Advances in Low Grade Iron

Ore Beneficiation

By
Kamal Kant Jain
Ravindra Kumar Verma
Khalid Razi

Presentation Layout
1. Background
2. Low grade Iron Ore Beneficiation Techniques
3. Current Iron Ore Beneficiation Practices in
India
4. Process Control and Automation
5. Advance Control Tools and Applications
6. New Developments in Iron Ore Beneficiation
7. Conclusion
8. References

1. BACKGROUND

 High grade reserves of Haematite are depleting & the

Indian iron ore mining scenario is changing.
In order to maximise the ore reserve utilization and meet
stringent product quality required by end users industry,
rigorous beneficiation techniques are employed.
If desired quality is not met then after Crushing,
Screening & Classification any one or in combination of
gravity concentration, magnetic separation, flotation,
selective flocculation and pelletisation techniques are
adopted to achieve desired quality

 Now there is trend of integrating geology, mineralogy,

mineral processing and metallurgy to build a spatiallybased model for production management
( GEO-METALLURGY)
Improved technologies for increasing production
efficiency while further reducing water, raw materials and
energy usage is prerequisites for balanced and
sustainable development
Now Iron Ore Beneficiation Plants designed with
advance level of automation and application
softwares

2. LOW GRADE IRON ORE

BENEFICIATION TECHNIQUES

Following principal process technologies/equipments

are available for iron ore beneficiation:
Scrubbers (Attrition & Drum) and Log Washers
Heavy Media Separation & Jig
Teeter Bed Separators (like Flotex density separator, All
flux Separator etc.)
Centrifugal Concentrator, Spirals & Reichert cone
Magnetic Separation (LIMS, MIMS, WHIMS, HGMS &
VPHGMS)
Floatation (Conventional & Column) & Selective
Flocculation
Pelletisation and Roasting

3. CURRENT TRENDS IN INDIA

Major High Capacity Iron Ore Beneficiation in

India Plant are :
Kudremukh, KIOCL (plant now stopped)
(SAG & BALL MILL, LIMS,SPIRALS, FLOTEX
CONV. & COL.)

Barsua, SAIL
(DRUM SCRUBBER,JIG)

Kirandul ,ESSAR
(BALL MILL,SPIRALS,LIMS,HGMS)

DENSITY SEPARATOR,FLOTATION-

 Toranagallu, JSW
(BALL MILL,ATTRITION SCRUBBER,LIMS,SLON/HGMS)

Barbil ,BRPL
(ROD & BALL MILL,ALLFLUX,WHIMS)

Rengali, Bhushan
(DRUM SCRUBBER,JIG,BALL MILL,SPIRALS,LIMS,HGMS)

4. PROCESS CONTROL
&
AUTOMATION

CHANGE IN OBJECTIVE OF
AUTOMATION

CONCENTRATE
TONNAGE
MAXIMISATION

TONNAGE
&
QUALITY AS PER
REQUIREMENT

TONNAGE, QUALITY
&
PEAK ECONOMIC
PERFORMANCE

Hierarchy in Process Control

PLANT
OPTIMISATION
CONTROL

OPTIMISATION

PROCESS OPTIMISATION
CONTROL
DCS/PLC
STABLISATION

INSTRUMENTATION

PROCESS

Overall Process Unit Control

Lower
MULTI
VARIABLE
TECHNIQUE

Knowledge
Based
Expert
Control

Production cost
CROSS
COUPLED
FEED FORWARD
& CASCADE

Higher
Lower

Feed Back

Higher

5. ADVANCE CONTROL TOOLS

&
APPLICATIONS

OBJECTIVE

The commonly used present system is the Distributed Control System (DCS).

It is made up of three main components, the data highway, the operator

station and the microprocessor based controllers.

Shift from maintaining quality to peak performance (often) requires something

more than a DCS / PLC: an optimizing control system

Objective for advanced process control - is to establish a dynamic

mathematical model, monitor the deviation from the model and finally restore
the original optimized conditions of operation.

