This study explores the combination of ultrasound technology with a detection algorithm to categorize flow regimes in bubble columns used for aeration in aquaculture.An ultrasonic velocity profiler is used to obtain t...This study explores the combination of ultrasound technology with a detection algorithm to categorize flow regimes in bubble columns used for aeration in aquaculture.An ultrasonic velocity profiler is used to obtain the standard deviation of the bubble velocity distributed throughout the column.The bubble velocity data for three known flow regimes were used to develop a probability density function(PDF)classification model.The experimental apparatus consisted of a circular tank equipped with a bubble generator and gas hold-up monitoring systems.The flow regimes of the experimental fluid were determined,and the classification was conducted via the PDF method.The results demonstrate that the classification accuracy is not lower than that of traditional machine learning methods.展开更多
基金supported by the Center of Excellence on Instru-mentation Technology and Automation(CEITA),Department of Instru-mentation and Electronics Engineering,Faculty of Engineering,King Mongkut’s University of Technology North Bangkok,Thailand。
文摘This study explores the combination of ultrasound technology with a detection algorithm to categorize flow regimes in bubble columns used for aeration in aquaculture.An ultrasonic velocity profiler is used to obtain the standard deviation of the bubble velocity distributed throughout the column.The bubble velocity data for three known flow regimes were used to develop a probability density function(PDF)classification model.The experimental apparatus consisted of a circular tank equipped with a bubble generator and gas hold-up monitoring systems.The flow regimes of the experimental fluid were determined,and the classification was conducted via the PDF method.The results demonstrate that the classification accuracy is not lower than that of traditional machine learning methods.
