0% found this document useful (0 votes)
33 views

milling-flotation-control-of-gold-circuit-linkedin-article.pdf.pdf

Mintek has developed two control modules, the MillStar Model Predictive Controller and the FloatStar Level Stabiliser, to enhance the stability and recovery of gold milling and flotation circuits at Mandalay Resources’ Bjӧrkdal Gold Mine. The implementation of these systems resulted in significant improvements, including a 28% increase in mill discharge stability and a 37% enhancement in flotation circuit stability, leading to additional monthly revenues of approximately US$ 62,000 and US$ 32,700, respectively. Furthermore, the FloatStar Flow Optimiser has been introduced to optimize flotation mass pull, achieving a preliminary 40% improvement in production rates.
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
33 views

milling-flotation-control-of-gold-circuit-linkedin-article.pdf.pdf

Mintek has developed two control modules, the MillStar Model Predictive Controller and the FloatStar Level Stabiliser, to enhance the stability and recovery of gold milling and flotation circuits at Mandalay Resources’ Bjӧrkdal Gold Mine. The implementation of these systems resulted in significant improvements, including a 28% increase in mill discharge stability and a 37% enhancement in flotation circuit stability, leading to additional monthly revenues of approximately US$ 62,000 and US$ 32,700, respectively. Furthermore, the FloatStar Flow Optimiser has been introduced to optimize flotation mass pull, achieving a preliminary 40% improvement in production rates.
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 3

Control and Optimisation of a Gold Milling and Flotation Circuit

Improving the stability of a minerals processing circuit results in better improved recovery of the valuable
mineral being extracted. The hurdles to achieving stability vary from plant to plant. As an example, the typical
challenges for mill discharge control include: sump integrating dynamics, instrumentation noise, maintaining
certain process variables within limits (as opposed to being controlled to setpoint), interaction between process
variables, and having an unequal number of manipulated and controlled variables. For a flotation circuit the
biggest challenges are disturbances in the feed flowrate and the lack of coupling between control loops of
downstream cells.

Mintek has developed two independent control modules that address the obstacles around stable operation of
a mill discharge and flotation circuit, namely the MillStar Model Predictive Controller (MPC) and the FloatStar
Level Stabiliser. Both these systems were commissioned at Mandalay Resources’ Bjӧrkdal Gold Mine. Once the
desired stability was achieved at this site, flotation optimisation was also implemented on the Rougher and
Scavenger banks. This paper demonstrates the benefits of these advanced systems and compares the results
against regular plant control.

Mill Discharge Stabilisation

Mintek’s MillStar MPC is a versatile tool that controls


a process based on mathematical relationships
between process variables (e.g. level, pressure) and
actuators (e.g. valve position, pump speed). An
objective function is formulated based on the
dynamic response that is desired from the process,
whether it be minimising the movement of certain
actuators, setpoint tracking of specific process
variables, keeping them within limits or a
combination of all.

The model-based controller solves this optimisation problem in order to determine the movements of the
manipulated variables to achieve the desired process response. This formulation enables it to overcome
obstacles such as the interaction between process variables, and having an unequal number of manipulated
and controlled variables.

The mill discharge circuit of Bjӧrkdal Gold


Mine is ideally suited to an MPC solution,
due to the limit handling that was
necessary to accommodate the competing
control objectives. The discharge sumps
are integrated in a recycle through the
gravity spiral separation circuit. Achieving
a stable flowrate and pressure feed to the
hydrocyclones, whilst simultaneously
balancing the volumes between the four
discharge sumps, poses a significant
challenge.
The MPC strategy was implemented with the objective of allowing all sump levels to vary, buffering the change
in volume through the system, so that the flowrate to the hydrocyclones could be kept constant. The controller
is able to prioritise this flowrate stability, while predicting the movement of the level and, pre-emptively, slowly
changes the pump speed if the level is predicted to violate a limit.

Implementation of this control strategy has


improved the stability to the gravity section by
28 % as measured by the minimised flowrate
and pressure feed fluctuations to the
hydrocyclones. This stability enhancement in
the mill discharge - gravity feed section has
resulted in an average recovery improvement of
1 %. This improvement is in spite of lower
average head grade, and the same consistent
feedrate and product grade. With the
assumptions1 listed, this recovery improvement
results in approximately US$ 62,000 additional
revenue per month, which is less than the total
cost of this system.

