Typical deep level mines use up to 27 ML water per day for mining operations. Multistage centrifugal pumps up to 2500 MW are used in an upward cascading manor to dewater the shaft. The dewatering systems at some mines are automated to enable surface control. Automation of the pumps is typically based on the best practice procedure known when implemented. Best practice procedures are used to ensure safe pumping operations. It was found that pump failures could still occur even with the best practice implemented.
Unexpected failures of pumps are of major concern because they can result in the flooding of a mine. Flooding increases the risk of environmental damage and injury to the mining personnel. An additional concern is the maintenance cost of multistage centrifugal pumps. Overhaul cost of a seized multistage centrifugal pump is almost R1-million.
The aim of this study was to improve established best practice procedures for pump automation. This could be achieved by investigating the general root cause of failures of automated pumps. Additional instrumentation and protection devices to prevent similar incidents were examined. Revised system control parameters were developed to ensure that the pumps operated within the design specifications.
The improved best practices proved to prevent failures as a result of overheating and cavitation. Increasing the pump reliability and availability enabled surface control. The control of the automated dewatering system realised an electricity cost saving of R6-million. The automated system also made it possible to calculate the real-time pump efficiency within 5%.
Previous best practice procedure was found to be inadequate to prevent all possibilities of failure. Additional precaution measurements were added to prevent pump failure. / MIng (Mechanical Engineering), North-West University, Potchefstroom Campus, 2015
Identifer | oai:union.ndltd.org:NWUBOLOKA1/oai:dspace.nwu.ac.za:10394/15483 |
Date | January 2015 |
Creators | Oberholzer, Phillip Johannes |
Source Sets | North-West University |
Language | English |
Detected Language | English |
Type | Thesis |
Page generated in 0.0024 seconds