Discovering key meteorological variables in atmospheric corrosion through an artificial neural network model

This paper presents a deterministic model for the damage function of carbon steel, expressed in μm of corrosion penetration as a function of cumulated values of environmental variables. Instead of the traditional linear model, we designed an Artificial Neural Network (ANN) to fit the data. The ANN numerical model shows good results regarding goodness of fit and residual distributions. It achieves a RMSE value of 0.8 μm and a R 2 of 0.9988 while the classical linear regression model produces 2.6 μm and 0.9805 respectively. Besides, F LOF for the ANN model were next to the critical value. The improved accuracy provides a chance to identify the most relevant variables of the problem. The procedure was to add/remove one after the other the variables and perform from scratch the corresponding training of the ANN. After some trial and error as well as phenomenological arguments, we were able to show that some popular meteorological variables like mean relative humidity and mean temperature shown no relevance while the results were clearly improved by including the hours with RH < 40%. The results as such might be valid for a limited geographical region, but the procedure is completely general and applicable to other regions.

[1]  Hal S. Stern,et al.  Neural networks in applied statistics , 1996 .

[2]  Brad Warner,et al.  Understanding Neural Networks as Statistical Tools , 1996 .

[3]  Robert A. Cottis,et al.  Phenomenological modelling of atmospheric corrosion using an artificial neural network , 1999 .

[4]  Ivan S. Cole,et al.  Holistic model for atmospheric corrosion Part 1 - Theoretical framework for production, transportation and deposition of marine salts , 2003 .

[5]  Philipp Slusallek,et al.  Introduction to real-time ray tracing , 2005, SIGGRAPH Courses.

[6]  T. Graedel,et al.  Gildes model studies of aqueous chemistry. II. The corrosion of zinc in gaseous exposure chambers , 1996 .

[7]  P. Rousseeuw,et al.  A fast algorithm for the minimum covariance determinant estimator , 1999 .

[8]  G. Lewicki,et al.  Approximation by Superpositions of a Sigmoidal Function , 2003 .

[9]  P. Rousseeuw.  Least Median of Squares Regression , 1984 .

[10]  P Albrecht,et al.  Composite Modeling of Atmospheric Corrosion Penetration Data , 1994 .

[11]  Sebastián Feliu,et al.  THE PREDICTION OF ATMOSPHERIC CORROSION FROM METEOROLOGICAL AND POLLUTION PARAMETERS--II. LONG-TERM FORECASTS , 1993 .

[12]  G. G. Stokes.  "J." , 1890, The New Yale Book of Quotations.

[13]  I. Cole,et al.  A Study of the Wetting of Metal Surfaces in Order to Understand the Processes Controlling Atmospheric Corrosion , 2004 .

[14]  T. Graedel.  Gildes model studies of aqueous chemistry. I. Formulation and potential applications of the multi-regime model , 1996 .

[15]  E. S. Pearson.  Biometrika tables for statisticians , 1967 .

[16]  Hal S. Stern,et al.  [Neural Networks in Applied Statistics]: Reply , 1996 .

[17]  Gustavo A. Cragnolino,et al.  Application of Accelerated Corrosion Tests to Service Life Prediction of Materials , 1994 .

[18]  N. Draper,et al.  Applied Regression Analysis , 1966 .

[19]  T. Graedel,et al.  Gildes model studies of aqueous chemistry. III. Initial SO2-induced atmospheric corrosion of copper , 1996 .

[20]  F. Lipfert,et al.  Atmospheric Corrosion Model for Galvanized Steel Structures , 1992 .

[21]  Norman R. Draper,et al.  Applied regression analysis (2. ed.) , 1981, Wiley series in probability and mathematical statistics.

[22]  Stuart Lyon,et al.  An approach to the modelling of atmospheric corrosion , 1995 .

[23]  Sebastián Feliu,et al.  The prediction of atmospheric corrosion from meteorological and pollution parameters—I. Annual corrosion , 1993 .