Traditional measurement methods of particle size and size distribution require sample preparation and mostly sample dilution. This causes time delay and measurement errors as suspension properties can change significantly at low concentrations. To date online measurement systems face significant challenges, for instance, multiple scattering, low sensitivity, lack of physical properties information, non-uniform mixing and aggregation. To solve these problems, understanding and characterisation of process suspensions are required.

This paper reports results of a project that tackles this measurement challenge by combining multimodel sensors to allow multidimensional and multiscale distributed characterisation of process parameters and product quality, integrating the outputs with sensor date mining system for interpretation and investigation, to achieve process optimisation and control.

This sensing system combines sensors for ultrasonic particle sizing, zeta potential measuring and process tomographic visualisation. Based on electrical and acoustic principles, the electroacoustic sensors are used in capturing the information of sound frequency attenuation resulting from dispersed particles. Zeta potential and tomographic sensors are used for monitoring suspension stability and visualising dispersed phase flow (concentration distribution), respectively. By combining the three types of sensors, particle flow and size distribution can be characterised.

This paper presents static and dynamic experimental results using samples of varied properties, size ranges and concentrations. A novel technique for deriving size distribution from multiple sensors is also presented.