The mining sector has seen an increase in costs associated with the use of energy in recent decades. Due to lower ore grade, deeper mineralization, or more remote location new mines generally require more energy to produce the same amount of mineral. Mining operations require reliable and cost-effective energy supply, without which extraction becomes economically risky, as well as unsafe for miners. Commercial software and research-oriented computer models are now available to assist in the decision making process regarding the optimal selection of Energy Supply Systems (ESS) and associated costs. However, software and models present limitations:​ some are designed to minimize the cost of supplying only heat and electricity, while others are custom applications for the residential and commercial sectors. Most computer tools assume invariable operating conditions, e.g. energy supply and demand profiles that do not change throughout the lifetime of

the mine, or conditions whose variations can be perfectly predicted. As a result, the optimization of ESS can yield designs that lack robustness to deal with real life, changing environments. Under the same approach, the Optimal Mine Site Energy Supply (OMSES) concept was originally developed as a deterministic mathematical programming tool to find the optimal combination of energy technologies and sources that could meet final energy demands. The solution also included the optimal operation strategy based on typical energy demands of a specific mine site. This thesis expands OMSES to address the robustness of the solution, by considering the uncertainty and variability of real operating conditions. A method is proposed herein, based on the optimal solution obtained by OMSES and utilizing Model Predictive Control (MPC). The

MPC-based simulation under changing environmental conditions ensures that energy demands are met at all times, taking into account energy demands and supply forecast, as well as their inherent variability. Results show that near optimal, more robust design solutions are obtained when the system is simulated under uncertain, more realistic operational conditions, leaving MPC in charge of exploring under-capacity events and of redesigning the system to ensure feasibility with minimum cost increase. This new method has been termed MPC-OMSES dynamic redesign. This thesis also reports on research work to adapt OMSES formulation to account for varying demands throughout the life of the mine, as a consequence of the natural process of mine development and extraction, which means deeper operations over time. This process entails a progressive increase in energy demands, and therefore the energy supply system must be

lanned accordingly. The proposed Long Term OMSES (LTOMSES) shows the advantages of considering an investment plan for the ESS, especially in the case of capital-intensive renewable energy technologies. Other concepts that have been integrated in OMSES and are covered in this thesis include:​ (i) material flows with considerable impact in the energy consumption have been included in the mathematical formulation, in combination with the corresponding technologies, such as pumps, fans and mobile equipment;​ (ii) energy and material storage have been also included, along with complex utility tariff structures, and grid and pipeline extensions. More innovative and integrated solutions can be considered by expanding the feasibility region of the optimization problem, as shown in a case study covering the integration of battery-powered electric underground mobile equipment. Overall, this thesis provides insight and tools to assist engineers in the important task of designing comprehensive and cost-effective energy supply systems for underground mines. Future work suggested includes:​ the development of a methodology to design fully adaptive

ESS (not considering a pre-existing optimal or sub-optimal design);​ the simultaneous optimization of the production plan (ore extracted per day) and the design and operation of the ESS;​ and a dynamic approach to review the investment plan in the face of long-term environmental operating conditions.