Improved Thickener Performance Through the use of an Extended Kalman Filter

Based on current sedimentation theory an on-line thickener model was developed and coupled with an extended Kalman Filter to obtain additional information such as the compression zone level and concentration, a corrected underflow concentration, the inventory, and some rheological information. The on-line model was installed and tested for robustness for periods up to three days in an industrial environment giving good results. Emphasis was given to signal processing and selecting the appropriate input variables for filtering. The analysis of the thickener from a phenomenlogical point of view provides for better operation, establishes the limiting operating conditions, and assists in the design of control strategies.