A Learning Algorithm For An Intelligent Decision Support System In A Dynamic Mining Environment

The mine environment is dynamic and operating conditions change continually and are difficult to predict. New equipment, variable geological conditions, modified layouts, and different equipment operators make monitoring and control system difficult to apply without continual software maintenance. The paper reports on algorithms that were developed, encoded with Knowledge Craft Tools, and tested on underground coal mine data to adapt to the dynamic environment. The system receives data from sensors in near real time and then identifies patterns in the data. Using a set of mode features the data is automatically checked to determine the present operating mode. When the operating mode changes from the previous shifts, the system automatically learns the new mode characteristics and uses that mode context for further data analysis. The concepts presented in this paper are applicable to a wide variety of mining scenarios, not just underground coal mining.