What Lies Ahead for AI in Flotation? Key Takeaways from Procemin 2023

Process Dynamics and Sustainability Still Threaten Flotation 

Flotation processes are inherently complex, with numerous variables and interactions. A barrier to data-driven discovery is that existing artificial intelligence methods often do not meet the needs of scientific models, especially in mineral processing, they must respect or incorporate physical laws, processes and operational constraints. Simplistic models may fail to represent the intricacies of the process accurately.

In the mining industry, controlling flotation variables and stabilizing column cells have always been challenging aspects of plant operations, particularly for metal recovery. With the increasing pressure towards more efficient and sustainable processes, and the constant increase in the cost of chemical inputs and energy, innovation in the field of real-time process control is under the spotlight.

In order to showcase the advances we have made in this field, we recently presented our methodology for the optimization of flotation at the Procemin GEOMET 2023 conference, in Santiago (Chile).


Mining and AI

Advances in Flotation Modeling Approach

In the realm of flotation, reaching the desired throughput, grade, and recovery objectives presents a big challenge. Operators typically adopt a reactive and manual approach, responding to the visual cues of froth behavior, which becomes more complicated due to frequent fluctuations in feed properties. Furthermore, the lack of real-time circuit visibility and limited feedback on optimal control setpoints pose a significant threat to flotation performance.

Improve Overall Cu Recovery by 1% ($38M)

Considering this aspects, IntelliSense.io studies proposed to analyze flotation behavior within a columnar cell using a method based on neural networks comprising:

1) data measured directly

2) other data calculated with the process of hydrodynamics

3) physics equations phenomena, all embedded in the main neural model. 

By combining the principles of physics phenomena with the power of machine learning using genetic algorithms, we aim to enhance our understanding and prediction capabilities in mining processing. This involves integrating hydrodynamic models with data-driven reasoning.

IntelliSense.io Procemin 2023

Utilizing Artificial Intelligence and Genetic Algorithms for Modeling Flotation Behavior in Columnar Cells

Outcomes, What to Expect?

Metallurgists can start investigating fundamental parameters of Flotation, including bubble surface area flux, bubble diameter, and gas hold-up, observing in almost real-time how step changes in air flow parameters, made by the operator, are having noticeable effects on key fluid mechanic properties in the flotation cells down the bank, and take quick changes if needed. By doing so, it’s possible to increase 1-3% metal recovery and ensure a stable operation.

Recently, we assisted a large Chilean Copper mine to optimize its Flotation Circuit, resulting in a 1% ($38M) improvement in metal recovery.

Interested in learning how we can assist you in improving your Flotation?

Mine to Market: Value Chain Optimization

The Stockpile & Inventory Optimization Application is one of a suite of real-time decision-making applications that uses AI to optimize each process; from mine to market.

Our Material Handling model connects these applications to drive even greater efficiency.

Our process optimization apps can be deployed on a specific process bottleneck or expanded across the entire value chain. They are powered by our Industrial AI Decision Intelligence Platform, brains.app.

Stockpile and Inventory Optimization

Stockpile & Inventory Optimization

Grinding Optimization App

Grinding Optimization

Thickener Circuit Optimization

Thickener Optimization

Flotation Circuit

Flotation Optimization

Solvent Extraction Optimization App

Solvent Extraction (SX) Optimization

Leaching Optimization App

Leaching Optimization