Przewiń w górę
We know the copper winners of the 2nd edition of the KGHM Hackathon

We know the copper winners of the 2nd edition of the KGHM Hackathon

Monday, 14 March, 2022
250 participants, 40 projects and countless hours of data analysis - the second Hackathon CuValley Hack 2022 organised by KGHM Polska Miedź S.A. and KGHM Analysis Centre has come to an end. 7 ideas were awarded. The prize pool for the winners amounted to PLN 120 thousand.

The hackathon was held in an online format, during the event many platforms known to programmers were used. The event was accompanied by a technology conference "Copper Valley" organised in the Library of the University of Zielona Góra. Experts presented participants with the latest solutions and examples of technology applications in industry.

The hackathon attracted 250 participants who created 60 teams. They continuously coded for over 40 hours to create 40 projects under 3 different task categories. The final teams took part in live sessions during which they answered questions from the jury about the proposed solutions.


List of hackathon winners:

In the category Preparation of a data model to determine cathode quality based on feedstock and process parameters, the best projects were:

1st place

Team: CUdHacki

Project: Interactive tool for cathode composition analysis

A few words from the team: Our tool can be used to select the composition of anodes before starting the process, as well as to correct parameters already during electrorefining. In the static part, we aggregated and analysed the available data in terms of correlations. In the dynamic part, we went one step further and created a model that, for the process input values selected by the operator, allows predicting the amount of white cathodes produced, as well as their composition. We then added another option that allows optimising input parameters to obtain the best quality cathodes.

Jury motivation: The winners have managed to combine an understanding of the technological process, data analysis and come up with a very interesting solution in the form of optimised parameter selection for the highest possible quality. A project with definite implementation potential.

2nd place

Team: Anovei

Project: Model for the prediction of cathode quality

A few words from the team: The project provides process parameters for a given chemical analysis of the anode in order to obtain a high quality cathode. A genetic algorithm and a custom recurrent network model. You can select which process parameters can be changed by applying a mask to the genes in the genetic algorithm.

Special award of the President of KGHM Polska Miedź S.A.


Project: White Crow (Biały Cruk)

A few words from the team: We have developed an application that, based on the chemical composition of the anodes, proposes technological parameters for the process to obtain 100% quality cathodes. A technology demo was produced as part of the project.

Jury's reasoning: The team was basically on par with second place in this category. It was hard to make a final decision. The team, however, competed in two tasks and were just a hair behind the winners in the next one, so we decided to reward them with a special prize for their diligence and creativity.


In the category Energy Optimisation of the Pump Unit of the Suspension Furnace Cooling System, the Jury selected the 2 best projects.

1st place

Team: Digital Twins

Project: Digital Twin of the Pump Unit

A few words from the team: As part of the task, a Digital Twin of the Pump Unit was created on the basis of catalogue data and available measurements, allowing the analysis of the system operation for different settings of the controlled parameters.

Justification from the jury: A clear winner in this task. They suggested not only improving the quality and efficiency of the current system but also identified possible future upgrades. The direction is also important - a digital twin for circuits or entire industrial systems is our target development direction in this area.

2nd place

Team: Copperheads

Project: CoolerOptymizer

A few words from the team: We have developed an application that, based on input data in CSV format and a diagram model, allows simulation of the operation of this system hour by hour. The model is flexible, with minor changes it allows modelling of any cooling system consisting of 2 types of pumps.


In the last category Artificial temperature analyser for slag inside the Suspension Furnace of Głogów smelter I and 2 winning teams:

1st place

Team: Miedziaki

Project: An eye on copper

A few words from the team: "An eye on copper" is a predictive model that has been based on a machine learning model, specifically Gradient Boosting.  This technique combines innovative technology based on the achievements of mathematical branches such as analysis, algebra and probability calculus with the simplicity of a decision tree.

Jury's reasoning: Once again a clear winner in its category. What makes us even happier is that there were 3 women in this team of 4. The team is made up of students from the Poznan University of Technology, who have demonstrated a very good understanding of industrial and technological processes. The algorithms used with the "clean coding" philosophy further convinced us. The model created was also characterised by the highest likelihood parameter and the lowest deviations.

2nd place

Team: Cuprum Insight

Project: Virtual measuring lance

A few words from the team: The aim of the project was to create a virtual measuring lance using artificial intelligence, with which the temperature of slag is determined with a 1 minute interval.

The second edition of the Hackathon CuValley Hack 2022, just like the first one, was under the honorary patronage of Prime Minister Mateusz Morawiecki.

Link to the broadcast of the hackathon conclusion and announcement of winners:

In order to make browsing this website more comfortable for our users, mainly to adapt the website's layout to their preferences, browser type and preferred language, we reserve the right to collect additional data (sent by the browser as standard) and use the so-called cookie files. We can use such data to analyse content attractiveness and for other statistical purposes.