bullet Sensors & Transducers Journal

    (ISSN 1726- 5479)

205.767

2008 e-Impact Factor

25 Top Downloaded Articles

Journal Subscription 2010

Editorial Calendar 2010

Submit an Article

Editorial Board

Current Issue

Sensors & Transducers

Sensors & Transducers Journal 2008

Sensors & Transducers Journal 2007

2000-2002 S&T e-Digest Contents

2003 S&T e-Digest Contents

2004 S&T e-Digest Contents

2005 S&T e-Digest Contents

2006 S&T e-Digest Contents

 

Best Articles 2009

 

 

 

Vol. 114, Issue 3, March 2010, pp. 41- 55

 

Bullet

 

A Neural Network Approach to Fluid Level Measurement in Dynamic Environments Using a Single Capacitive Sensor

 

Edin TERZIC, 2Romesh NAGARAJAH, 1Muhammad ALAMGIR

1Delphi Corporation, 86 Fairbank Road, Clayton South VIC 3169, Australia

*Tel.: +(61) 418 139 277

2Swinburne University, Burwood Hwy, Hawthorn VIC 3122, Australia

Tel.: +(61) 3 9214 8530

E-mail: edin.terzic@delphi.com, nragarajah@swin.edu.au, muhammad.alamgir@delphi.com

 

 

Received: 11 January 2010   /Accepted: 22 March 2010   /Published: 29 March 2010

 

Abstract: A measurement system has been developed using a single tube capacitive sensor to accurately determine the fluid level in vehicular fuel tanks. A novel approach based on artificial neural networks based signal pre-processing and classification has been described in this article. A broad investigation on the Backpropagation neural network and some selected signal pre-processing filters, namely, Moving Mean, Moving Median, and Wavelet Filter has also been presented. An on field drive trial was conducted under normal driving conditions at various fuel volumes ranging from 5 L to 50 L to acquire training samples from the capacitive sensor. A second field trial was conducted to obtain test samples to verify the performance of the neural network. The neural network was trained and verified with 50 % of the training and test samples. The results obtained using the neural network approach having different filtration methods are compared with the results obtained using simple Moving Mean and Moving Median functions. It is demonstrated that the Backpropagation neural network with Moving Median filter produced the most accurate outcome compared with the other signal filtration methods.

 

Keywords: Accurate level measurement, Liquid slosh, Backpropagation Neural Network, Intelligent sensor

 

Acrobat reader logo Click <here> or title of paper to download the full pages article (1.09 Mb)

 

 

 

 


1999 - 2010 Copyright , International Frequency Sensor Association (IFSA). All Rights Reserved.


Home - News - Links - Archives - Tools - Standardization - Patents - Marketplace - Projects - Wish List - Subscribe - Search - Membership - Submit Press Release

 Members Area -Sensors Portal -Training Courses - S&T Digest - For advertisers - Bookstore - Forums - Polls - Sensor Jobs - e-Shop - Site Map