0% found this document useful (0 votes)
37 viewsTraining Simulator For Flotation Process Operators - Elsevier Enhanced Reader
ined
Uploaded by
Angel LopezCopyright
© © All Rights Reserved
Available Formats
Download as PDF or read online on Scribd
0% found this document useful (0 votes)
37 viewsTraining Simulator For Flotation Process Operators - Elsevier Enhanced Reader
ined
Uploaded by
Angel LopezCopyright
© © All Rights Reserved
Available Formats
Download as PDF or read online on Scribd
/ 6
Proceedings of the 18th World Congress,
‘The international Federation of Automatic Control
Milano (aly) August 28 - September 2, 2011,
TEAC
Training Simulator for Flotation Process Operators
‘Timo Roine*, Jani Kaartinen, Pertti Lamberg”
‘Aalto University, Department of Automation and Systems Technology
Espoo, Finland (e-mail: timo.roine@tkk fi, jani kaartinent@tkkf)
**Luled University of Technology, Division of Sustainable Process Engineering, Luled, Sweden (e-mail: [email protected])
‘Abstract: This paper presents a novel simulation concept for operator training in the field of mineral
processing, The simulations are carried out with a dynamie process simulator HSC Sim? of HSC
Chemistry developed by Outotec Research Oy. The simulator is fitted to mimic an existing copper
flotation cireuit as accurately as possible by using metallurgical models and then integrated into a larger
mulation environment, providing the operator trainees a realistic experience of the process. The
simulation environment is designed to be scalable and very flexible, allowing many different usage
scenarios and thus aiding in t
transfer of the tacit knowledge from operator generation to the next.
‘Coneurrent work is being done on higher level analysis, uilizing the results reported in this paper.
Keywords: process simulators, operators, process models, training, process automation
1 INTRODUCTION
Mineral flotation is a complex separation process that
typically contains several stages and multiple feedback loops
(e. circulating loads). Also, the reagents that are used vary
and often have opposing effects. This makes the process
difficult to control, atleast in an optimal manner (for further
details about flotation and mineral processing in general, see
eg, Wills and Napier-Munn, 2006, Finch and Dobby, 1990,
and King, 2001), For these reasons, the actions of the process,
operators and differences in their operating behaviour play a
significant role in the performance of the flotation plant.
Training of the operators. in mineral processing has
traditionally been carried out by teaching the basics of the
process to the students and then letting them follow more
experienced operators at work. Due to the increase in
computing power and decrease in prices of computer
hardware, training simulator software is becoming an
important factor in different application areas. This type of
simulation software has been in use, for example, in nuclear
power plants and in aviation for a long time. However,
mineral flotation, the use of such simulators has been limited
This is not to say that simulation has not been utilized in dhe
mineral industry; there are many simulation based
applications. in common use, for example: JKSimMer®
(MeKee and Napier-Munn, 1990), Dynafiag (Desbiens et al.
1997, Flament et al, 1997), JKSimFloat® (Schwarz and
Alexander, 2006), USIM-PAC® (Brochot et al. 2002) HSC
Sim® 7.0 (Outotec, 2006, Roine and Kotiranta, 2007,
Lamberg and Bernal, 2009), but they are typically focused on
aiding in design or control of the process rather than helping
in operator taining. Furthermore, flotation models in these
solutions have typically been empirical and have allowed
only steady-state analysis capabilities
Modelling of the flotation process is very difficult due to the
complex physicochemical reactions and feedbacks of the
process. The micro processes that can be identified in
978-3.902661-93-7111/820.00 © 2011 IFAC
sane
flotation are: 1) particle-air bubble collision, 2) particle
bubble attachment, 3)rise of the bubble, 4) detachment of
particle ftom bubble, 5) froth processes (King, 2001). It is
very challenging to create a physical model even in simple
controlled systems (Miettinen et al, 2010). Flotation,
however, involves chemistry, too. To make minerals
hhydrophobie. ie. floatable, they are treated with collector
chemicals which change the surface of mineral particles.
Chemically these reactions are complex and _ their
measurement in industrial applications is difficult. Finally,
there isa challenge from the complexity of the material. True
flotation plant feeds have wide size distribution, complex.
