Dynamic simulation model of rotary lime kiln

It is often difficult for the operators and researchers to completely understand the complex behavior of the kiln. This is because of the many different phenomena inside the lime kiln and because of the slowness of the process. Dynamic simulation has been adapted as a powerful method for studying transient behavior of the process. When the most important chemical and physical phenomena are modelled using basic conformities of physical and chemical laws, the process can be studied in the computer environment and a better general understanding of the kiln can be gained. The model illustrated in this paper has been developed as a part of a larger simulation environment, APMS (Advanced Paper and Pulp Mill Simulator). The physical and chemical phenomena modeled in this work are water evaporation, calcining reaction, solid movement, gas flow, heat transfer inside the kiln, heat transfer through the wall and mass transfer between solid and gas phases. Also the burner, the draft fan and the automation related to these components have been modeled. The goal in the modeling process was to keep the models of the individual phenomena simple enough in order to achieve fast simulation, which is essential for training purposes. On the other hand, the used mechanistic models describe the process in a detailed way. Another aim was to make the model so general that it could be applied to different kilns. This means that the structure of the model is flexible enough for fast reparametrization. The validation of the model was done using process data from the Metsä-Sellu Äänekoski Mill. There are only a few measurements from the kiln so the comparison of the model with the real process had to be done with limited data. Temperatures in the hot and cold ends of the kiln as well as the outside temperature profiles measured with an infrared camera were the bench marks to compare with. These tests prove that the dynamic behavior of the model is logical and also the quantitative results are accurate in reasonable limits. Both steady state and dynamic results are shown in this paper. Also comparison to measured data is presented. When the simulator is used for operator training purposes a graphical user interface has a significant importance. In this work also a web user interface for the model has been developed. This enables flexible demonstration and training sessions.