Air pollutant emissions prediction by process modelling - Application in the iron and steel industry in the case of a re-heating furnace

Monitoring air pollutant emissions of large industrial installations is necessary to ensure compliance with environmental legislation. Most of the available measurement techniques are expensive, and measurement conditions such as high-temperature emissions, difficulty of access, are often difficult. That is why legislation can not impose a permanent emission monitoring in many countries. The possibility to replace it with predictive models based on the routine measurements of the main control parameters of the installation is analysed in this paper. In order to identify these models, a special measurement campaign of emissions must be performed or, alternatively, a deterministic modelling of the process can be developed. This study was carried out in the case of a real installation in the steel industry i.e. a billet re-heating furnace. Physical phenomena involved in combustion within the furnace were complex enough to prefer an empirical black-box modelling of the furnace over a deterministic approach. A 3-week monitoring campaign of fume emissions at the stack was performed; furnace process parameters during the same period were available. The relationship between CO"2 emissions and furnace process parameters could successfully be expressed linearly, while NO"2 emission modelling required a non-linear model. Artificial neural networks modelling revealed a good ability to predict NO"2 and CO"2 emissions.

[1]  Heekuck Oh,et al.  Neural Networks for Pattern Recognition , 1993, Adv. Comput..

[2]  Stephen Dorling,et al.  Statistical surface ozone models: an improved methodology to account for non-linear behaviour , 2000 .

[3]  G. David Garson,et al.  Interpreting neural-network connection weights , 1991 .

[4]  M. Kolehmainen,et al.  Neural networks and periodic components used in air quality forecasting , 2001 .

[5]  Gilbert Saporta,et al.  Probabilités, Analyse des données et statistique , 1991 .

[6]  A. T. C. Goh,et al.  Back-propagation neural networks for modeling complex systems , 1995, Artif. Intell. Eng..

[7]  Wray L. Buntine,et al.  Bayesian Back-Propagation , 1991, Complex Syst..

[8]  M. Gardner,et al.  Neural network modelling and prediction of hourly NOx and NO2 concentrations in urban air in London , 1999 .

[9]  M. Georgiopoulos,et al.  Feed-forward neural networks , 1994, IEEE Potentials.

[10]  Saleh M. Al-Alawi,et al.  Assessment and prediction of tropospheric ozone concentration levels using artificial neural networks , 2002, Environ. Model. Softw..

[11]  D. M. Titterington,et al.  Neural Networks: A Review from a Statistical Perspective , 1994 .

[12]  Klaus-Robert Müller,et al.  Asymptotic statistical theory of overtraining and cross-validation , 1997, IEEE Trans. Neural Networks.

[13]  A. Comrie.  Comparing Neural Networks and Regression Models for Ozone Forecasting , 1997 .

[14]  Holger R. Maier,et al.  Neural networks for the prediction and forecasting of water resource variables: a review of modelling issues and applications , 2000, Environ. Model. Softw..

[15]  Gavin C. Cawley,et al.  A rigorous inter-comparison of ground-level ozone predictions , 2003 .

[16]  Kurt Hornik,et al.  Degree of Approximation Results for Feedforward Networks Approximating Unknown Mappings and Their Derivatives , 1994, Neural Computation.

[17]  Nejib Smaoui,et al.  The Box-Jenkins analysis and neural networks: prediction and time series modelling , 2003 .

[18]  Gavin C. Cawley,et al.  Maximum likelihood cost functions for neural network models of air quality data , 2003 .

[19]  Laurene V. Fausett,et al.  Fundamentals Of Neural Networks , 1994 .

[20]  Kurt Hornik,et al.  Multilayer feedforward networks are universal approximators , 1989, Neural Networks.

[21]  S. M. Lo,et al.  Application of evolutionary neural network method in predicting pollutant levels in downtown area of Hong Kong , 2003, Neurocomputing.

[22]  Yoshua Bengio,et al.  Pattern Recognition and Neural Networks , 1995 .

[23]  Ali Elkamel,et al.  Measurement and prediction of ozone levels around a heavily industrialized area: a neural network approach , 2001 .

[24]  David E. Rumelhart,et al.  Predicting the Future: a Connectionist Approach , 1990, Int. J. Neural Syst..

[25]  Sandhya Samarasinghe.  Fundamentals of Neural Networks and Models for Linear Data Analysis , 2006 .

[26]  Roberto Serra,et al.  Neural Networks for Sulphur Dioxide Ground Level Concentrations Forecasting , 2000, Neural Computing & Applications.

[27]  Gabriel Ibarra-Berastegi,et al.  Regression and multilayer perceptron-based models to forecast hourly O3 and NO2 levels in the Bilbao area , 2006, Environ. Model. Softw..

[28]  Hao Zhou,et al.  Modeling and optimization of the NOx emission characteristics of a tangentially fired boiler with artificial neural networks , 2004 .

[29]  Soteris A. Kalogirou,et al.  Artificial intelligence for the modeling and control of combustion processes: a review , 2003 .

[30]  Holger R. Maier,et al.  The effect of internal parameters and geometry on the performance of back-propagation neural networks: an empirical study , 1998 .

[31]  Roberto Baratti,et al.  Monitoring pollutant emissions in a 4.8 MW power plant through neural network , 2002, Neurocomputing.

[32]  Laurene V. Fausett,et al.  Fundamentals Of Neural Networks , 1993 .

[33]  L. Piroddi,et al.  Estimation of NOx Emissions in Thermal Power Plants Using Neural Networks , 2001 .

[34]  Timothy Masters,et al.  Practical neural network recipes in C , 1993 .

[35]  Pierre Borne.  Les réseaux de neurones. , 2006 .