Comparison of simulation and test data of carbon fibre composites with advanced measurement methods as DIC and OFDR

Kurzfassung

This thesis deals with experimental investigations and measurement data processing for the area of application of “Damage detection” as well as “Advanced FE-Simulation calibration”. The context of this work is the usage of the advantages of low intrusive, high-resolution measurement tools (as OFDR and DIC) for the application in new vehicle concepts for structural security observation as well as faster structural simulation calibration due to an increase in complexity but a decrease on the required amount of experimental test series. One innovative vehicle concept that is under development at DLR’s Institute of Vehicle Concepts is the hydrogen-powered Interurban Vehicle (IUV). For lightweight reasons, its body in white is mainly out of fiber reinforced thermoplastics. Within laminates and compared to a metallic material, damage can occur without being visible from the outside. Optical fiber sensing technology as OFDR can be implemented in security-relevant structural areas of the vehicle to measure strain and potentially detect low degradation of the laminate that could lead to damage, failure and safety issues. Another field of research is the application of sustainable materials in the automotive industry. New materials or new material compositions have always the difficulty that they have to be characterized before being able to use them in finite element simulations to evaluate their potential in complex application-oriented situations. Nowadays this is a laborious process, especially for fiber-reinforced materials in crash-relevant scenarios. The before mentioned measurement methods as OFDR and DIC could - in combination with optimization methods - help to reduce the extent of this process. In this thesis, experimental investigations, based on literature research, are proposed and partly carried out to evaluate some potentials and limits of these two measurement tools in practical applications. For strain data procurement, LUNA ODiSI fiber optics and GOM ARAMIS Adjustable measurement devices are used. LUNA’s ODiSI uses the principle of OFDR (Optical Frequency Domain Reflectometry) to measure heat signature and/or strain values along with the optical fiber glued/embedded in the specimen. GOM’s ARAMIS Adjustable uses the principle of DIC (Digital Image Correlation) to obtain element-wise strain readings. Few preliminary tests using 3D printed samples, and few tests from the set of proposed tests out of CFRP laminates are conducted. A comparison is drawn between the data obtained from three devices: DIC, OFDR and tension test machine. Limitations and capabilities of these devices are noted, and relevant data is exported for optimization in mechanical parameter identification. In the next step, a test is carried out where a coupon with an integrated optical fiber on multiple layers is impacted with an increasing amount of energy until breakage. OFDR signal results are shown and interpreted for damage detection. Aiming towards fast FE-calibration, an optimization tool for parameter identification is set up. First, reference simulation models with which the test data will be compared are built. Many simulation models, according to the proposed tests, are developed and adapted to the experiments via QD-Library from the python environment. Second, for proper difference comparison, a python tool is developed which projects measured strain data on-to the simulation data. Third, an optimization python tool that iteratively varies material parameters to reduce differences between test and simulation data is developed. To validate the capability of the optimization algorithm, it is applied to reference simulations. Applied on experimental data, mechanical parameters are predicted with high accuracy, and it can be concluded that the method can be followed for further advanced experiments.