Non-invasive real-time measurement of fluid properties in industrial processes is often considered very challenging. In this R&D project with Acosense AB the signal processing part of an active acoustic sensor is being optimized.
Acosense sensing technology consists of a clamp-on instrument measuring complex fluids in the process industry. The innovative technique can measure a multitude of properties simultaneously and can also handle properties of opaque, viscous or corrosive fluids.
The data analysis involved is to compare emitted and received acoustic spectra and translate this into a signal corresponding to fluid properties of interest. This is where advanced methods for model based signal processing, of which FCC has a longstanding knowledge and experience, comes in. Methods in this area includes linear and non-linear regression, classification, and statistical machine learning.