KGHM International Ltd.
Copper hackathon resolved - meet the winners of the programming marathon on the occasion of the 60th anniversary of KGHM Polska Miedź SA
The first hackathon organized by KGHM CuValley Hack 2021 has come to an end - nearly 300 participants took part in the competition, who managed to create 53 projects. Participants coded continuously for 40 hours across 3 different task categories. The prize pool in the hackathon was 100,000. PLN.
- KGHM is a tycoon in the mining industry, but most of all it is a technology company. We harness technology in what we do and the more such innovations, especially recently digital ones, the better for us. As KGHM, we are extremely proud that we were able to conduct a hackathon, which was received very positively and with wide interest in the IT world. We thank the participants for their projects. All of them were interesting, therefore the Jury had a difficult task in selecting the presented solutions. However, these are the laws and rules of the hackathon that only the best can win - said Marcin Chludziński, President of KGHM Polska Miedź SA
CuValley Hack participants focused on data analysis and the use of AI, Machine Learning or BigData in industrial automation systems. In practice, the developed solutions are to reduce the failure rate in SMG machines, stabilize the operation of the flash furnace and support shock prediction.
The first edition of the CuValley Hack hackathon was held under the patronage of the Prime Minister Mateusz Morawiecki. Additionally, the winners of the hackathon received an award from the President of Poland, Andrzej Duda, in the form of pens with a dedication to the winners.
KGHM counts on the practical implementation of the awarded projects, including ideas for the optimization of the flash furnace at the Głogów Copper Smelter and the prediction of failures in mining machines. Hackathon CuValley Hack has successfully connected the world of data science with industrial engineering.
Hackathon winners list:
In the SMG MACHINE DATA ANALYSIS category, the best projects were:
1st place
Team: PREDYKCYJNI KRK
Project: Detection of gear failure in SMG.
Jury's justification: Professional work, presenting advanced data analysis from self-propelled mining machines with the potential for application and development in the KGHM environment. Extensive substantive knowledge in the field of machine systems operation guarantees understanding of domain issues and selection of appropriate methods for further analysis.
A few words from the team: As part of the task, an exploratory data analysis was carried out on the basis of which guidelines for data processing into a structure enabling the identification of the variability of selected parameters for two operating states of the machine in relation to specific operating conditions in the daily cycle (normal and pre-emergency, 4-shift system).
2nd place
Team: PBP TEAM
Project: First the sage's glass and eye - then the perceptrons.
Jury's reason: The team presented in the presentation the effect of forecasting and alerting the occurrence of failures. We see the potential of application here and we want to verify the proposed methods for a larger data set.
A few words from the team: The project aims to facilitate quick visual verification of failure detection theory hypotheses, provide an intuitive 'data drill' interface, use unsupervised and supervised learning algorithms to predict failures / identify pre-failures. The developers focused on gearbox failures in WOS machines.
3rd place
Team: CUPRUM INSIGHT
Project: Failure Time Forecast.
Jury's reason: An encouraging solution for the use of prediction of failure of self-propelled mining machines. The obtained high prediction rates made the Jury want to continue cooperation and verify the application of the solution in the production environment in the near future.
A few words from the team: On the example of a gear failure in a haul truck, the authors of the project show that it is possible to build a forecast of the remaining working time using recursive neural networks.
Additional distinction - DATA MINERS
Jury's reason: An interesting idea of the application layer solution and a comprehensive approach to the project, from the model to the application. The jury, appreciating this approach, would like to establish further cooperation with the team.
A few words from the team: MADSztygar is an advanced analytical tool, using solutions in the area of machine learning, time series analysis, feature engineering, in the service of searching for relationships between failures of mining machines and indicators describing these machines. The solution has a user-friendly graphical interface based on RShiny, which facilitates interaction with the modeling results.
In the STABILIZATION OF OPERATION OF THE SUSPENSION OVEN category, the jury selected two best projects. The submitted concepts were assessed in terms of the greatest production implementation potential and the most innovative approach to solving the problem.
1st place
Team: DATA DRIVERS
Design: Furnace stabilizer.
Jury's reason: The first prize was awarded to the DATA DRIVERS team, which presented a project involving a fairly accurate model that produces losses and an optimization algorithm that is quite simple to implement .
A few words from the team: The project consists of three parts: source data processing scripts, a model of total losses of a flash furnace and a flash furnace stabilizer, as well as simulation and visualization of its work.
2nd place
Team: ANZONIA
Project: Optimization of flash furnace operation.
Jury's reason: The second place was awarded to the ANZONIA team for a relatively optimal prediction model and an original, genetic optimization algorithm.
A few words from the team: Based on the regression model, we were able to simulate the operation of the furnace to find the most favorable parameters at the moment. For optimization, we have prepared a genetic algorithm that has the ability to learn and adjust variables over time. With the help of the generic algorithm, the helmsman can simulate and react to the heat balance in a way that is as beneficial as possible.
In the last category PREDICTION OF SHOCK jury selected two winning teams.
1st place
Team: ŚWIEŻAKI
Project: Calm Before the Storm: A Fresh Look at Shock
Jury's reasoning: "A fresh look at tremors in the Rudna mine" according to The Jury of the competition most accurately and most accurately defined the problem of an attempt to predict tremors by analyzing the total amount of energy given up by the rock mass during operation. Of course, the proposed model still needs to be validated, because the effectiveness of the forecast depends on the organization of work and the mining plant's operations.
A few words from the team: In our project, we created a machine learning model for prediction of shocks, which is based on observing the "energy pool" defined by us. We assume that orogenic forces and rock stresses must find their outlet, and that seismic phenomena are natural.
2nd place
Team: KGIS
Project: Prediction of the seasonality of the occurrence of the phenomenon
Jury's reason: The team presented the trends and seasonality of the tremors in the RG, RZ and RP mining areas of the Rudna mine. As in the case of "Świeżaków", it will be crucial to check the algorithm in movement tests.
A few words from the team: Although it is not possible to accurately predict the place, time and energy of the shock, it is possible to narrow the range and provide periods of increased risk, which may translate into improved safety and protection of employees' lives.
Source: https://media.kghm.com/pl/informacje-prasowe/miedziowy-hackathon-rozstrzygniety-poznaj-zwyciezcow-maratonu-programowania-z-okazji-60-lecia-kghm-polska-miedz-s-a
|
Senior Mine Engineer
The Senior Mine Engineer is responsible for the Technical Services Department, including short and mid-range mine planning; mine surveying; ore control; blast hole and stacker sampling.
Senior Environmental Engineer
- Plan, organize, implement, monitor, evaluate and support all environmental projects on the mine.
Senior Benefits Specialist
KGHM International Ltd., a subsidiary of KGHM Polska Miedź S.A., serves as a holding company and an administrative services company for KGHM’s North American assets, which include underground mines and a large underground project in Canada, and two open-pit mines in the United States. KGHM International Ltd produces above 120 million pounds of copper, as well as gold, silver, platinum, palladium, molybdenum, and nickel in North America annually.
Payroll Supervisor
KGHM International Ltd., a subsidiary of KGHM Polska Miedź S.A., serves as a holding company and an administrative services company for KGHM’s North American assets, which include underground mines and a large underground project in Canada, and two open-pit mines in the United States. KGHM International Ltd produces above 120 million pounds of copper, as well as gold, silver, platinum, palladium, molybdenum, and nickel in North America annually.