The process of controlling a dynamic system is complicated especially in iron

ore processing systems where a number of variables are involved
simultaneously

SYSTEM
Intelligent control, including ES and fuzzy logic
Model predictive control, using linear or non-linear models
originating in phenomenological
or empirical models
adjusted on the basis of operating data
Attempts have been made to combine them into a single
integrated solution (Hybrid) , with the algorithms known as
fuzzy model predictive control

Intelligent Control
Expert systems (ES) integrate the knowledge of one or more
process specialists into a set of rules or a knowledge base
that defines the actions of an expert controller who acts
similarly to a proportional (P), proportionalintegral (PI) or
proportionalintegralderivative (PID) automatic control
algorithm
One of the most frequently adopted alternatives for improving
the robustness of expert control systems in handling
uncertainties and errors is fuzzy logic.
The most commonly used membership functions are
triangular, trapezoidal or Gaussian
Expert systems that incorporate fuzzy logic into processing
rules are known as fuzzy ES.

Intelligent Control..contd.
Notable among the non-linear models are neural
networks, which are used to numerically approximate a
highly complex non-linear function by interconnecting
simpler processing elements such as adders, multipliers
and sigmoid functions.
As with linear time series models, the neural model must
be calibrated by adjusting its parameters to the operating
data, a task generally performed by a back propagation
gradient algorithm.
Genetic Algorithm is also one alternative in Intelligent
control

Model Predictive Control

Model predictive control (MPC) embraces a complete family of
controllers whose basic concepts are:
Use of an explicit dynamic model (predicts process outputs at
discrete future time instants over a prediction horizon)
Computation of a sequence of future control actions through the
optimization of an objective function with given operating
constraints and desired reference trajectories for process outputs
Repetition of the optimization process at each sampling instant
and application of the first value of the calculated control
sequence (receding horizon strategy)
Above three characteristics allow MPC to handle multivariable,
non-minimum phase, open-loop unstable and non-linear
processes with a long time delay and the inclusion, if necessary,
of constraints for manipulated and/or controlled variables.

MPC SUPPLIERS

ABB -Expert Optimizer

Andritz- BrainWave6
Emerson Process Management- Delta V
Honeywell -Profit7 Suite
Invensys -Connoisseur
Metso Minerals -Optimizing Control System
Mintek -StarCS
Rockwell -Pavilion Technologies
SGS -MinnovEX Expert Technology

6. NEW DEVELOPMENTS IN
IRON ORE BENEFICIATION

New developments and products are in in the following

areas:
Visual sensors with greater accuracy and robustness.
On-line hardness and mineralogy analyzers
New sensors for the measurement of grinding,
classification and flotation variables
Additional new tools for advanced control that combine
expert system (ES) with model-based control and
continuous with discrete control (hybrid systems)
Dynamic optimization applications for integrated
processes and plant interconnection

7. CONCLUSION

 Use of latest beneficiation techniques in iron ore

industry in India has immense scope to cater the
burgeoning demand of steel industry.
Region specific integrated approach is to be
developed to prepare the ore characterization
database and standardization of beneficiation
technology.
Robust Design and model based optimized
mineral beneficiation techniques would be key
enlightener for decision making to choose the
right path for beneficiation of low grade iron ore.

 The optimization based on models predictive control

(MPC) is widely applicable state of art feature of
advance beneficiation technology. It is economically
viable & getting importance globally as well in India
Combined expert system (ES) with model-based control
and discrete control (hybrid systems) hold good future in
Low Grade Iron ore beneficiation plants
Training of professionals and technicians charged with
design, supervising and operating mineral processing
plant and automation equipments is essential element
for robust design and optimsed operation of iron ore
beneficiation plant especially for low grade ore utilization.

8. REFERENCES

Web site www.steel.gov.in

Burt, R.O. & Mills, C. (1984), Gravity Concentration Technology, Advances in
Mineral Processing Series, Volume 5, Elsevier, and Amsterdam
J. Lynch (January 1977), Mineral Crushing and Grinding Circuits: Their
Simulation, Optimization, Design and Control, Elsevier Scientific
Napier-Munn TJ,Morrell S,Morrison RD, Kolovic T (1996),Mineral Comminution
Circuits:Their Operation and Optimisation. JKMRC, University of Queensland,
Brisbane
Mular AL, Barratt DJ, Halbe DN (2002) Mineral Processing Plant Design,
Practice, and Control (2-volume set). Society for Mining Metallurgy &
Exploration, New York
Barry A. Wills, Tim Napier-Munn,(2006), An Introduction to the Practical Aspects
of Ore Treatment and Mineral Recovery, Elsevier Science & Technology Books
A.Gupta and D.S.Yan, (2006), Introduction to Mineral Processing Design and
Operation
Daniel Sbrbaro Ren del Villar (2010), Advanced Control and Supervision of
Mineral Processing Plants, Springer-Verlag London Limited

