IJE TRANSACTIONS B: Applications Vol. 15, No. 2 (July 2002) 157-166   

downloaded Downloaded: 46   viewed Viewed: 1474

H. B. Bahar

Faculty of Engineering, Tabriz University
PO Code 5166614776, Tabriz, Iran
( Received: December 27, 2001 – Accepted in Revised Form: May 09, 2002 )

Abstract    A functional relationship between two variables, applied mass to a weighing platform and estimated mass using Multi-Layer Perceptron Artificial Neural Networks is approximated by a linear function. Linear relationships and correlation rates are obtained which quantitatively verify that the Artificial Neural Network model is functioning satisfactorily. Estimated mass is achieved through recalling the trained Artificial Neural Network model on a set of waveforms resulting from applied masses over the operating range of the weighing platform. In this work the Least-Squares Fit (LSF) method for straight line and correlation rate R between the applied and estimated masses are used to investigate the accuracy of the estimated masses. The slope of the linear functions together with correlation rates R are computed for both simulation and experimental data. The numerical results confirm the correctness of neural network technique in estimating the applied mass m(t).


Keywords    Correlation Rate, Least-Squares Fit, Artificial Neural Networks, Dynamic Weight Estimation, Platform Model, Time Series Data, System Identification



1. Ferguson, C., “From Beam to Load Cell”, Engineering, (1974), 202-207.

2. Lolley, R. A., “A Review of Industrial Weighing Systems Part 1”, Meas. and Cont., (1976), Vol. 9, 411-416.

3. Lolley, R. A., “A Review of Industrial Weighing Systems Part 2”, Meas. and Cont., Dec. 1976, Vol. 9, 435-439.

4. Norden, K. E., “Electronic Weighing in Industrial Processes”, Granada Publishing, Technical Books, (1984).

5. Shi, W. J., White, N. M. and Brignell, J. E., “AdaptiveX10 - Vol. 15, No. 2, July 2002 IJE Transactions B: Applications Filters in Load Cell Response Correction”, Sensors and Actuators, Vol. 37, (1993), 280-285.

6. Shu, W. Q., “Dynamic Weighing Under Non-Zero Initial Conditions”, IEEE Trans. on Inst. and Meas., Vol. 42, No. 4, (1993), 806-811.

7. Danaci, M. and. Horrocks, D. H., “A Non-linear Regression Technique For Improved Dynamic Weighing”, Proc. Eur. Conf. on Circuit Theory and Design, Istanbul, Turkey,(August 1995), 507-510.

8. Danaci, M., “Signal Processing Techniques Applied to Dynamic Weighing Systems”, Ph.D. Thesis, School of Engineering, Cardiff University of Wales UK, (December1996).

9. Bahar, H. B. and Horrocks, D. H., “Dynamic Weight Estimation Using Artificial Neural Network”, Artificial Intelligence In Engineering, Vol. 12, (1998), 135-139.

10. Pandya, A. S. and Macy, R. B., “Pattern Recognition with Neural Networks in C++”, CRC Press Inc., (1996).

11. Haykin, S., “Neural Networks, A Comprehensive Foundation”, Macmillan College Publishing Company, (1994).

12. Pham, D. T. and Liu, X., “Neural Networks for Identification, Prediction and Control”, Springer-Verlag, 2nd Edition, (1995).

13. Masters, T., “Practical Neural Network Recipes in C++”, Academic Press, Boston, (1993).

14. Bevington, P. R., “Data Reduction on Error Analysis forthe Physical Sciences”, McGraw-Hill, 2nd Edition, (1994).

15. Chatfield, C., “The Analysis of Time Series, An Introduction”, Chapman & Hall, Fifth Edition, (1996).


International Journal of Engineering
E-mail: office@ije.ir
Web Site: http://www.ije.ir