Abstract
Earlier work has shown how differences in material transit times from surge units to transfer points in a mine conveyor network enable precise forecasts of load levels that will be intersected by the discharge of a controlled surge unit over future time intervals. Using these forecasts, control policies can be determined that specify the unit's discharge rates over a planning horizon in a manner that prevents peaks in the combined flow at the transfer point from exceeding belt capacity limits while simultaneously insuring adequate unit through-flow to prevent it from becoming a production bottleneck. This paper first extends this control approach to the case where the differences in transit times between surge units are small. It is shown that multiple units should be considered simultaneously when determining control actions, and a dynamic programming approach is given for determining the optimal control policy in such a situation. Subsequently, the application of this approach is considered for longwall mines where flow from the longwall is not controlled but the discharge rates of continuous miner section feeders are adjusted in real time to prevent belt overloading while minimizing any production constraint from the continuous miner units. This application requires special treatment for the highly variable material flow pattern produced by longwalls. The practical utility of these approaches is evaluated through detailed simulation studies.
Original language | English |
---|---|
Pages (from-to) | 2110-2117 |
Number of pages | 8 |
Journal | Conference Record - IAS Annual Meeting (IEEE Industry Applications Society) |
Volume | 3 |
State | Published - 1994 |
Event | Proceedings of the 29th IAS Annual Meeting. Part 3 (of 3) - Denver, CO, USA Duration: Oct 2 1994 → Oct 5 1994 |
ASJC Scopus subject areas
- Control and Systems Engineering
- Industrial and Manufacturing Engineering
- Electrical and Electronic Engineering