Growing demand for electricity forces suppliers to expand their generation capacity. Financing these expansion programmes results in electricity cost increases above inflation rates. By reducing electricity consumption, additional supply capacity is created at lower costs than the building of conventional power stations. Therefore, there is strong justification to reduce electricity consumption on the supplier and consumer side. The mining and industrial sectors of South Africa consumed approximately 43% of the total electricity supplied by Eskom during 2012. Approximately 10% of this electricity was used to produce compressed air. By reducing the electricity consumption of compressed air systems, operating costs are reduced. In turn this reduces the strain on the South African electricity network. Previous energy saving projects on mine compressed air systems realised savings that were not always sustainable. Savings deteriorated due to, amongst others, rapid employee turnover, improper training, lack of maintenance and system changes. There is therefore a need to improve projects that have already been implemented on mine compressed air systems. The continuous improvement of equipment (such as improved control valves) and the availability of newer technologies can be used to improve existing energy saving strategies. This study provides a solution to reduce the electricity consumption and operating costs of a deep level mine compressed air system. This was achieved by modernising and improving an existing underground compressed air saving strategy. This improvement resulted in a power saving of 1.15 MW; a saving equivalent to an annual cost saving of R4.16 million. It was found that the improved underground compressed air DSM project realised significant additional electrical energy savings. This resulted in ample cost savings to justify the implementation of the project improvements. It is recommended that opportunities to improve existing electrical energy saving projects on surface compressed air systems are investigated. / MIng (Mechanical Engineering), North-West University, Potchefstroom Campus, 2014

Growing demand for electricity forces suppliers to expand their generation capacity. Financing these expansion programmes results in electricity cost increases above inflation rates. By reducing electricity consumption, additional supply capacity is created at lower costs than the building of conventional power stations. Therefore, there is strong justification to reduce electricity consumption on the supplier and consumer side. The mining and industrial sectors of South Africa consumed approximately 43% of the total electricity supplied by Eskom during 2012. Approximately 10% of this electricity was used to produce compressed air. By reducing the electricity consumption of compressed air systems, operating costs are reduced. In turn this reduces the strain on the South African electricity network. Previous energy saving projects on mine compressed air systems realised savings that were not always sustainable. Savings deteriorated due to, amongst others, rapid employee turnover, improper training, lack of maintenance and system changes. There is therefore a need to improve projects that have already been implemented on mine compressed air systems. The continuous improvement of equipment (such as improved control valves) and the availability of newer technologies can be used to improve existing energy saving strategies. This study provides a solution to reduce the electricity consumption and operating costs of a deep level mine compressed air system. This was achieved by modernising and improving an existing underground compressed air saving strategy. This improvement resulted in a power saving of 1.15 MW; a saving equivalent to an annual cost saving of R4.16 million. It was found that the improved underground compressed air DSM project realised significant additional electrical energy savings. This resulted in ample cost savings to justify the implementation of the project improvements. It is recommended that opportunities to improve existing electrical energy saving projects on surface compressed air systems are investigated. / MIng (Mechanical Engineering), North-West University, Potchefstroom Campus, 2014

This dissertation presents a pumping system operation efficiency improvement solution that includes optimal selection and control of the water pump. This solution is formulated based on the performance, operation, equipment and technology (POET) framework. The focus is on the minimization of the operational energy cost. This efficiency improvement solution is divided into three stages in accordance with the operation category of the POET framework. The first stage is to select the optimal pump capacity by considering both energy efficiency and load shifting requirements. The second stage is to develop a flexible pump controlling strategy that combines and balances the contributions from energy efficiency and load shifting. The last stage is to improve the robustness of the control system using the closed-loop model predictive control approach. An optimal pump capacity selection model is formulated. In this model, additional capacity requirements for load shifting are considered along with the traditional energy efficiency requirements. By balancing the contributions from load shifting and energy efficiency, the operational energy cost can be reduced by up to 37%. An optimal pump control is formulated. The objective of this control model is to balance the energy efficiency and load shifting contributions during the operation and minimize the operational energy cost. This control model is tested under different operational conditions and it is compared to other existing control strategies. The simulation and comparison results show that the proposed control strategy achieves the lowest operational energy cost in comparison to other strategies. This optimal pump control model is further modified into the closed-loop model predictive control format to increase the robustness of the control system under operation uncertainties. A mixed integer particle swarm optimization algorithms is employed to solve the optimization problems in this research. AFRIKAANS : Hierdie verhandeling bied ’n verbeterde oplossing vir die operasionele doeltreffendheid van pompstelsels wat die optimale keuse en beheer van die waterpomp insluit. Hierdie oplossing is geformuleer op ’n raamwerk wat werkverrigting, bedryf, toerusting en tegnologie in ag neem. Die oplossing fokus op die vermindering van bedryfsenergie koste. Hierdie oplossing is onderverdeel in drie fases soos bepaal deur die bedryfskategorie gegrond op die bogenoemde raamwerk: Die eerste fase is die keuse van die optimale pompkapasiteit deur beide energiedoeltreffendheid en lasverskuiwing in ag te neem. Die tweede fase is om ’n buigbare pompbeheer strategie te ontwikkel wat ’n goeie balans handhaaf tussen die onderskeie bydraes van energiedoeltreffendheid en lasverskuiwing. Die derde fase is om die stabiliteit van die beheerstelsel te verbeter deur gebruik te maak van ’n geslote-lus beheermodel met voorspellende beheer (Predictive Control). ’n Model vir die keuse van optimale pompkapasiteit is geformuleer. In hierdie model word vereistes vir addisionele pompkapasiteit vir lasverskuiwing sowel as vereistes in terme tradisionele energiedoeltreffendheid in ag geneem. Deur die regte verhouding tussen die onderskeie bydraes van energiedoeltreffendheid en lasverskuiwing te vind kan ’n besparing van tot 37% op die energiekoste verkry word. Optimale pompbeheer is geformuleer. Die doel van die beheermodel is om die bydraes van energiedoeltreffendheid en lasverskuiwing te balanseer en om die bedryfsenergie koste te minimiseer. Hierdie beheermodel is getoets onder verskillende bedryfstoestande en dit is vergelyk met ander bestaande beheerstrategiee. Die simulasie en vergelyking van resultate toon dat die voorgestelde beheerstrategie die laagste bedryfsenergie koste behaal in vergelyking met ander strategiee. Hierdie optimale pomp beheermodel is verder aangepas in ’n geslote beheermodel met voorspellende beheerformaat om die stabiliteit van die beheerstelsel te verbeter onder onsekere bedryfstoestande. ’n Gemende heelgetal partikel swerm optimisasie (Mixed interger particle swarm optimization) algoritme is gebruik om die optimiseringsprobleme op te los tydens hierdie navorsingsoefening. / Dissertation (MEng)--University of Pretoria, 2011. / Electrical, Electronic and Computer Engineering / Unrestricted