Date

2017-08-30

Author

Yardımcı, Ahmet Güneş
Karpuz, Celal

Metadata

Show full item record

Item Usage Stats

520
views

0
downloads

Cite This

Majority of traffic in an underground mine flows inside the main haulage way, which is called as the mine ramp. As a navigable haulage path between the portal and production levels, mine ramp is composed of straight and spiral sections with a fixed maximum gradient and minimum turning radius. Most commonly, skilled experts design a so-called satisfactory path based on their professional experience. However, there is no attempt to check whether the design is an optimum solution or not. Considering the initial capital investment and operating cost parameters are directly related with the path, an optimum design minimizing the path length is a necessity. Although, length is the main consideration, there may be other constraints that a ramp is required to satisfy. For example, the path must avoid some undesired underground spaces like aquifers, safety buffers, in order to keep the stability. In addition, some desired regions must be strictly connected by the ramp. Also, rock quality should be considered to drive the ramps inside a higher quality rock zone by decreasing the support cost and increasing the safety and speed of progress. Besides the mentioned constraints some kinematical constraints must be satisfied for the sake of navigability. All these objectives transform the ramp path optimization from a shortest path problem into a least cost estimation problem. This paper presents a genetic algorithm implementation for predicting the near optimum navigable haul road. The algorithm was validated by a benchmark problem and tested on a manually designed real underground mine ramp and improvements were noted.

URI

https://hdl.handle.net/11511/72149

Collections

Department of Mining Engineering, Conference / Seminar