Local gas holdup measurement of a bubble column using SONIA-ultrasonic non-invasive method

A non-invasive technique for local gas holdup measurement of a bubble column using Self-Organized Network inspired by Immune Algorithm (SONIA) neural network and ultrasonic method is investigated. The energy attenuation and the transmission time difference of ultrasound are used as measurement parameters to obtain the local gas holdup in an air–water dispersion system using SONIA neural network reconstruction. Bubble size distributions in the bubble column are obtained by using a photographic method. The experimental results and simulations on three different mean bubble size condition show that the errors of SONIA neural network method is 1/9 times lower than those of the conventional back-propagation neural network. The results show a good agreement with measured data.