Dynamic Simulation and Model-based Control System Design for Comminution Circuits - Advanced Control and Supervision of Mineral Processing Plants - page 215

5.2 Dynamic Simulation of Comminution Circuits Based on
MATLAB ® /Simulink ®

As described in the previous section, in order to model a comminution circuit for
control system design, it is necessary to have a flexible modular simulation environ-
ment. This environment must have the flexibility to configure and blend processes
and control equipment models. There are many simulation tools available on the
market that satisfy these requirements. One of these is MATLAB ® /Simulink ®1 ,as
described in [5]. This tool has many desirable features, such as a modular approach,
an open model structure, ease in changing circuit configurations, a powerful graphi-
cal interface, advanced nonlinear system solvers and control design toolboxes. Sev-
eral authors have recognized these advantages and they have proposed simulators
for comminution operations [5, 6]. This section describes the efforts carried out at
Universidad de Concepcion during the last eight years in the building of a flexible
dynamical simulator for control system design. The simulator includes a library of
equipment and interconnection models which can be blended to configure complex
plants. The library also considers sensor modules that extract information from the
interconnections. This library has been under constant development and refinement
since its conception. Figure 5.1 depicts the modules available to date.
The process modules are interconnected by a flow of material with the following
attributes:

• solid mass flowrate f ;

• water volumetric flowrate (for wet circuits only) q ;

• particle size distribution f ;

• hardness γ.

To change these attributes, there are special mixing modules. For example
MIXWP adds water to a specific flow.
Each model represents the transformation of flow attributes as they advance
through the process equipment. In the following section, main assumptions and
equations considered in the simulator are described.
The notation of this chapter follows these simple rules: lower-case Latin letters
are scalar quantities representing flows, lower-case Latin bold letters are vectors rep-
resenting size distributions, overline lower-case Latin bold letters are vectors repre-
senting size distributions in terms of mass fraction, and bold Latin capital letters
represent matrices associated with equipments. For instance f represents ore feed
flow at n -size intervals and f

∑ i = 1 f i represents total ore feed flow. Thus, the
discrete size distribution associated with f can be written as f

=

= f / f .

1 MATLAB ® and Simulink ® are registered trademarks of The MathWorks, Inc., 3 Apple Hill
Drive, Natick, MA 01760-2098, USA, http://www.mathworks.com