Heating, Ventilation and Air Conditioning (HVAC) systems generally have high energy requirements, hence there is considerable interest in the development of modelling techniques, optimisation tools and micro-climate control algorithms for buildings. Pertinent to this research area is the Lancaster University Main Campus, for which a central energy centre supplies the hot water used to heat around 50% of the buildings. The Building Management System provides an abundance of energy data for the entire campus, which is further enhanced by the availability of data from the Hazelrigg weather station. This project concerns the development simple, flexible models suitable for improving control system robustness and overall system optimisation. Lancaster's energy centre provides multiple methods of heat production, such as gas boilers and a biomass generator. The models are being used to explore options for hierarchical control, with a focus on optimising the use of the boilers and generator. To achieve this, non-minimal state space model predictive control methods are being adapted for this application. A novelty of the research is the incorporation of weather forecasting and human occupancy data into the control calculations. The research considers the Charles Carter Building as a case study example. This building has a central atrium, surrounded by lecture theatres, offices, meeting rooms and break-out spaces.