Pedro DeSouza Bergamo
The training of metallurgical plant staff for optimal operation of plant is currently based on a mixture of classroom teaching and limited hands-on exposure. Limitations of this approach are the slow response of a real plant and the inability to include critical events faced in real processes (e.g. power failure). By contrast, training on Outotec virtual experience (digital model plants) exposes staff to a greater variety of critical situation and the models can be run at up to 5x speed. This project target is the development of dynamic and predictive process models for virtual training in mineral processing plants.
The work involves reviewing existing metallurgical and mineralogical data for case plant, conduct complementary lab studies to add missing data and build full and integrated dynamic models for a plant case. Therefore, these models are verified and validated against training cases in order to design the training program for plant staff. Improved dynamic process model with predictive capability will result in more relevant and realistic training enabling better technology transfer across generations. Provision of a library of training sessions will cover a wide array of real-life scenarios.