Cement finish grinding consumed the most electrical power in cement industry. The total power consumption however depends on the fineness of the grind, the size distribution and the efficiency of separation of the finely ground particles. Such high power consumption means that improving the efficiency of grinding and separation of the ground clinker can yield significant improvement in energy efficiency and cost saving.
Cement finish grinding mills are usually done in two or three compartment ball mill in series connected by a diaphragm. This thesis describes the development of a mathematical model that can realistically describe the behavior of the two-compartment ball mill in cement finish grinding. Based on previous study, efficiency curves were used to model the high efficiency separator. Modelling approach conducted in the present study involved extensive data from laboratory and full-scale plant. Industrial surveys were done to obtain data in the two-compartment ball mill around open and closed circuits. These data provide important information and understanding on how the mill behaves and help in constructing good models.
Due to lack of study being conducted to understand the material transport in many types of mill, a simple experiment using grate- discharge mill has been designed to study the factors affecting material transportation in the two-compartment mill. Due to its simplicity, discharge rate function was used to describe the material transport function in the mill. It was found that the discharge rate depended on particle size (finer material tended to move faster), ball load and mill load. Results form this experiment was used as a basis in describing the material transport function in the development of the two-compartment ball mill model developed.
The diaphragm effect in the mill has become important recently and its effect has been included in the present model. Based on extensive data before and after diaphragm, it was found that the diaphragm can be modelled using two-stage classification function. The classification function which could be described by two maximum sizes and grate size were found to work very well.
The breakage distribution function was also compared between single size fraction technique and drop weight test method to obtain its suitability in the model. Through detail analysis, it was found that there was a significant amount of low energy breakage within the mill. It was therefore decided to use combine breakage function consist of impact (high energy) and abrasion (low energy breakage) as being used in SAG/AG mill model.
The newly developed approach considered the two-compartment mill as perfectly mixed slices in series. In this approach, instead of using lumped parameter (breakage and discharge rate function being combined together as r/d) as being used in Whiten's perfect mixing ball mill model, an individual function was used. The breakage function was assumed to be constant for each slice. The cubic spline method was used to calculate the breakage rate function from laboratory mill. Due to the simple modelling approach as described in this thesis, it was found that the approach was able to be used in modelling the two-compartment mill in open or closed circuit.