This thesis focuses on computer-related and lighting energy consumption in post-secondary educational institutions. In this respect, artificial intelligence and data association mining are proposed as tools to identify and reduce energy waste. First, an artificial intelligence-based method for forecasting computer usage is proposed. Based on the models' forecast, workstations can be turned on and off, in order to strike a balance between energy savings and user comfort. The models are evaluated on different datasets and their results compared to commercially available alternatives.;Second, a data association mining-based approach is proposed to uncover possible relationships between occupancy patterns and lighting-related energy waste in classrooms. A wireless data collection system is used to log data from both lighting consumption and occupancy states during a year. Next, energy savings results of using the proposed approach are compared to those of an occupancy-activated lighting control system for classrooms. |