Froth Flotation (Sulphide & Oxide)

Froth Flotation (Sulphide & Oxide)

  • To participate in the 911Metallurgist Forums, be sure to JOINLOGIN
  • Use Add New Topic to ask a New Question/Discussion about Flotation.
  • OR Select a Topic that Interests you.
  • Use Add Reply = to Reply/Participate in a Topic/Discussion (most frequent).
    Using Add Reply allows you to Attach Images or PDF files and provide a more complete input.
  • Use Add Comment = to comment on someone else’s Reply in an already active Topic/Discussion.

Improving Recovery with Smarter Reagent Dosing (9 replies)

Gruppen
8 years ago
Gruppen 8 years ago

We just finished a flotation project and needed to improve recovery and reduce reagent usage if possible. With the sliding price of many metals, this has been a common request lately. The end results were fantastic, with recovery improving by as much as 14%, and reagent usage down by 9%.http://is.gd/ml7p1o

One of the assumed constraints when doing this project was that we must maintain the concentrate grade within the clients’ specification. Part of the challenge is that we must strategically push against the recovery/grade curve so that we can increase the recovery while minimizing the impact to the concentrate grade.

S
Sturmbann
8 years ago
Sturmbann 8 years ago

Impressive results, but is difficult to relate the metallurgical results in your article and the linked webpage to financial results. What was the effect of your system on overall circuit performance (metallurgical and financial)? You mention that the client previously attempted (ineffectively) to optimize each circuit on its own - yet your article does not report on total circuit performance. Lastly, what was the feed condition during the benchmark and test stages? You note that ores from different areas in the pit require different processing settings (hence the need for your system) - unless the feed conditions for the control and test conditions were statistically identical, the results are not very useful. It may be more appropriate to present the results after effects such as feed grade have been accounted for.

Gruppen
8 years ago
Gruppen 8 years ago

The article was intended to display the use of automation in a flotation strategy, and to shed some light on how the rules and limits can be set to have a smooth response without the system reacting too slowly. I think that too often we forget how much the ops team really has to monitor at any given moment, and as a result, how many set points are kept in place well beyond what the feedback from the circuit is suggesting. This seems to be especially true in flotation, and the consequences are amplified in sites that have a lot of ore variation. Each site is always very unique, and so while these results are impressive, each site will require a slightly different approach. The general trend in the histograms is rather rewarding though. In the histograms for reagent addition, you can see that the dosage is much more variable under the expert control, but on average lower, and the grade and recovery less variable, and higher. This is exactly what we want. The operator typically cannot make as many changes, and makes bigger ones. Thus operator control stick with certain set points and the performance varies, while the expert system moves the set points much more, and the performance is more stable and improved.

S
Sturmbann
8 years ago
Sturmbann 8 years ago

I would be interested in seeing the paper you mention. My own interests are very similar to what the article discuss - how much can you automate the decisions that are made by operators today (using readily available process data), and what are the benefits of doing so.

(unknown)
8 years ago
(unknown) 8 years ago

The results are amazing, is really interesting how combining the physical flotation control (air and level) with the chemical aspects like reagent dosage we can get significant results, congratulations for your team!

In my experience implementing flotation optimisation strategies from a process control perspective, the real improvement is reducing concentrate variability and to produce a concentrate grade of a specific and acceptable quality while maximizing recovery. Typically optimizing the cells mass pull and improving reagents dosage can result in a plant performance increment. Many times the systems must be flexible to adapt the mineral market conditions; sometimes you need a better quality and scarify a little in recovery and other time the opposite.

Expert systems in my opinion have proved to be highly applicable to find optimization opportunities in the process every minute compared with the humans who have not the ability to be multi-tasking.

Rahil Khan
8 years ago
Rahil Khan 8 years ago

I think its key to balance Recovery and Concentrate Quality. May you please indicate the grade recovery curves, and also show the quality of the Products.

O
Obersturmbann
8 years ago
Obersturmbann 8 years ago

The work is great. I'm interested in listening to your thoughts regarding fully automated control for the collector dosage in the flotation circuit vs. semi-automated, i.e. sending reminds, guidelines for the operators but let them make the decision? Is there survey or study that has been done on this that I could take a look at?

Gruppen
8 years ago
Gruppen 8 years ago

I'm not sure we've ever tested against the type of semi-automation that you are asking about. Of course, most operators typically have guidelines and subtle reminders that they operate under, but I don't know if we've ever compared to a well established "semi-automated" system like the one you describe. My assumption is that an experience operator, with full attention, and established metallurgical guidelines can mimic what our system does, and because humans are so intelligent, they could very well do it better. The reality is though, that while humans are intelligent, they are forgetful, inconsistent, and typically can't multi task as well as a computer. For this reason, we typically see significant improvements when switching to full automation.

(unknown)
8 years ago
(unknown) 8 years ago

In my experience programming and implementing solutions based on expert system what I always say is that the Expert System reasoning is not better than the human reasoning, the difference is that the expert system has two main characteristics that make the difference: it is consistent and is full time dedicated to control the process. Expert systems need two different sources for its knowledge base: heuristics (from operators and process people), and established knowledge (from the process rule of thumb). Finally an expert system has the ability for explain and feedback the operators about the concluded actions and the conditions supporting it.

In conclusion semi versus full automation strategies has significant differences when the optimization objectives depends of find the suitable conditions at a specific time for improving.

O
Obersturmbann
8 years ago
Obersturmbann 8 years ago

I like your reasoning. Is there any fully automated expert system that is currently in use that you are aware of? I'm interested in this area.

Please join and login to participate and leave a comment.