Data analytics is one of the largest areas of our activity. However, it is important to distinguish between the analyses related directly to machine diagnostics, and ones that serve other purposes. In this context, DMC undertakes the challenges related to process monitoring, efficiency and reliability analysis, maintenance management and others.
Due to the wide variety of analyzed data types (raw input data as well as intermediate data structures during the analytical process), there is a need to be able to wield a wide variety of data processing techniques. Hence, constant learning and improvement is an unquestionable policy of DMC. Experience gained over the years by our members allows utilizing advanced analytical methods that allow to achieve desired goals.
Regardless of the project or specific solution, there are always critical questions to be answered, with the most important one from our point of view:
How to transform raw data into information interpretable by human user?
The image above presents an example of two different ways that the same portion of raw data (top panel) can be processed and analyzed to obtain high-level information about the process. The technique presented on the bottom left panel utilized the analysis of the map constructed using Anderson-Darling statistic. Related research:
Technical condition change detection using Anderson--Darling statistic approach for LHD machines--engine overheating problem Journal Article In: International Journal of Mining, Reclamation and Environment, vol. 32, no. 6, pp. 392-400, 2018. |
Bottom right panel describes the utilization of short-time data distribution density map. Related research:
Condition monitoring of loading-haulage-dumping machines based on long-term analysis of temperature data Proceedings Article In: pp. 157–164, STEF92 Technology Ltd., 2016, ISSN: 1314-2704. |
Examples of developed solutions
Monitoring of mining machinery has a long tradition at the Faculty of Geoengineering, Mining, and Geology with over 20 years of experience. For years we have been able to provide complete solutions for e.g. conveyor drive monitoring.
Another solution created for the KGHM is DiagManager system designed for the monitoring and preliminary diagnostics of spatially-distributed belt conveyor network.
In recent years we have been working on major projects funded by the EU, such as DISIRE (funded by Horizon2020 framework) or MaMMa (KIC Raw Materials). They focused on the utilization of IoT solutions and cloud technology for the maintenance of horizontal ore transportation process (DISIRE) as well as LHD machines (MaMMa).
Predictive maintenance of mining machines using advanced data analysis system based on the cloud technology Proceedings Article In: pp. 459–470, Springer International Publishing, 2019. |
