Practical aspects of model predictive control implementation on an industrial lime kiln

Rotary lime kilns are significant energy consumers. In the past, little attention has been paid to kilns resulting in poor efficiency and lime quality. With the adoption of stringent environmental regulations and increased competitiveness, the need for improved lime kiln control has become significant. Lime quality and granular size are primarily determined by the temperature profile along the kiln and the solids residence time. Total reduced sulphur (TRS) and energy consumption can be minimized by maintaining the excess oxygen between constraints. A constrained model predictive control algorithm was implemented on an industrial lime kiln to improve lime quality, decrease energy consumption, reduce maintenance cost and eliminate TRS emissions. The control strategy is very successful in controlling the temperature profile along the kiln while respecting the constraints in excess oxygen and in the manipulated variables. The strategy is also successful in controlling the temperatures during recurrent upset conditions caused by regular washing of the lime mud filter.