A Nonlinear Model Based Control Strategy for the Aluminium Electrolysis Process
Important factors for the aluminium industry for succeeding in reducing greenhouse gas emissions and increase energy efficiency is not only the speed in which the organization is able to utilize new knowledge, but also the development and use of new advanced process control systems. New advanced process control systems imply utilizing state of the art process control systems as e.g. Nonlinear Model Predictive Control (NMPC). Although the conventional control structures are dominating the aluminium industry, several authors have addressed advanced process control structures for controlling the Hall-Heroult process. This includes the adaptive control of alumina addition, 9 -Box Matrix Control, LQG Control, Model Predictive Control and control structures involving the Neural network approach. Recently Hydro has been active in developing an NMPC control structure for controlling the Hall-Heroult process. The Hydro NMPC control structure and results from operational practice on Hydro’s HAL275 and Hal4e cells are presented.
KeywordsAluminium electrolysis NMPC Process control Estimation
Unable to display preview. Download preview PDF.
- 1.Grjotheim, K. and Kvande, H. (1993). Introduction to aluminium electrolysis. Aluminum-Verlag.Google Scholar
- 2.Foss, B. and Schei, T. S. (2005). Putting Nonlinear Model Predictive Control into Use. Workshop on Nonlinear Model Predictive Control, Freudenstadt(Germany). IAI, World Aluminium Newsletter, December 2006.Google Scholar
- 6.Rieck, T., Iffert, M., White, P., Rodrigo, R., and Kelchtermans, R. (2003). Increased Current Efficiency and Reduced Energy Consumption at the TRIMET Smelter Essen using 9 Box Matrix Control. Light Metals, pp 449–456.Google Scholar
- 7.Stevens Mc Fadden, F. J., Welch, B. J., and Austin, P. C. (2006). Multivariable Model-Based Control of the Non-Alumina Electrolyte Variables in Aluminum Smelting Cells. Journal Of Metal, 58(2):pp. 42–47.Google Scholar
- 8.Shuiping, Z., Jinhong, L., and Qiangqiang, P. (2008). Model Predictive Control of Superheat for Prebake Aluminum Production Cells. Light Metals, pp. 347–351.Google Scholar
- 9.Meghlaoui, A., Thibault, J., Bui, R. T., Tikasz, L., and Santerre, R. (1998). Predictive control of aluminum electrolytic cells using neural networks. Metallurgical and Materials Transactions, 10(1).Google Scholar
- 10.Kolås, S. (2007). Method and means for controlling an electrolysis cell. Norsk Hydro Patent Application, NO 20075933.Google Scholar
- 11.Drengstig, T. (1997). On process model representation and AlF3 dynamics of aluminium electrolysis cells. Dr.Ing. thesis, Norwegian University of Science and Technology (NTNU), 1997:94Google Scholar
- 13.Saksvikrønning, T., Gran, E., and Vee, K. (1976). Estimation of States in Aluminum Reduction Cells Applying Extended Kaiman Filtering Algorithm Together with a Nonlinear Dynamic Model and Direct Measurements. TMS, AIME-meeting, Las Vegas, USA.Google Scholar