A new focused ultrasonic sensor is proposed, based on which the measurement system for particle size distribution measurement of suspended sediment is established. Compared with the traditional ultrasonic sensors, the one used in this paper is equipped with piezoelectric transducer (PZT) on an arc-shaped shell, to concentrate ultrasonic beams on one measurement point. The sensor is used to measure the particle size distribution of suspended sediment. The experiments were carried out on water-sediment mixtures with different particle size distribution. Due to multiple parameters and non-linearity of the ultrasonic attenuation model, the artificial bee colony (ABC) inversion algorithm is used to estimate particle size distribution, thus improving measurement accuracy. The particle sizes obtained by sieving method are seen as reference values. The results indicate that whether the suspended particles are subject to a unimodal distribution, uniform distribution or random distribution, the particle size distribution obtained by ABC inversion algorithm is consistent with the result obtained by the sieve method. The results demonstrate that the method has good utility and accuracy within the low concentration range.
Precise and fast microdroplet size distribution measurement using deep learning