THANKS

Back up
slides

AUTOMATION BENEFITS

Increased production
Process stability improvements
Better use of raw materials
Reduced maintenance and improved safety
Improved process knowledge

LEVEL OF AUTOMATION
Level 1- Basic block
Level 2-Supervision block
Level 3-High level block
Level 4- "Watch dog"

AUTOMATION LEVEL

Level 1
This is the regulatory level where basic controls loops like, P+I control loops
include control of feed tonnages from bins, conveyors, manipulating of bins,
water addition loop (in milling circuit) pump speed and sump level controls,
thickener overflow density control etc are involved (depending on the process
circuit).
Level 2
This is a supervisory control stage that includes process stabilization and
optimizing, usually using cascade loop and ratio loops. For example, in a ball
mill circuit the ratio loop controls the ball mill water while the cascade loop
controls the particle size of product by manipulating the tonnage set point.
Level 3
Controls at this level include maximizing circuit throughput, limiting circulating
load (where applicable).
Level 4
This is a higher degree of supervisory controls of various operations including
plant shut downs for maintenance or emergency. It has been referred to level 4
controls as "watchdog" control.

LIST OF VARIABLES
Variety of process variables are measured by Sensors :
Feeder frequency, conveyor load and crusher chute level
(crushing)
Tonnage, water flow rate, mill speed, pulp level, pump speed,
pulp volumetric Flow rate, pulp density, cyclone and mill
pressure for Screens, pumps and cyclones (grinding), the
power draw of mills
Pulp flow rate, cell and column pulp levels, air flow rate,
reagent flow rate, wash Water flow rate and pH (flotation).
Pulp particle size distribution sensors in grinding and grade
analyzers in flotation/Gravity/Magnetic Separators

Mathematical Models
Models can be used to improve efficiency and
sustainability of mineral processing in many ways. They
can be used, for example, in process research and
development, design, optimization and control.
If the model is time-dependent, it is dynamic while static
(or steady-state) models do not depend on time.
Dynamic models are typically represented as differential
(or difference) equations.
Mechanistic models are based on the actual or
assumed mechanisms of studied phenomenon while
empirical models are based on observations.
Simulations based on either mechanistic or data-based
models operating in steady-state or dynamic conditions
have also been used commonly in mineral industry.

Model Predictive Control..contd.

The control sequence is obtained by optimizing an objective
function that describes the goals the control strategy is
intended to achieve.
In classical MPC, an objective function minimizes the error
between predicted outputs and the set-points during the
prediction horizon as well as the control effort during the
control horizon.
The function may be expressed as The optimization process
may involve hard or soft constraints.
For linear unconstrained systems this optimization problem is
tractable and convex and can be solved analytically, but in
general applications it is common to take into account
constraints or non-linearities in the process, and in such
cases the optimization problem must be solved using iterative
numerical methods.

Model Predictive Control..contd.

A fundamental element in MPC is the model used to
characterize the dynamic behavior of the process.
The origins and formulations of such models are diverse, but
may be classified as follows:
Phenomenological or first principle models, in the vast
majority of cases nonlinear and continuous time
Models obtained through numerical adjustments based on
operating data using discrete time series, either linear or
non-linear
Model parameters are obtained mainly by two methods :
Regression
Curve fitting method

MPC APPLICATION
In recent years, the application of MPC to hybrid
dynamic systems has emerged as a significant area of
research.
In these systems, continuous dynamic sub processes
interact with discrete event detection elements and
start/stop commands .
Characterizing this type of system involves combining
continuous with discrete variables and differential or
difference equations with finite state automata or
switching theory.
Although this approach increases the complexity of the
model, its potential for accurately capturing the dynamic
of an industrial process is much greater.