A Monte Carlo-Based Method for Assessing the Measurement Uncertainty in the Training and Use of Artificial Neural Networks

Open access

Abstract

When an artificial neural network is used to determine the value of a physical quantity its result is usually presented without an uncertainty. This is due to the difficulty in determining the uncertainties related to the neural model. However, the result of a measurement can be considered valid only with its respective measurement uncertainty. Therefore, this article proposes a method of obtaining reliable results by measuring systems that use artificial neural networks. For this, it considers the Monte Carlo Method (MCM) for propagation of uncertainty distributions during the training and use of the artificial neural networks.

[1] Haykin, S. (1999). Neural Networks: A Comprehensive Foundation. 2nd ed., Prentice Hall.

[2] Singaram, L. (2011). ANN prediction models for mechanical properties of AZ61 MG alloy fabricated by equal channel angular pressing. Int. J. of Res. and Reviews in Appl. Sciences, (8), 337−345.

[3] Ghobadian, B., Rahimi, H., Nikbakht, A.M., Najafi, G., Yusaf, T.F. (2009). Diesel engine performance and exhaust emission analysis using waste cooking biodiesel fuel with an artificial neural network. Renew. Energy, (34), 976−982.

[4] Ertunc, H.M., Hosoz, M. (2006). Artificial neural network analysis of a refrigeration system with an evaporative condenser. Appl. Therm. Eng., (26), 627−635.

[5] Arcaklioğlu, E., Çavuşoğlu, A., Erişen, A. (2004). Thermodynamic analyses of refrigerant mixtures using artificial neural networks. Appl. Energy, (78), 219−230.

[6] Russel, S., Norvig, P. (2003). Artificial Intelligence: A Modern Approach. 2nd ed., Prentice Hall.

[7] BIPM, IEC, IFCC, ILAC, ISO, IUPAC, IUPAP, OIML, (2008). JCGM 100: Evaluation of measurement data - Guide to the expression of uncertainty in measurement, 134.

[8] BIPM, IEC, IFCC, ILAC, ISO, IUPAC, IUPAP, OIML, (2008). JCGM 200: International vocabulary of metrology - Basic and general concepts and associated terms, 104.

[9] Neto, A.C.R., Neves, C.A.M., Roisenberg, M. (2013). Comparative study on local and global strategies for confidence estimation in neural networks and extensions to improve their predictive power. Neural Comput. & Appl., (22), 1519−1530.

[10] Zapranis, A., Livanis, E. (2005). Prediction intervals for neural network models. Proc. 9th WSEAS International Conference on Computers, Athens, Greece, 7.

[11] Methaprayoon, K., Yingvivatanapong, C., Wei-Jen, L., Liao, J.R. (2007). An Integration of ANN Wind Power Estimation Into Unit Commitment Considering the Forecasting Uncertainty. IEEE Trans. Ind. Appl., (43), 1441−1448.

[12] Chryssolouris, G., Lee, M., Ramsey, A. (1996). Confidence interval prediction for neural network models. IEEE Trans. on Neural Netw., (7), 229−232.

[13] Hwang, J.T.G., Ding, A.A. (1997). Prediction intervals for artificial neural networks. J. of the American Statistical Association, (92), 748−757.

[14] De Veaux, R.D., Schweinsberg, J., Schumi, J., Ungar, L.H. (1998). Prediction intervals for neural networks via nonlinear regression. Technometrics, (40), 273−282.

[15] Papadopoulos, G., Edwards, P.J., Murray, A.F. (2001). Confidence estimation methods for neural networks: a practical comparison. IEEE Trans. on Neural Netw., (12), 1278−1287.

[16] Bakhary, N., Hao, H., Deeks, A.J. (2007). Damage detection using artificial neural network with consideration of uncertainties. Eng. Structures, (29), 2806−2815.

[17] Fluke Corporation. (1994). Calibration: Philosophy in Practice. 2nd ed.

[18] BIPM, IEC, IFCC, ILAC, ISO, IUPAC, IUPAP, OIML, (2008). JCGM 101: Supplement 1 to the “Guide to the expression of uncertainty in measurement” - Propagation of distributions using a Monte Carlo method. 90.

[19] Moscati, G., Mezzalira, L.G., Santos, F.D. (2004). Measurement uncertainty using the Monte Carlos method in the context of GUM-Supplement 1. Proc. ENQUALAB National Meeting for Laboratories Quality, São Paulo, Brazil, 4.

[20] Gusman, C.S.A. (2011). Measurement uncertainty evaluation for neural networks applied on predictive maintenance of power transformers. Master Degree Dissertation on Metrology. Pontifícia Universidade Católica, Rio de Janeiro.

[21] Zhu, J., Wang, Z., Xia, X., Lei, J. (2005). One parameterized model of indirect measurement based on neural network and its sensitivity coefficient computing. Proc. 7th Int. Symp. on Measurement Technol. and Intell. Instrum., J. of Phys.: Conf. Series 13.

[22] Fortuna, L., Giannone, P., Graziani, S., Xibilia, M.G. (2007). Virtual instruments based on stacked neural networks to improve product quality monitoring in a refinery. IEEE Trans. on Instrum. and Measurement, (56), 95−101.

[23] Hu, Y.H., Hwang, J. (2002). Handbook of neural network signal processing. CRC Press LLC, USA, 384.

[24] Edwards, P.J., Peacock, A.M., Renshaw, D., Hannah, J.M., Murray, A.F. (2002). Minimizing risk using prediction uncertainty in neural network estimation fusion and its application to papermaking. IEEE Trans. on Neural Netw., (13), 726−731.

[25] Ahmad, Z., Zhang, J. (2002). A comparison of different methods for combining multiple neural networks models. Proc. Int. Joint Conf. on Neural Netw., (1), 828−833.

[26] Efron, B., Tibshirani, R. (1993). An introduction to the bootstrap. Chapman & Hall, NY, 450.

[27] Tibshirani, R. (1994). A comparison of some error estimates for neural network models. Department of Preventive Medicine and Biostatistics, University of Toronto.

Metrology and Measurement Systems

The Journal of Committee on Metrology and Scientific Instrumentation of Polish Academy of Sciences

Journal Information


IMPACT FACTOR 2016: 1.598

CiteScore 2016: 1.58

SCImago Journal Rank (SJR) 2016: 0.460
Source Normalized Impact per Paper (SNIP) 2016: 1.228

Metrics

All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 226 188 8
PDF Downloads 108 98 3