Application of Neural Network in Analysis Automation of Ultrasonic Cavitation Efficiency

View at publisher Summarise with VIRAT
Authors: Alena A. Vjuginova, Dmitry I. Ivanov, Anton Bunakov
Cited by: 0
References: 6

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

Ultrasonic cavitation makes it possible to implement a wide variety of ultrasonic technologies in liquid media such as ultrasonic dispersion, emulsification, extraction, lysis, homogenization, deagglomeration and many others. Evaluation of cavitation efficiency is necessary to ensure the repeatability of technological processes and the scaling of laboratory results to industrial ones. Traditional methods for evaluating the effectiveness of cavitation by erosion of aluminum foil are often characterized by insufficient speed and accuracy, and there is a need to automate the process. The paper explores the possibility of using neural network technologies to analyze the area of cavitation erosion for aluminum foil with a small number of training samples. The neural network architecture was developed and optimized to determine the erosion coefficient at a rate of several seconds for a sample, the lowest relative error was 3%, the results were analyzed, limitations and directions for further improvement were identified.

Fields of Study & Subjects

Ultrasound and Cavitation Phenomena Neural Networks and Applications Ultrasound and Hyperthermia Applications

Publication details

Journal / source
IEEE Conference of Russian Young Researchers in Electrical and Electronic Engineering
Pages
415–418
Publication date
3 February 2026
Language
EN