Part 1 of this article discussed the standard control strategies for minerals thickening.
In part 2 we will discuss advanced control strategies.
Advanced control strategies
Given their importance and criticality in water recovery, the development of expert control systems applied initially in concentrator plants to mills and flotation has now spread to the tailings and concentrate thickeners. Expert control is based on the comprehensive vision of the thickening process, application of fuzzy control logic, implementation of logical rules based on methodologies used by operators and high-level implementation of input and output variable instrumentation.
Fuzzy control logic
Fuzzy control logic is a method of handling and processing information in the same way that a human being does. Thus allowing the management of inaccurate information and uncertainty through diffuse logic. The great advantage of this new methodology is use of expert knowledge in the automation of the tasks to be carried out. Because there is knowledge of the process, it allows it to be modeled using fuzzy logic.
In the expert control system, operating rules are defined to take control of the internal variables (thickener layers) among them, clear water level, mud layer height and rake torque and then maximize the concentration of solids in the thickener discharge and minimize the consumption of the flocculant polymer.
In this control strategy, unlike the PID control, there are no fixed set points, but dynamic ones that change depending on the use of available resources and the gap to optimise. The expert control system applied to the thickeners has been a powerful tool to achieve stable operation in the tailings thickening, despite constant disturbances in the feed, keeping the equipment in a safe operating range.
Standard measurements of density (g/l), volumetric feed flow, particle size, motor current (amps) or rake torque, recovered water volumetric flow and turbidity along with additional continuous measurement of levels in the thickener, are required for expert control.
The additional level measurements in the thickener can be obtained from a SmartDiver. The system provides accurate, real time clarity, interface, and mud level readings. This instrument can also provide the entire density profile of a thickener.
Other measurements that can be included in expert control are recent developed in-line yield stress and/or viscosity (CSIRO). These allow to decouple the thickener control based upon the density and use rheology instead as an additional output variable.
An example of a successful application of an advanced control strategy is the design and implementation of expert control in the Gold Fields, St Ives in Western Australia. Here, a SmartDiver was installed (figure 1) to get the thickener level measurements. Then expert control was implemented for underflow density and flocculant control.
Figure 1. SmartDiver installed on thickener at St Ives
The under-flow density control uses measurements of heavy mud level, rake torque, bed pressure, underflow density, and feed to the CCD/thickener as control and constraint variables with the underflow flow rate as the main manipulated variable.
The flocculant density control uses measurements overflow clarity, interface settling band, and as control and constraint variables with the flocculant addition rate as the main manipulated variable.
The plant’s main objective was to improve overall tank stability producing a consistent underflow density. The objective was met, resulting in a lower variability of the underflow density (figure 2). The flocculant control resulted in less flocculant usage whilst keeping a stable interface reducing flares and ultimately reducing costs associated with overuse of flocculant.
Figure 2. Performance trends during thickener shutdown and start-up.
The traditional PID control strategies are still the most popular strategies used in the industry. However they are limited to optimise because they do not control all involved variables. Thus, the need for advanced strategies that include fuzzy logic and operator expertise.
The current trend in improving the operation and control in thickeners is applying advanced expert control techniques. To implement these systems, measurements are critical. Thus, it is required to populate all points of the circuit with the necessary instrumentation to characterize input and output flows.
Also, to complement the control tasks is necessarily to use modern instrumentation that can monitor the internal layers of the thickener and even the in-line measurement of the yield stress or viscosity of the pulp.
For more information contact our technical team at email@example.com or call us on +61 3 9786 1711