ABB, the leading power and automation technology group, announced that it has successfully installed
and commissioned its newest version of the Raw Mix Preparation (RMP) solution at the Buzzi Unicem, Guidonia plant in Italy. To improve quality and handle process disruption RMP was installed for raw mix proportioning at the Guidonia plant in June 2006.

Buzzi Unicem chose their Guidonia plant to have ABB‚s RMP solution installed because the plant uses raw materials that have a high variability of chemistry and difficult handling properties.

RMP has successfully overcome these problems by applying a model based on likely causes.  The model detects the gap between what is measured and what is expected and then takes appropriate corrective action to mitigate disturbances and the associated time delays.

Difficulties with the clay supply (related to the clay reclaimer reaching the ends of the bed) have also been successfully overcome. Thanks to the unique capabilities of Model Predictive Control (MPC), RMP is able to anticipate this event and take optimal corrective actions in advance and
afterwards. The resulting benefit is that quality deviations have been reduced without any modifications to the hardware.

The RMP solution is ensuring that the set point for product quality is achieved. Further, the standard deviation has been reduced substantially and is consistently better than the expected 25% reduction that it was first anticipated could be achieved after the benefit analysis. Additionally the system controls the overall flow rate to the buffer silos between the mill and the kiln.  Whenever the mill is running, RMP is online resulting in a run time of 100%. As a result of implementing ABB‚s RMP solution on the raw meal stages of production the plant has increased its product quality and their run times. Reflecting on the RMP installation Mr Luigi Buzzi said ABB’s RMP solution gave extra-flexibility to our Guidonia plant, to overcome difficult and specific raw material handling and dosing obstacles, in good combination with an existing PGNA analyzer".