possibility of the structure having multiple local optimums. In addition, the strategy can be used for models involving different kinds of elements, like shell elements including layered composite, sandwich or isotropic (metallic) shell sections, and also fastener elements.

This approach by itself could lead to local concentrations of mass caused by high stresses concentrations. These stresses might be artificial, depending on the constitutive law used, and can occur near sharp corners, boundary conditions or load introduction areas. To reduce these peak masses a smoothing algorithm is developed which smooths the stress concentrations by averaging over multiple elements. The domain over which smoothing occurs for a specific element depends on a certain maximum distance from the element centroid and the similarity of neighbouring elements regarding how their stresses change with increasing load. Using this approach, artificial peak stresses can be smoothed out without the risk of over-smoothing and underestimation of stresses. Additionally, stiffness driven stresses can be smoothed or shifted as well with this algorithm.

In order to handle the large amount of data involved in the optimization process, SQLite-databases are used, which act as a means of exchanging information between the algorithms and the FE-software. With these databases the full optimization procedure remains traceable and, for example, the convergence behaviour of the process can be investigated. Also, it allows for the model itself and all of its properties and results to be visualized and filtered in an efficient manner. Finally, the layup stored in the database can easily be used for an automated model build, resulting in a complete and robust design/optimization work flow.

In this work, the theoretical backgrounds, the logic of the work flow and practical issues of the applications are discussed. An industrial example illustrates distinct features of the whole process and shows the advantages of the approach for industrial applications.

**Optimization of a finite element model for a lightweight structure**

*Thomas Zwanowetz (University of Innsbruck)*

In order to determine the accuracy of finite element models, their predictions are compared to real world tests. If the results do not reach the desired level of consensus and these differences cannot be explained by engineering judgment alone, experimental data is used to upgrade the model. This step is known as parameter calibration.

Based on these results conclusions can be drawn to make further enhancements of the model. The presentation refers to a computational model of lightweight structures developed by the company INTALES GmbH Engineering solutions as well as its comparison with practical tests (loading) of the structure.

The presentation will include the following main topics:

● Full scale test and finite element model

● Parameter calibration with emphasis on the Nelder-Mead algorithm

● Adaptions of the model in particular the inclusion of friction

**Stochastic parameter calibration in a finite element model for lightweight structures**

*Daniel Sirianni (University of Innsbruck)*

In this presentation a statistical approach towards inverse problems will be employed to determine specific material parameters based on experimental data. In statistical inverse problems the unknown parameters are modelled as random variables such that Bayes’ theorem can be used to obtain the posterior distribution of the parameters. A widely used method to solve statistical inverse problems is the Markov chain Monte Carlo method such as the Metropolis-Hastings algorithm. The Metropolis-Hastings algorithm produces a Markov chain whose equilibrium distribution is the sought after posterior distribution, realized asymptotically by the end pieces of the chain. This approach will be applied to a large scale finite element model, developed by INTALES. This model contains some poorly known material parameters such as damping and stiffness parameters. To determine these parameters the previously mentioned method is used. Once the empirical distribution of the parameters is known, one can perform a stochastic investigation of the deviation of measurement and prediction. Furthermore, statistical features of the parameters can be determined with the help of Monte Carlo methods.

**On the influence of changing snow properties onto the structural behaviour of a snowboard undergoing a carved turn**

*Benoit Caillaud (University of Innsbruck)*

The purpose of this study was to numerically investigate the influence of different snow conditions onto the structural behaviour of a snowboard undergoing the conditions of a carved turn. The properties of different types of snow were numerically idealized and the deformations and pressure distribution along the contact edge of a snowboard were observed, in an attempt to understand their sensitivity to the environmental parameters. A static load bench was developed in-house and a simplified snowboard prototype was manufactured in order to represent the in-situ conditions. The experimental set-up was idealized in a finite element model, representing the composite structure and its loading environment. A method for the validation of the numerical model was proposed, based on the comparison of the experimental and numerical displacement fields, and consisted in a best-fit algorithm to superimpose both deformed shapes. The congruence between the two deformed surfaces was expressed with statistical means, and constituted the target function for optimization frameworks. Additionally, the contact pressure at the interface was experimentally assessed with the use of pressure measurement tape, and compared with the numerical predictions. The results of the finite element simulation were then explored to give an insight into the influence of the snow properties onto the structural behaviour.

**MissionToTheMoon**

*Jürgen Brandner, Manuel Schleiffelder (PTScientists GmbH)*

PTScientists GmbH is an aerospace engineering company and unique in Europe as the only privately owned and funded SME that is developing a commercial lunar payload delivery system. The company was founded in 2008 and has been developing its spacecraft and lander concept, ALINA autonomous landing and navigation module, since 2010 with a goal of showing that lunar exploration is commercially viable. PTScientists is already working on a highly innovative technology demonstration mission that aims to be the first private mission to land on the Moon. “Mission to the Moon” will include several key proof-of-concept elements, with a view to developing a sustainable lunar transport and communications infrastructure, which may also be used to further humanity’s exploration ambitions. The company is headquartered in Berlin, Germany with 55 fixed employees and a team of part-time contributors. The team behind the mission also features some senior engineers of the original Apollo program such as Jack W. Crenshaw who was in charge of the flight trajectories for Apollo. Technical cooperation partners include Audi AG, Vodafone Germany and Nokia Bell Labs, the German Aerospace Center (DLR), the European Space Agency (ESA) as well as technical universities in Germany and Austria.

**Conclusion of the event**

### Downloadable Material

### Location

### Venue and Date

The workshop will be held in the Hotel **dasMEI** at Mutters on Thursday, October 18, 2018.

### Arrival

**By car:** Highway A13, direction Brenner / Italy, exit Innsbruck-Süd, direction Mutters

**By public transport:** Train until Innsbruck main station, then change to the tram line "STB" (direction Fulpmes or Kreith)

You can find the exact location here

### Hotel booking

Accommodations can be found here: