The material parameters required to describe material behavior can change with the age of the components due to chemical and physical aging processes. The finite element method (FEM) is a tool for designing components for later use. The aim of this study is to correlate the change in the mechanical properties of silicone elastomers from standard tests over a longer period of time with the parameters of the Mooney–Rivlin model. To date, there are no publications on the development of the Mooney–Rivlin parameters of silicone elastomers over a storage period. For this purpose, the Shore A hardness, rebound elasticity, compression set and tensile properties were investigated over an aging period of approx. 200 days on two liquid silicone rubbers (LSRs) and one room-temperature-vulcanizing (RTV) silicone rubber. Depending on the silicone elastomer used, different trends in the characteristic values can be observed over the storage period. In general, increases in the Shore A hardness, rebound resilience and stress at a 100% strain with a decrease in the compression set can be determined. In addition to standard tensile tests, the material’s multiaxial behavior under tension was probed, and it was found that the similarly stress at a 100% strain increased. Finite element simulations verified the standard tensile test and corresponding Mooney–Rivlin model parameters. These parameters from the uniaxial tensile were validated in the multiaxial behavior, and the model’s accuracy in representing material properties and the influence of aging on the FEM simulation were affirmed.
Maintenance measures are widespread in the industrial environment, and various approaches to maintenance using artificial intelligence are increasingly gaining ground. Predictive assessments of system conditions ensure greater reliability and cost reductions through longer service life. The implementation of a machine learning and a deep learning algorithm for predictive maintenance through early damage detection on an electric rear axle test bench is presented in this paper. The algorithms were selected based on extensive literature research. This paper deals with the question of whether the approach of condition-based or statistical predictive maintenance provides more benefit for a development test bench with highly varying tests. The chosen method of deep learning and machine learning can predict damage for a specific device under test with an accuracy of up to 99% using only one torque signal. In fact, the machine learning approach was found to be sensitive to abnormal behavior on the test bench as well, leaving no abnormality undetected. Although the deep learning model was more resistant to other damage, a pretrained model can be applied to any similar device under test and deliver almost identical results. The methods presented can be adapted to various industrial applications with some adjustments, even without access to big data. This enables predictive maintenance from the very first applications.