‘mineralogy and wide range of different liberated, binary and
multi-mineral particles. ‘Therefore, for flow sheet
development and process improvement, empirical and more
practical approaches are used (Runge et al., 1997). Tt is,
‘common to bind all pulp sub processes under a simple kinctie
flotation model and eall this part of the model true flotation
‘The froth processes are combined under the froth recovery
models. The third important component in the empirical
models is to handle water and entrainment. Entrainment is
defined as the unclassified part of sotid material that is
carried by water into the concentrate
To improve the training of process operators, a training
simulator environment has been ereated and is deseribed in
this paper. It consists of 1) flotation process simulations
performed in Chemistry® (Outotec, 2006), 2) process
logic emulation by means of software developed in Matlab®,
and based on Outotec’s Proscon® automation system, and
3) Proficy’HMI Cimplicity® automation software for control
and visualization. In addition, supervisory teacher software
has been developed to manage the student training
environments
Different scenarios can be used in the training simulator to
train inexperienced operators, as well as to improve process
knowledge of senior operators. The eavironmeat can also be
used to collect information of the operator actions and
analyse and compare the performance of different operators.
10.3182/20110828-€:17-1002.02171 418th IFAC World Congress (IFAC'I1)
Milano (Italy) August 28 - September 2, 2011,
In addition, the system and the collected data can serve as a
valuable means to convey important silent knowledge to
following operator generations. Another valuable asset of the
system is that if ean be used to train operators even before the
construction of the actual plant is completed.
Although the HSC Chemistry® simulation software can be
used in a variety of different applications, in this particular
scenario it is used as an integral part of a virtual training
environment in order: 1) to get the trainees acquainted with
metallurgical unit processes, 2) to provide a realistic feel and
response to the changes in metallurgical parameters and to
control actions made by the trainees, and 3) to provide a tool
for improving strategies and scenarios for process control and
development.
‘The training simulator presented in this paper utilizes the
same simulation engine as the Virtual Experience of Outotee
(Moilanen and Lamberg, 2010), but with a completely
different design. Firstly, the simulation model has been fitted
to match the copper flotation cireuit of Inmet Mining
Conporation’s Pyhasalmi mine in central Finland (details in,
Section 2). Secondly, the simulation environment is designed
to be flexible enough to comply with different usage
scenarios, scalable in the number of concurrent simulations,
and distributed so that simulation speed ean be inereased by
running CPU intensive tasks sinmultancously in several
computers. Furthermore, the distributed nature of the system
allows also physical distribution, meaning that teaching can
be done via Infemet. One example of the several possible
setup scenarios is shown in Fig. 1
6 86
é-&
6 86
é»
Fig. 1. Example setup for the training environment.
As it ean be seen, one teacher ean control several simulations
and each of them is realized with two virtual machines
(Smith and Nair, 2005). The virtual machines can be
distributed into one or several physical machines, depending
fon the performance requirements as well as on the
capabilities of the physical machine(s). In this ease, each
simulation is rin in a separate physical machine and an
arbitrary number of students ean conneet to each simulation,
making it possible to have dedicated simulations for each
student, or to let the students work as a team. The structure,
implementation, and communication aspects of the training
environment are eovered in Section 3
‘The basie idea in the training system presented in this paper
is to mimic the operational behaviour of an existing flotation
cireuit as closely as possible and then use the generated
‘model with copies of the existing displays being used in the
plant. This makes it possible for the operator trainees to get
very realistic experience with the simulated process. Another
usage scenario is to run the simulation model in parallel with
the actual process and use it to give foreknowledge of ~ say —
the implications of a given control action. These and few
other usage scenarios are described further in Section 4.
2. SIMULATION MODEL,
Dynamic model of flotation built in HSC Sim® is largely
based on the AMIRA P9 models (Vera, 1999, Zheng et al.,
2006, Welsby et al., 2010), As the P9 models have been
developed for steady-state simulation the dynamie model
uses differential equations with small (1 to 5 seconds)
simulation time steps. The entrainment and froth recovery
models have been adopted from Neethling (2003) and
Neething and Cilliers (2002a, b, 2009). These are described
in more detail later in the text. Flotation cell is divided in two
‘mass balance areas: pulp and froth, Particles flow from pulp
to froth by two mechanisms: true flotation and entrainment.
