| This thesis is dedicated to the study of load models and control of residential appliances. A load model base on computing an average dwelling augmented by a diversification process is introduced. This model is well-suited for studying the load group's natural and controlled dynamics. To validate this model, extensive simulations are employed. Two different strategies are proposed for the control of residential appliances: (1) aperiodic ON/OFF switching of the appliances; (2) continuous control of the thermostat temperature settings.;A mathematical framework based on Pulse Width Modulation theory is developed for the remote ON/OFF switching of residential appliances. Load Management System performs the switching actions on load groups in such a way that the resulting net demand of the appliances matches with a Load Reduction Target. The Load Reduction Target is obtained by modelling load groups as new generation resources into generation scheduling functions. The results of the proposed approach are compared with the results of the current practices and it is shown that the proposed scheme is an effective and flexible method that not only can realize the general form of Load Reduction Targets but also it is capable of error compensation, payback power regulation, switching schedules, incremental update and handling unequal peak powers of load groups.;For continuous control schemes, it is shown that during the control period, diversified states of the appliances do not change drastically and room temperatures within each load group are largely similar. Optimal control strategies are generated by employing Priority List, Linear Programming and Forward Dynamic Programming. The results show a close match between the Load Reduction Target and appliances' demand during control period.;Both methods are efficient and flexible. They allow incremental updating of schedules. Therefore, they are good candidates for use in real time operations. |
