Energy management strategies for residential buildings under demand response

The evolution of the existing power grid from a conventional and unidirectional structure into an automation\&optimization-based and bi-directional structure, defined as the smart grid (SG), may serve to the goals of conservation, efficient utilization, optimal management of energy at the end-user side. SG simply refers to the smarter, more efficient, and augmented communication facility-based operation of the existing power grid. SG exploits some means to attain its goals. One of the major means of SG is demand response (DR) programs. DR programs aim to incorporate the integration of end-users into power system operation, and residential end-users have the greatest potential among other end-user types from DR perspective since their electrical energy consumption can be flexibly managed by automation\&optimization-based residential energy management systems (REMs). This thesis, therefore, investigates the energy management strategies for residential end-users under DR programs. To this end, cost and comfort-efficient solutions are developed for the various kinds of appliances in residential buildings. As the first study in this thesis, an energy management strategy for heating, ventilation, and air conditioning (HVAC) loads which have a substantial portion in the energy demand of a residential building is developed. This strategy is based on the adjustment of the thermostat setting of air conditioners (ACs) in a residential building community using end-user thermal flexibility levels, time-varying energy prices, and outdoor weather conditions. The results of this study reveal that end-users can acquire cost and energy savings, while the peak load of the power grid can be mitigated through the proposed strategy. On the other hand, it is shown that end-users can improve their cost savings if they agree to be more thermal flexible, through a fair cost-saving allocation strategy. As the second study in this thesis, a cost and comfort-efficient utilization of HVAC loads from a multi-objective optimization framework integrated with a residential microgrid scheme are discussed based on the proposed thermostat setting strategy in the previous study. To this end, the optimization problem is decomposed through a decoupling process, and an algorithm to cost-efficiently allocate the available power sources (i.e., solar, wind, and battery storage) is proposed. The results of this study show that HVAC loads and a residential microgrid can be managed in such a way that there remains no requirement for the prediction of daylong energy prices. On the other hand, the proposed methodology indicates that the peak-to-average ratio of the grid can be improved through the implementation of a bi-directional DR program. The third study in this thesis is conducted to extend the multi-objective optimization proposed for HVAC loads to other appliances in a residential building. In this regard, a REMS is designed to comprehensively obtain DR and relevant discomfort models of various types of household appliances. The motivation of the designed REMS is to enable residential end-users to make an explicit cost-comfort analysis of REMSs by observing how much a comfort loss occurs in return for cost-saving. The results demonstrate that end-users can get accurate knowledge of their cost and comfort balance under different DR programs. The results of this study also reveal that electrical energy prices that change at shorter regular intervals produce prominent results in reducing the cost of end-users, while the electrical energy prices that change at longer time intervals yield better results in relieving the discomfort of end-users. All the energy management strategies proposed in this thesis focus on developing cost and comfort-effective solutions to household appliances, thus to the household load demand, under various pricing tariffs and operating conditions. Through the results, it is shown that efficient usage of electrical energy inside residential buildings produces considerable benefits from both end-user and grid perspectives.