This thesis has investigated the Acospector, an active acoustic spectroscopy system produced by the Swedish company Acosense. The purpose was to find a transferability model of the system, such that the calibration time at (re)installation of the system could be reduced. Through theoretical investigations of the viscous wave equation for suitable geometries an experimental plan was devised in order to find a working transferability model.
It was concluded that component-wise first order linear multiplicative and additive models were insufficient for the task of finding the transferability of the system. Further frequency shift models through compositions could be rejected on the grounds of insufficient capabilities to capture the complexity of the system. Based on the theoretical investigation of the viscous wave equation, together with the conducted experiments, it could be concluded that the system is susceptible to changes in the mounting positions of the components as well as changes in certain media parameters.
Future work into the transferability of the system is proposed to focus on new ways of using the system, be that in terms of the signals used or the mounting set up. These might provide sufficiently simple transferability models for practical and valuable purposes.