Model Predictive Control For Mining: Process Optimization Solution

Model Predictive
Control for mining

Process optimization solution
BEGIN
Mining powers the world
Minerals and metals play a central role in the
global economy and will continue to provide
the materials the world needs.
THE WORLD IS SEEING INCREDIBLE CHANGES that require resource-intensive
goods of all sorts. Overall population is growing and shifting to urban areas,
the middle class is expanding and infrastructure development is increasing.
With this growth comes a need for resources, meaning metals like iron, copper,
aluminum and nickel will come with a much higher demand.
A new report from World Bank Group states that the production of minerals,
such as lithium, graphite and cobalt, could increase by nearly 500% by 2050, in
order to meet the growing demand for clean energy technologies. The report
estimates that more than 3 billion tons of minerals and metals will be needed to
deploy wind, solar and geothermal power, as well as energy storage, required for
achieving a below 2°C future.
SOURCE: World Bank Group
The global mining market Model predictive Model predictive control

PG 2 control solutions in mining applications 2
PG 5 PG8
Mineral processing challenges
Ore grades are declining and becoming more complex
Today’s process characteristics
Mineral processing plants are inherently complex, facing internal and external disturbances,
many recirculating loads and multiple interactions. And the declining of the ore grades and
continuous variability of raw material are pushing the limits of processing plants even more.
Process objectives
These process characteristics that miners are dealing with today provide opportunity to improve:
• Reduce cost per ton • Continue to meet regulatory

requirements
• Lower plant costs
• Overcome lack of experienced and
• Lower energy intensity
trained instrumentation technicians,
• Improve energy and water consumption operators, metallurgists
• Extend the life of existing ore bodies
How will you achieve these process

objectives in such complex operations?
Solutions to meet The connected Increase Knowledge driven Modernize and Design, build and
The
steelglobal
needsmining market Model predictive
steel plant operationalModel predictive control
operations empower your upgrade with
PG
PG 2
2 PG 5 control solutions
efficiency in mining applications
PG 14 workforce confidence 3
PG 5 PG 9 PG8 PG 19 PG 22
Although traditional optimization strategies provide adequate control
in terms of plant safety, it rarely achieves optimal control of quality and
economical, efficient operation.
AN Why? Mineral processing plants face inherent complexities that can’t be
TH
75%
addressed through traditional technologies:
Multiple constrains
E
MOR
Time delay processes
Multiple variables
of mineral processing Key variables difficult to measure
plants still use basic Other potential technologies, such expert systems, can address some of
optimization strategies these challenges but with price tag:
HIGH MODEL OPERATORS HARD TO

MAINTENANCE (INSTEAD OF PROCESSES) COMPREHEND
The
steelglobal
PG
PG 2
2 PG 5 control solutions
Pavilion8 MPC solution
Model Predictive Control benefits for mining
INCREASED throughput
Handling constraints, maximizing economics, and
maintaining stable process LOWER reagents consumption
Pavilion8® Model Predictive Control (MPC) from Rockwell Automation reduces process BETTER recovery
variability and enhances stability over and above what is currently possible with more
traditional control schemes. This is accomplished through our multi-variable, nonlinear, OPTIMUM use of water and energy
model predictive control capabilities.
IMPROVED process stability
Maximum profitability is realized by operating a process as close to its constraints as
possible while maintaining an appropriate margin of safety.
KEY PRODUCT PROPERTY PERCENT OF LIMIT
100 Process/specification limit

BEFORE MPC ON REDUCES variability
98 ACHIEVES “plant obedience”
96 MANAGES the process within constraints
ACHIEVES UPLIFT - operate closer to

94 specifications and performance limits
while maintaining safety margins
92
Variability under Variability reduction Push process
90 operator control toward limits
TIME
The
steelglobal
PG 2
PG 2 PG 5 control solutions
Pavilion8 Model Predictive Control
architecture overview
Controller / optimizer: Pavilion8 Model Predictive
Control (MPC) optimizes the process making predictions
about future plant outputs responding to changes in the
process input variables and disturbances.
PROCESS
The model: In order to operate properly, we need an HISTORICAL DATA PLANT TEST DATA KNOWLEDGE
accurate model of the process. Pavilion8 MPC leverages

best in class machine learning technologies to build
robust models of the process incorporating and MODELING, IDENTIFICATION, ANALYSIS
combining all available knowledge of the process, such CONSOLE
as historical data, process equations, data from plant VIRTUAL ONLINE ANALYZER CONTROLLER/OPTIMIZER
tests and operator knowledge.
Virtual online analyzer: The efficiency of the

optimization tool is related to the richness and quality
of the field data. However, in most of the process some PLANT CONTROL SYSTEMS
measurements are unprecise, not present, or delayed

