Model based virtual acoustic sensing for augmented system monitoring of flexible multibody systems

The sustainable development and maintenance of vehicles and machines plays an increasingly important role. This involves the challenging task of monitoring and balancing the often conflicting criteria related to human health, profitability and environment. The knowledge of the forces, torques and stresses in components is of great importance to make sure that machines and vehicles stay within safety limits and retain their reliability over time. However, these quantities are typically hard to measure directly, since existing stress, force and torque sensors are intrusive and expensive.
Therefore, augmented sensing strategies have been gaining momentum to estimate these quantities from easier to measure quantities. This is achieved by intelligently coupling physics based numerical models with affordable measurements. However, the potential of these physical models is not yet exploited to its full extent. The goal of the project is to investigate how affordable acoustic measurements can be used to estimate hard to measure quantities in complex multibody systems. To achieve this goal, accurate multi-physical models in which multibody dynamics and acoustics are coupled need to be defined and implemented. To reduce the computational complexity dedicated model order reduction techniques will be developed. Several augmented sensing configurations will be analyzed from both a theoretical and experimental perspective to demonstrate the power of this innovative way of virtual sensing.