In-Line Monitoring and Control of Rheological Properties through Data-Driven Ultrasound Soft-Sensors

• Published in: Sensors (Basel, Switzerland) Vol. 19; no. 22; p. 5009
• Main Authors: Tronci, Stefania, Van Neer, Paul, Giling, Erwin, Stelwagen, Uilke, Piras, Daniele, Mei, Roberto, Corominas, Francesc, Grosso, Massimiliano
• Format: Journal Article
• Language: English
• Published: Switzerland MDPI AG 16-11-2019; MDPI
• Subjects: Acoustics; Algorithms; Batch processes; data-driven; decision support; hybrid approach; industry 4.0; Ingredients; Inverse problems; neural network; Neural networks; Newtonian fluids; Non Newtonian fluids; non-newtonian fluid; Principal components analysis; Rheological properties; Rheology; Sensors; Signal processing; Travel; Ultrasonic imaging; Ultrasound; ultrasound sensor; Velocity; Viscosity; viscosity curve; hybrid approach; non-Newtonian fluid; data-driven; viscosity curve; Industry 4.0; ultrasound sensor; decision support; neural network
• ISSN: 1424-8220; 1424-8220
• DOI: 10.3390/s19225009

The use of continuous processing is replacing batch modes because of their capabilities to address issues of agility, flexibility, cost, and robustness. Continuous processes can be operated at more extreme conditions, resulting in higher speed and efficiency. The issue when using a continuous process is to maintain the satisfaction of quality indices even in the presence of perturbations. For this reason, it is important to evaluate in-line key performance indicators. Rheology is a critical parameter when dealing with the production of complex fluids obtained by mixing and filling. In this work, a tomographic ultrasonic velocity meter is applied to obtain the rheological curve of a non-Newtonian fluid. Raw ultrasound signals are processed using a data-driven approach based on principal component analysis (PCA) and feedforward neural networks (FNN). The obtained sensor has been associated with a data-driven decision support system for conducting the process.