Virtual Sensing Application Cases Exploiting Various Degrees of Model Complexity Karl Janssens, Bart Forrier, Roberta Cumbo, Enrico Risaliti, Bram Cornelis, Tommaso Tamarozzi, and Wim Desmet 1 Introduction Several approaches exist to obtain insight in the performance of a product. A first approach is the analysis of measurement data gathered in a test campaign. The major advantage of this approach is that measurements reflect the performance of the real system in operating conditions. However, setting up a full measurement campaign can be costly and time-consuming, and not all locations of interest are always accessible. Alternatively, numerical simulations can be used to analyze the system. While simulations provide clear insight in the underlying physical processes and generate full-field information, they often only approximately describe the real behavior of the system. Virtual sensing is a new emerging approach which attempts to combine the best of both worlds by coupling numerical simulation models with easily attainable measurement data. This enables the model-based estimation or virtual measurement of physical quantities at any desired location and blends the real-world accuracy of measurements with the insights and flexibility of models. The fusion of measure- ment data and simulation models in an estimator framework creates tremendous application potential for the industry, both in the product design cycle and product usage phase. K. Janssens () · R. Cumbo · B. Cornelis Siemens Industry Software NV, Leuven, Belgium e-mail: Karl.Janssens@siemens.com B. Forrier · E. Risaliti · T. Tamarozzi Siemens Industry Software NV, Leuven, Belgium Katholieke Universiteit Leuven, Heverlee, Belgium W. Desmet Katholieke Universiteit Leuven, Heverlee, Belgium © Springer Nature Switzerland AG 2021 S. Oberst et al. (eds.), Vibration Engineering for a Sustainable Future, https://doi.org/10.1007/978-3-030-47618-2_20 161