(for example data from laboratory). In this instance,
Pavilion8 MPC uses an innovative virtual online analyzer LABORATORY
that’s able to estimate such variables closing the gap of
this information gap.
The console: A powerful visualization tool allowing PROCESS
operators to visualize the process and model KPIs for

better situational awareness of the system.
The
steelglobal
PG 2
CV4
The controller matrix MV1 MV4 DV4 MV5 CV4
MV2
The controller matrix shows dynamic relationships

between process outputs and inputs. The system DV1
predicts future values of the outputs by movement DV2
of all the inputs including manipulated variables and DV3
disturbance variables.
Controlled variables (CVs)

Process variables to maintain at a target or within a
CV1
range (can be considered outputs) CV2
Constrain variable (CCVs)

The state is forbidden to penetrate or may have CV3 MV3
physical limitations
Manipulated variables (MVs) Illustrative MV1 MV2 MV3 MV4 MV5 DV1 DV2 DV3 DV4
example of
A manipulated input is one that can be adjusted by the a flotation Collector Depressor Frother P80 from Feed Feed
Trailings density
control system (or process operator)
reagent reagent Air flow reagent grinding flow from pH
matrix model flow ratio flow ratio value flow ratio circuit grinding from
grinding
Disturbance variables (DVs) CV1 Concentrate

grade
Disturbance variables - these are also called

CV2 Metal
“load” variables and represent input variables that can recovery
cause the controlled variables to deviate

CV3 Froth depth
Bubble
CV4 speed of air
hold up
Bubble size
CV5 distribution
or surface
area
The
steelglobal
PG 2
Upper & lower Upper Lower
Maximize Target
Constraint Constraint Constraint
Crusher circuits
Manipulated Disturbance Controlled Constraint
MV Variable
DV Variable
CV Variable
CCV Variable
PROCESS CHALLENGES MV Speed

From primary
• Balance between the individual crusher
crushers and the overall circuit CV Level

goal Feed bin 1 MV Speed
• Maintain circuit efficiency CV Level
MODEL PREDICTIVE CONTROL

Weight, Level, Speed,
CCV
Weight, Level, Speed, Position, Current
• Ensure that crushers are choked CCV
Temperature, Position, Current
Feed bin 2
• Crusher capacity vs. feed Belt 1 MV Speed
• Equipment wearing Screen 1 MV Speed
2nd cone 3rd cone MV Speed

• Power consumption crusher crusher
MV Speed CV Level
Screen 2 Belt 4
RESULTS CCV T Oil

MV Speed Belt 3
CV Level
• Maximized crusher circuit

CCV
CCV T Oil Position, Current
throughput CCV
Position, Current
• Process stability
Belt 5
• Optimum energy efficiency Belt 2
• Decrease equipment wear
• Fewer trips over current,
high-level silos
The
steelglobal
needsmining market operationalModel predictive
Model predictive
steel plant control
PG 2
PG 2 PG 5 efficiency in mining applications
control solutions PG 14 workforce confidence 8
Comminution /
Maximize Target

MV DV CV CCV
grinding circuits
Variable Variable Variable Variable
PROCESS CHALLENGES CV P80
Flotation feed
• Complex behavior
CCV Flow
• Process disturbances
CCV Pressure
• Relationship between variables is Crusher

MV Speed
Cyclones
non-linear SAG MILL

DV F80 CV Density
Number of DV
• Energy waste due to overgrinding hydrocyclone
Fresh feed MV
• Requires operators with extensive

experience Ball mill
CCV Power
• Balance between throughput and CV Density

CV Weight
downstream constraints such as

optimum particle size
MV Flow CCV Power
MV Flow
Water
RESULTS CCV Density
• Maximized throughput rate Cyclone feed tank
• Decrease energy/ton DV Flow MV Flow
• Increased stability Sump level

CV
• Reduction in particle size variation

• Delivers grinding circuit stability
to achieve throughput and grind POTENTIAL GAINS:
Up to 10% increased throughput, 10%
size targets
on reduced specific energy, with a P80
variability reduction of up to 50%
The
steelglobal
Model predictive
steel plant control
PG 2
Flotation
Maximize Target

MV Variable
DV Variable
CV Variable
CCV Variable
PROCESS CHALLENGES
• Complex process that is not
fully understood
• Still relies on operator control
• A small disturbance in one process CV Bubble speed
variable propagates to the final CV Bubble size
process output

MV Air MV Air
RESULTS
Concentrate CV Concentrate grade
CV pH
• Increased recovery DV Ore Mineralogy MV Frother Mass/metal
CV
recovery
• Increased and stabilized DV Feed flows
CV Cell levels Tailings
cleaner grade DV Ore grades
• Optimized reagent consumption DV P80

MV Valve position MV Valve position
MV Modifier
MV Depressor
MV Collector
POTENTIAL GAINS:
Increased recovery of up to 3%,
reduction of reagents of up to 3%
The
steelglobal
Model predictive
steel plant control
PG 2
Thickeners
Maximize Target

MV Variable
DV Variable
CV Variable
CCV Variable
PROCESS CHALLENGES
• Water conservation
• Optimum reagents consumption
• Little control and local knowledge DV Feed density
• Not often integrated into control DV Feed pH
system DV Feed flow

