Abstract:
As amperages have increased on existing plants and environmental and energy constraints have tightened on both new and existing ones, smelter control has become the biggest leverage point for not only cost reduction and product quality improvement, but also for environmental compliance. The license to operate smelters now is becoming increasingly dependent on the reputation of companies for close control of their operations. This licence is even more important for companies building new smelters. The present article addresses some of the issues smelter practitioners, designers, and engineers face on a daily basis concerning the drivers of smelter process variation and the levers to reduce it. A step change in the understanding and thinking about control is required first at a managerial level, and this change is discussed. Better education concerning process control is essential for engineers and scientists so that more rigorous analysis of plant control defects and failures can occur on a broad front. This is the second important advance required. The third major advance concerns research and development to provide tools for multivariate, non-stationary processes that have extreme observability, controllability, and criticality characteristics. This describes some but not all of the features that are encountered in the control of aluminium smelters. Some promising research directions and applications are discussed.