With current model the entrainment passes directly through,
the fioth into the concentrate. Flux by true flotation is,
‘modelled using first order kinetic equation and the flux from
the froth to the concentrate with the froth recovery model.
Solid material is deseribed as particles, each representing a
particle elass, and having properties like size, composition
and specific gravity. In principle the model is capable to
handle multiphase particles but as a first approximation a
floatability component approach has been used. Fach mineral,
is divided in each size class into three components: fast
floating, slow floating and non-floating — in the Pyhusatmi
cease a total of 75 particle classes (5 minerals x 5 size classes
x 3 components). Liquid phase includes water and reagents:
collector and frother. Collector reacts in the conditioning
stages immediately and resets the mass proportions of
components for each mineral by size class. Frother follows
liquid phase but in a flotation eel itis divided with a fixed
ratio between the froth (concentrate) and liquid tail).
In HSC Sim* the unit model is a DLL file and the main
program takes care of material transport between the units.
Pyhasalmi copper circuit simulation consists of 17 flotation
cells, 2 conditioners, 14 pump/sumps and two on-line
analysers. Delay caused by pipes is currently ignored. In each
flotation unit the calculation within a simulation step goes:
1) take the new input into the cell and mix it totally with the
pulp, 2) calculate the flux of each particle type into the froth
by true flotation, 3) calculate, according to froth recovery, the
flux of each particle class from the froth to the concentrate
and, through drainage, back to the pulp phase, 4) for the
‘current pulp calculate water flux into the concentrate and flux
of each particle class info the concentrate by entrainment,
5) for the remaining pulp calculate the flux of each particle
class and water into the tail according to tailing valve
‘opening, 6) for the remaining pulp caleulate the pulp level in
the cell, 7)caleulate the new value for the tailing valve
‘opening according to pulp level PID control.
s219 418th IFAC World Congress (IFAC'I1)
Milano (Italy) August 28 - September 2, 2011,
2.1 Equations for recovery by true flotation
In the dynamic flotation model the mass flow rate (flux) of a
particle (class) / transferred fiom the pulp to the concentrate
(7, in a time interval (df) is calculated according to equation
a
‘where my is the mass of particle lass jin the pulp and ke is,
the flotation rate constant of particle class / in the collection
zone.
mye= My pk tlt
The flotation rate constant of a particle class i, ky. is a
finetion of floatability of the particle class (P,) and bubble
surface area flux (Sy) (Gorain 1997, 1999)
Hig = PSy 2
‘The bubble surfice area flux S, is related to airflow speed in
the flotation cell (J,, ms) and bubble diameter (ds, Sauter
average, m) as follows Finch and Dobby, 1990)
ly
_
@)
Superficial gas velocity J, (mvs) ean be calculated from the
air flow rate Qyyg (m''s) and cross sectional area of the
flotation cell (A, n*) using equation (Gorain, 1997, 1999)
Que
4
Io @
2.3 Froth recovery
Froth recovery R, , for particle classi, ic, mass proportion of
particles passing through the froth of the true floated ones is:
a nceasley ret
Gey Gey reas
‘where f is the fiaction of attached material that becomes
detached from lost surface area during coalescence, and vis
vertical gas velocity, 7 and re are the bubble sizes in the
bottom and top of the froth, respectively (Neethling. 2008).
2.2 Entrainment
Mass flow of particle class i directly from the pullp through
the froth to the concentrate by entrainment m,~ (i.
unclassified particles coming with water) is caleulated as
Edw ©)
where F; is the degree of entrainment of a particle class ¢ and
Ju is water flux into the concentrate (Savassi et al., 1998).
‘The degree of entrainment is caleulated for each particle class,
using the approach of Neethling & Cilliers 2002a, b, 2009),
) ifac
me
oo( Uisee froth
PM DasarVToe =)
2vhieehproeh
OO Desanlle
”
fae
where U9: is partiele settling velocity, My; oen is froth height,
Dasiat is axial dispersion coefficient and a is air recovery.
Water flux is ealeulated using the formula (Neethling, 2003)
1
ea ifa
