Global site

ABB's website uses cookies. By staying here you are agreeing to our use of cookies. Learn more

Optimal process operation round-the-clock across 12 process areas at Dyckerhoff cement plant

Share this page

ABB Ability™ Expert Optimizer software achieves stable and optimal process operation at a level that even the best operators are not able to match, 24 hours a day, 7 days a week. Running a more stable process means more stable clinker and cement quality and a need for less fuel and electrical energy.

Key Facts

Customer: Dyckerhoff AG
Country:Germany
Customer need:World energy prices soaring and competition between cement manufacturers remaining strong required decisive steps to ensure advantageous operating conditions
ABB solutions
Customer benefits:Optimal process operation round-the-clock across 12 process areas including raw mill, driers and material blending and optimized costs
Year2009

Ensuring that nothing is left to chance

With world energy prices soaring and competition between cement manufacturers remaining strong, Dyckerhoff AG, a member of the Italian Buzzi Unicem Group, is taking decisive steps to ensure advantageous operating conditions.

Dyckerhoff AG had successfully introduced Expert Optimizer at its Deuna plant in Germany, where positive experiences encouraged the company to install the machinery at Göllheim. 

At the Göllheim plant, Expert Optimizer was installed on two raw mills, two raw material dryers, the raw meal proportioning, two kilns and coolers, and three cement mills at Göllheim. The installation includes the ground-breaking application of model-based state estimation MPC technology to ensure that Göllheim’s kilns and mills can be optimized to the highest level.

ABB’s Expert Optimizer has proved itself worthy for this very large installation with some unique aspects, like unusual configuration of the raw mills and raw material driers

Advanced Process Control (APC) with Model Predictive Control (MPC) technology

MPC is a multivariable control technique. It is based on the ‘receding horizon’ principle where the controller uses a process model to predict process responses to actuator moves and is thus in a position to calculate the best moves for the process evolution over a relevant period.

The MPC controller then sends this set of independent variable moves to the corresponding regulatory (PID) controller, to be implemented in the process as setpoints. It is important to note that a sequence of optimum actions can be calculated while incorporating the dynamics of the system.

MPC technology can, unlike many other controller strategies, explicitly take account of lag and delay times directly in the model. Moreover, it can optimally handle process and actuator constraints.

MPC constructed in a graphical model building toolkit increases understanding and facilitates maintenance

ABB Ability™ Expert Optimizer uses model-based state estimation in the moving horizon approach. Through this, the history of process inputs (fuel, rate, feed, air rates) and of measurements (back end temperature, burning zone temperature, oxygen level) together with the plant model, are processed in a mathematical algorithm in order to estimate magnitudes that cannot be observed directly, like the midkiln temperature. Expert Optimizer adapts the model continuously, and by sending frequent but small set point changes to the controller, stabilises the process.

The models in Expert Optimizer are not black boxes like in ‘standard MPC tools’, but are instead constructed in a graphical model building toolkit. They are a clear representation of the real system with components that have clear interpretation for versed process engineers. This increases understanding and facilitates maintenance. The mathematical complexity is hidden, and only relevant process knowledge is presented

Reaching previously unattainable process stability

Model-based state estimation MPC technology builds on existing MPC methods and experience, but improves on the model by being able to interpret information that would not otherwise exist. In other words, it can recreate, through modelling, a process input value that cannot otherwise be measured.

In this context, the estimation of mid-kiln temperature has already been mentioned, but another important example is estimation of the mill load. Here, Expert Optimizer uses a mill mathematical model, together with the history of fresh feed rate, elevator power consumption and return rates, in order to produce a reliable estimate of the amount of material inside the mill at a given point in time. This additional information means that improvements in process stability can reach levels that were previously unattainable. 
Running such a stable process allows the Göllheim plant to operate closer to its operating constraints than has previously been possible. This means more stable clinker and cement quality and a need for less fuel and electrical energy, which of course means an improvement in operating costs.

Running a more stable process means more stable clinker and cement quality and a need for less fuel and electrical energy, which of course means an improvement in operating costs.

Learn more

  • Contact us

    Submit your inquiry and we will contact you

    Contact us

Popular links

ABB Web feeds