Flotation Stabilisation
Further processing of the gold ore at Bjӧrkdal includes a Knelson separator which feeds a flotation circuit. Due
to the flushing cycle of the Knelson separator, the flotation circuit intermittently experiences extreme volume
fluctuations. The lack of compensatory control on the flotation circuit resulted in this disturbance causing the
pulp level of all the flotation cells to deviate drastically from setpoint.

Mintek’s FloatStar Level Stabiliser was developed to address the instability


around a flotation circuit, most often caused by varying feed flowrates. The
Level Stabiliser is a model-based, feedforward scheme with stabilising
Proportional-Integral-Derivative (PID) controller that allows the downstream
flotation cells to pre-emptively react to a change in slurry volume that is being
propagated through. This approach has the benefit that disturbances are
mitigated in downstream flotation cells.

The FloatStar Level Stabiliser, with additional


compensation for the Knelson separator’s flushing
cycle, was commissioned on all nine flotation cells at
Bjӧrkdal. This implementation has resulted in a 37 %
improvement in stability of the flotation circuit,
quantified by the reduction in integral absolute error
between the setpoint and measured level values. As
a result of this improved stability and disturbance
mitigation, the flotation circuit recovery has
improved on average by 2.1 %, despite lower
average head grades being processed, as
summarised in the figure to the left.

This improvement results in an additional 35 g of gold produced daily from the flotation circuit. With the
assumptions2 listed, this would equate to US$ 32,700 additional revenue per month. Given the cost of the entire
control system, this improvement alone would allow Bjӧrkdal to realise a return on their investment within only
3 months.
1
Production of 3,600 tonnes/day; head grade of 1.54 g/ton; market value of gold = US$ 40.75/g; concentrate sold at 90 %
of market value
2
Average market value of gold at US$ 40.75/g; concentrate sold at 75% of average market value
Flotation Optimisation
Once process stability of a circuit has been addressed, the opportunity for optimisation can be investigated.
Concentrator plants often process different ores from various sections of their open pit or underground
operations – sometimes changing on a daily basis. Floating or separating physically or chemically different
material requires the operating regions to be adapted continuously. This is most important for the flotation
circuit where the optimal reagent recipe, froth depth and air addition rates are dependent on the type of ore
being processed.

In the absence of online grade analysers, manipulating the reagent recipe is based on visual monitoring of the
froth and an in-depth understanding of the ore characteristics. Therefore, in most cases it is best that this
remains within the scope of the plant metallurgist and operators. The Mintek FloatStar Flow Optimiser is ideal
for finding the optimal operating region for the froth depth and air addition rates to each flotation cell, in order
to achieve a desired mass pull target – all while managing concentrate sump capacities.

Processing of a different ore type that may change on a daily basis has operators and metallurgists at Bjӧrkdal
constantly busy with achieving the correct reagent recipe for the flotation circuit. This means that the
operational staff seldom have the opportunity to consider the adjustment of the froth depth and aeration rates.
The FloatStar Flow Optimiser was therefore implemented on the Rougher-Scavenger section, in order to
optimise the concentrate mass pull from these banks. The strategy uses the flowrate measurement on the
combined concentrate streams to determine the optimal slurry level setpoints and aeration rates for each
individual cell in the two flotation banks, in order to achieve a desired production rate. With this, the plant
personnel are able to focus their attention on the reagent recipe, while the FloatStar Flow Optimiser assists in
working towards the same optimal mass pull target.

Preliminary results indicate that the


implemented strategy has improved
the overall flotation circuit mass pull
by 40 % comparing a 3.5 month
period of ON and OFF data.

Bjӧrkdal is in the process of installing


additional flow meters in order to
expand the successful Mintek
flotation optimisation system to
include other parts of the plant.

This white paper has been brought to you by Mintek. Feel free to pose any questions you might have to the
author at [email protected]. To connect with Mintek Process Control experts on LinkedIn, click here.

You might also like