Browsing by Author "Kästner, Markus"

Now showing 1 - 3 of 3

Open Access
Data: Neural networks meet anisotropic hyperelasticity: A framework based on generalized structure tensors and isotropic tensor functions
(Technische Universität Dresden, 2024-11-21) Kalina, Karl Alexander; Brummund, Jörg; Sun, WaiChing; Kästner, Markus
This collection provides homogenized datasets including deformation gradient, free energy, stress tensor and material tangent for anisotropic hyperelastic composites. Five different representative volume elements (RVEs) are included: An RVE of a fiber reinforced material with stochastic fiber distribution (stochastic fibers), a unit cell with a hexagonal fiber arrangement (hexagonal fibers), a unit cell with one spherical inclusion (cubic sphere), an RVE with a plane-like arrangement of particles (plane spheres), and an RVE with an arrangement of particles in a chain-like structure (chain spheres). All components, i.e., matrix, particles and fibers are assumed to be compressible and isotropic. For all, a two-parametric neo-Hookean model was chosen. The data have been generated by using an in-house finite element code based on Matlab. The data belongs to the work "Neural networks meet anisotropic hyperelasticity: A framework based on generalized structure tensors and isotropic tensor functions" by Kalina et al.; https://doi.org/10.48550/arXiv.2410.03378. Further information on the data can be found there.
Open Access
eBike measurements for fatigue monitoring using acceleration differences
(Technische Universität Dresden, 2025-06-26) Heindel, Leonhard; Hantschke, Peter; Kästner, Markus
In fatigue monitoring, the aim is to approximate the fatigue damage accumulation in a system, for example to schedule maintenance intelligently. Virtual sensing can be deployed to obtain the required information more efficiently by estimating them from readily available sensor data. This dataset contains acceleration and strain measurements from a sensor equipped eBike, designed to develop such methods. Using the acceleration data, the aim is to estimate the fatigue damage accumulation at the strain gauge positions. Here, the sensors are positioned such that differences between multiple acceleration sensors provide information that is related to the deformation of the eBike frame, so that a connection to strain gauge data can be established by computing differences between integrated displacement data. The dataset includes data intended for training and testing of data-driven modeling approaches, which is obtained from regular eBike usage. Additional labeled data enables the analysis of specific riding maneuvers.
Open Access
Predictive maintenance demonstrator dataset with individual load histories
(Technische Universität Dresden, 2025-04-14) Heindel, Leonhard; Hantschke, Peter; Kästner, Markus
Predictive maintenance aims to develop methods that are capable of predicting component failure before it occurs. Virtual sensing methods predict unmeasured physical quantities from available measurement data. These methods offer significant benefits to predictive maintenance, since virtual sensors can be used to estimate quantities that are difficult to measure. In many real applications, the time to failure is in the range of years, complicating the development and validation of predictive maintenance and virtual sensing approaches. This dataset provides a demonstrator example where failure occurs based on individual load histories. The sensor setup consists of simple notched steel specimens, which are clamped between two servo-hydraulic cylinders of a fatigue test bench. It is designed to provide a virtual sensor use case with independent training and testing data, so that the dataset can be used for algorithm development and benchmarking purposes.