Home Insights Blog Mining and metals blog Advanced thickener control
Mining and metals refining
Jun 13, 2017

Advanced thickener control

In earlier Minerva articles we looked at thickener control from the perspective of measurement and simple reactions to events. To enable better outcomes, control systems are required because in a thickener it is quite difficult to see or measure what is happening and because the reaction time from upset to output change can be long.
Thickener Outotec

The control of the thickener must accommodate fluctuations in the process stream. This control is achieved by manipulating both the flocculant addition and the rate of underflow pumping. However fluctuations in ore type cause random changes in the required flocculant dosing. 

With PID loop control the thickener inventory can be controlled through feedback from either bed level or Clarometer and flocculant dosing can be controlled against mass flow with feedback from bed level. By controlling flocculation and solids inventory, the thickening process is stabilized and consistent, optimum; underflow density can be achieved through manual optimization.

FLOCCULANT DOSING

Flocculant dosing should be matched to the mass flow to the thickener with the rate adjusted to suit the ore being processed. Once the optimum flocculant or coagulant is selected the correct range of dosing can be determined. Simple cylinder tests indicate the effect of dosing on the initial settling rate and give a good guide to the effects on the settling rate and overflow clarity, the effects of flocculants on the underflow properties are less obvious.

For flocculant to be controlled the dosing should be well designed and tailored to the site needs. The location of the injection point should be such that turbulence will ensure a good dispersion of the flocculant with the dilution liquor and feed. Multiple points of addition may be require as this usually improves the contact of the flocculant with the system, coagulants should be added early and where froth is present care should be taken to avoid stabilizing the froth with flocculant. Dilute solutions very often give better results however over dilution will increase the up flow rate overriding the benefits of the enhanced settling rate. 

In most thickeners’ it is possible to check the efficiency of flocculation through sample observation by sampling the flocculated slurry from the feed well.

Principle of advanced thickener control
Figure 1. Principle of advanced thickener control

UNDERFLOW CONTROL

Thickener underflow density is a function of the solids ability to settle and the residence time in the thickener. Systems that attempt to control density from the density measurement tend to cause cyclic instability. By selecting inventory control that controls the underflow pumping to maintain a constant mass the underflow density can be stabilized, then adding underflow density as a feedback to modify the inventory set point consistent high underflow density can be achieved within the limits of the thickener itself. Increasing underflow density results in a material with a higher yield stress being raked increasing the rake torque and limiting the achievable underflow density. In a similar way the depth of the bed is also limited by the thickener geometry.

PHILOSOPHY

Typically, thickener controls are implemented as single loop controllers in DCS/PLC systems. These systems are challenged by long response dynamics and interaction between variables, leading to difficulties in maintaining underflow density. Underflow pumps and flocculant dosing pumps must be manipulated to control the thickener at the desired operation point in varying situations and quality targets. Whilst some classical techniques can handle these situations, regular operator intervention is required. The tuning becomes difficult, typically requiring high-level experts to remain onsite. Another challenge is in allowing for process constraints and the desired prioritization between controlled variables.

Automation Control Tools (ACT) overcomes the inability of Proportional, Integral, Derivative (PID) loops to handle slow response dynamics and cross actions between controlled variables. An appropriate control system architecture is required. Here, the advanced control calculations can take place on a Higher Level ACT system that sits above the site’s process control system (PCS).

Separating advanced control from sequences and the general human machine interface (HMI) on the PCS makes both systems easy to maintain and troubleshoot. Figure 2 shows the results that have been achieved after installing ACT-based advanced control on a concentrate thickener. This compares favorably with PID based approaches that can take many months to fine-tune to an acceptable level of control.

Advanced thickener control
Figure 2. Underflow density, turbidity and bed pressure before and after installation of thickener optimizer.

NICKEL CONCENTRATE EXAMPLE

Nickel concentrate thickener (HRT, 11 m) with 2 underflow lines and UF recirculation

The existing situation included:

  • High variation in incoming mass flow
  • Underflow consistency target not constantly met
  • Overflow dirty - solids load to circulation waters
  • Bed pressure (level) never stabilizes 
  • Lot of underflow recirculation 
  • Lots of operator intervention required

Turbidity and incoming flow rate measurements were missing. The modernization involved the addition of a Software product providing improved controllability of single thickener (Outotec ACT platform).

Controlled variables:

  • Underflow density
  • Bed pressure
  • Overflow turbidity

by adjusting the underflow flow rate and flocculant dosage. Changes in the incoming (mass) flow taken into account as disturbance variable. Rake torque is also monitored to avoid overload situations.

  • Before (red): Manual control, remarkable incoming variation
  • After (green): Thickener optimizer, remarkable incoming variation

After installation of thickener optimizer: 

  • Need for re-circulation reduced 65%
  • Underflow density average + 6% (+0.12 kg/l) 
  • Underflow density variation -24 % 
  • Overflow turbidity overshooting remarkably reduced
  • Overall controller usage high; >94% since the start-up with less operator intervention.
Advanced thickener control
Figure 3. Relative distribution of underflow density before (red) and after (green) installation of thickener optimizer.
Back to Mining and metals refining