• Conventional control is MV Flocculant flow
inefficient (long residence time,

large disturbances and non-
linear behavior)
CV Bed pressure
Thickener overflow
RESULTS CCV
Rake arm
torque
DV Rake height
• Increased underflow density CCV Turbidity
• Decreased flocculant
consumption
CV Underflow pump amp
• Improved overflow clarity
• Better water utilization MV Underflow
CV Mud density
POTENTIAL GAINS:
Increased water recovery of up to 3%,
reduction of flocculant of around 2%
The
steelglobal
Model predictive
steel plant control
PG 2
Material flow
management
PROCESS CHALLENGES
• Inconsistent production rates
• Loss of production at shift change
• Conveyor system trips

RESULTS
• Maximizing ore flow
• Operate to maximum constraints
• Minimize equipment trips
• Extending equipment life
• Minimize interruptions during shift
changes
• Energy conservation and
maintaining the maximum
throughput
POTENTIAL GAINS:
20% increased throughput
The
steelglobal
Model predictive
steel plant control
PG 2
Material refining
PROCESS CHALLENGES
• Product under specified grade
• Change in feed characteristics
• Final product variability
• Temperature control

RESULTS
• Stabilization of the operation
• Improved quality
• Product consistency
• Energy savings
• Reduce emissions
The
steelglobal
Model predictive
steel plant control
PG 2
Expanding human possibility
Expandingis a human
registered trademark
possibility, of Rockwell
Pavillion8, andAutomation. Alphabetically
Rockwell Automation are list all Rockwell
trademarks Automation
of Rockwell trademarksInc.
Automation, used in the publication.
Positioning should beAll approximately 0.3” (7.62mm)
other trademarks from above
are property of theirlistrespective
of locations and from below publication number.
companies.
Font: Barlow Condensed, Regular, 8 pt., 80% black, centered.
Publication MIN-BR007B-EN-P January 2021

Publication ENCOMP-QR004I-EN-P - Month 20XX | Supersedes Publication ENCOMP-QR004H-EN-P - Month 20XX
Copyright © 2021 Rockwell Automation, Inc. All Rights Reserved. Printed in USA.
Copyright © 2020 Rockwell Automation, Inc. All Rights Reserved. Printed in USA.
BACK

Model Predictive Control For Mining: Process Optimization Solution

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Model Predictive Control For Mining: Process Optimization Solution

Uploaded by

Copyright:

Available Formats

Model Predictive

Control for mining

SOURCE: World Bank Group

The global mining market Model predictive Model predictive control

• Reduce cost per ton • Continue to meet regulatory

• Extend the life of existing ore bodies

How will you achieve these process

AN Why? Mineral processing plants face inherent complexities that can’t be

Time delay processes

of mineral processing Key variables difficult to measure

optimization strategies these challenges but with price tag:

HIGH MODEL OPERATORS HARD TO

100 Process/specification limit

96 MANAGES the process within constraints

ACHIEVES UPLIFT - operate closer to

accurate model of the process. Pavilion8 MPC leverages

Virtual online analyzer: The efficiency of the

measurements are unprecise, not present, or delayed

The console: A powerful visualization tool allowing PROCESS

operators to visualize the process and model KPIs for

The controller matrix MV1 MV4 DV4 MV5 CV4

The controller matrix shows dynamic relationships

Controlled variables (CVs)

Constrain variable (CCVs)

Disturbance variables (DVs) CV1 Concentrate

Disturbance variables - these are also called

cause the controlled variables to deviate

PROCESS CHALLENGES MV Speed

crushers and the overall circuit CV Level

• Maintain circuit efficiency CV Level

MODEL PREDICTIVE CONTROL

• Equipment wearing Screen 1 MV Speed

2nd cone 3rd cone MV Speed

RESULTS CCV T Oil

• Maximized crusher circuit

Manipulated Disturbance Controlled Constraint

PROCESS CHALLENGES CV P80

• Relationship between variables is Crusher

MODEL PREDICTIVE CONTROL

• Requires operators with extensive

• Balance between throughput and CV Density

downstream constraints such as

• Maximized throughput rate Cyclone feed tank

• Decrease energy/ton DV Flow MV Flow

• Increased stability Sump level

• Reduction in particle size variation

Manipulated Disturbance Controlled Constraint

variable propagates to the final CV Bubble size

MODEL PREDICTIVE CONTROL

cleaner grade DV Ore grades

• Optimized reagent consumption DV P80

Manipulated Disturbance Controlled Constraint

• Not often integrated into control DV Feed pH

system DV Feed flow

MODEL PREDICTIVE CONTROL

inefficient (long residence time,

• Increased underflow density CCV Turbidity

MODEL PREDICTIVE CONTROL

MODEL PREDICTIVE CONTROL

Publication MIN-BR007B-EN-P January 2021

You might also like