Professional in Training: Metallurgical Engineering

• Jan 2017 - Present

Vacation Student

• Dec 2015 - Jan 2016

A time series analysis of coal ash as measured by online analyzers and laboratory tests was done. An ARIMA model was developed using R GUI. This model was used to predict ash measurements for both testing methods. The model uses the relative density (RD) of the medium at that time to predict what the ash will be one measurement in the future. The model was limited to predicting only one measurement into the future. For the online analyzer, the mean absolute deviation between the predicted and actual ash values was 0.25%. For the laboratory analysis, the mean absolute deviation between the predicted and actual ash values was 0.45%. To improve accuracy the model would need to use mining dispatch data to determine how many previous data points the model should use to predict the next measurements. This system can be improved to simulate changes in quality based on RD shifts. This system should be able to assist with quality control if it runs live in a control room environment. Show less

Vacation Student

• Dec 2014 - Jan 2015

I worked with the metallurgist on the DMS plant. I was involved with two projects. One that focused on reducing input costs in the day to day running of the plant. The other focused on what influence the use of spirals to beneficiate coal fines would have on the overall yield of the plant. I worked with the metallurgist on the DMS plant. I was involved with two projects. One that focused on reducing input costs in the day to day running of the plant. The other focused on what influence the use of spirals to beneficiate coal fines would have on the overall yield of the plant.