Applicability of Artificial Neural Network in Hydraulic Experiments Using a New Sewer Overflow Screening Device

Abstract

During wet weather conditions, sewer overflows to receiving water bodies raise serious environmental, aesthetic and public health problems. These issues trigger the need the most appropriate device/system for a particular installation, especially at unmanned remote locations. A new sewer overflow device consists of a rectangular tank and a sharp crested weir with a series of vertical combs is presented. A series of laboratory tests to determine trapping efficiencies for common sewer solids were conducted for different flow conditions, number of combs layers and spacing of combs. To overcome physical limitations inherent in laboratory studies such as significant cost and time. Artificial neural model was adopted as it has the capacity to accurately predict the outcome of complex, non-linear physical systems with relatively poorly understood physicochemical processes. A series of laboratory tests were conducted with 55 different sets of data. Forty-seven sets of experimental data are used with 60% for training, 20% each for testing and validation of the model. A separate validation data sets were used to judge the overall performance of the trained network. The model can successfully predict the experimental results with more than 90% accuracy with an average absolute percentage error of around 7%.

Key Words:

• Screening device
• sewer solids
• capture efficiency
• artificial neural network (ANN)

Additional information

Notes on contributors

M A Aziz

Md. Abdul Aziz is working as floodplain officer at Wimmera Catchment Management Authority, Horsham, Victoria, Australia. He received the Swinburne University Post Graduate Research Award for his doctoral study in 2009, and Japanese Government Monbukagakusho Scholarship in 2002 for his Master of Engineering at Tokyo Institute of Technology, Tokyo, Japan. He graduated as a Civil Engineer from Bangladesh Institute of Technology, Rajshahi, in 2001. His key research interests include artificial neural network application in hydraulic experiments, computational fluid dynamics modelling, and floodplain modelling.

M Imteaz

Dr Monzur Imteaz is a Senior Lecturer and Postgraduate Program Coordinator within the Civil Engineering group of Swinburne University of Technology, Melbourne, Australia. He completed his PhD in 1997 on Lake Water Quality Modelling from Saitama University, Japan. After his PhD, he worked with the Institute of Water Modelling, Bangladesh, in collaboration with the Danish Hydraulic Institute. Later he has completed his post-doctoral research at University of Queensland, Brisbane. Before joining Swinburne he was involved with several Australian local and state government departments. At Swinburne, Also, he has been actively involved with various research on sustainability, water recycling and modelling, developing decision support tools, and rainfall forecasting using artificial neural networks.

T A Choudhury

Tanveer Ahmed Choudhury is currently doing his PhD at Swinburne University of Technology, Melbourne, Australia. He secured the Swinburne University Post Graduate Research Award for his doctoral study. His expertise is in the fields of artificial neural networks, machine learning and expert systems. He received his BSc in Electrical and Electronic Engineering from Islamic University of Technology, Gazipur, Bangladesh, in 2007.

D Phillips

Dr Donald Phillips has since spent over 55 years in the water industry as a civil engineer, consultant, educationalist and researcher. He is a director of an intellectual property company. As a civil engineer he has engaged in many aspects of the design and construction of irrigation pipelines and structures, high dams, works of flood control, urban water supplies, bridges, highways, and freeways. As a consultant, he has been engaged in a wide range of water-related activities for both private and public companies throughout his professional life. As an educationalist, he spent some 27 years as a Senior Lecturer at the now Swinburne University of Technology lecturing on all civil engineering subjects, with emphasis on water engineering at graduate and post-graduate levels. As a researcher, he developed the methodology of control of urban runoff from intensively developed sites, for which he was awarded his doctorate in 1994.