A series of tests at Boliden’s Systems Technology division has indicated that artificial intelligence (AI) could unlock further gains from its productivity efforts at the Aitik copper mine, in Sweden.

The company, which partnered up with ABB for these tests, conducted the AI studies to see if technology is available today that could make its concentrators “self-learning,” it said.

The trial took place during the autumn and took a closer look at how AI could be used by Boliden to optimise its concentration processes.

Aitik, meanwhile, is in the middle of an expansion plan that will see production increase from 36 Mt/y to 45 Mt/y of copper ore starting in 2020.

Development Engineer and Project Manager, Johannes Sikström, explained: “At Systems Technology, we develop dynamic simulations of our processes. These simulations can be used in the same way as a game where we define what is a win and what is a loss.

“In the case of self-learning algorithms – so-called deep learning or reinforcement learning – the challenge is the great quantity of data necessary for the algorithm to learn enough about the system for it to make effective decisions.

“This is why games are such a major area within AI research. Games are well suited to enable algorithms to train themselves, and what constitutes a successful result – a win – is also well defined,” he said.

The simulation models enable the company to re-create data equivalent to several decades in just a few hours, according to Boliden.

In its previous projects, Boliden primarily researched machine-learning techniques that analyse data without allowing the algorithm itself to influence it. The aim of the latest project was to allow the algorithm to self-learn instead.

Following initial studies into suitable tools together with Anders Hedlund from data analysis firm BI Nordic, the project led to a degree project in a collaboration involving ABB and Boliden. Max Åstrand from ABB was appointed Supervisor, with his colleague Mattias Hallén taking the lead.

Sikström said: “We directed our attention to the grinding process in Aitik, where we have a well-developed simulation model. We wanted to see if AI was able to do better than our existing control strategy.

“Mattias did a fantastic job setting up the architecture and getting the various environments to ‘play ball’ with each other. We were then able to test various algorithms and different goal functions.”

To begin, Boliden tested a “Q-learning algorithm” which had a goal of trying to control the mill’s load within a given range. After around 40 attempts, the algorithm taught itself to do just that, according to Boliden, acknowledging that it solved the task using a method that would not work in the real world.

In the next step, Boliden investigated the ability of the algorithm to optimise a “gain” instead of optimising a process variable. The goal function for the gain was created as a theoretical model using metal prices, grinding and throughput, for example.

Sikström said: “With this goal function, the AI algorithm succeeded in beating our PID (project initiation documentation) structure to produce a greater gain. So-called wall time was around 80 hours before AI had learned to run the process profitably, in this case equivalent to a plant operating time of more than 300 years.

“The study highlights the value of simulations, and the AI technology shows exciting development opportunities for Boliden’s future process control.”

While the test results were positive, with AI performing better than Boliden’s current control method, Sikström said further studies were necessary before the company considers approaching a viable production solution.

He concluded: “Several technical details need to be resolved, and it is important to use accurate simulation models and well-defined goal functions.

“Because an algorithm is only able to solve the problems formulated for it, process knowhow and experience are at least as important in this type of development as classic process control.”