With a growing concern on environmental protection and resource conservation, the electric power industry is evolving towards smart grids, empowered by various technologies developed recently including demand side management, energy storage and renewable energy generation. Meanwhile, they are also useful in energy management of smart homes and smart cities. For instance, in-home energy management gives the occupants an intuitive view of the energy consumption and enables smart control of appliances for the purposes of cost saving and security. Smart cities with adoption of electric vehicles (EVs) require intelligent load control strategies to coordinate the charging activities in order to relieve the burden on the power grid. Interesting challenges that are specific to application scena...[ Read more ]

With a growing concern on environmental protection and resource conservation, the electric power industry is evolving towards smart grids, empowered by various technologies developed recently including demand side management, energy storage and renewable energy generation. Meanwhile, they are also useful in energy management of smart homes and smart cities. For instance, in-home energy management gives the occupants an intuitive view of the energy consumption and enables smart control of appliances for the purposes of cost saving and security. Smart cities with adoption of electric vehicles (EVs) require intelligent load control strategies to coordinate the charging activities in order to relieve the burden on the power grid. Interesting challenges that are specific to application scenarios thus occur. This thesis aims to address some issues arising from applying smart grid technologies to the environment of smart homes and smart cities. In particular, this thesis designs a smart home energy management system using non-intrusive load monitoring (NILM), and studies the optimal power dispatch (OPD) of a centralized EV battery charging station (BCS) with renewable generation integration.

To make smart home systems more practical, we design a novel framework of smart home energy management systems incorporating both centralized NILM and decentralized smart control components. It features the capability of automatically mapping the appliances to the corresponding sockets, enabling intuitive human-to-appliance interaction without manual initial setup. Numerical simulations prove the accuracy and efficiency of the proposed framework.

To integrate renewable energy to a centralized EV BCS operating in the battery swapping mode, we develop a two-stage stochastic optimization framework for the OPD to either the conventional or the renewable power system. The conventional power is acquired from both day-ahead and real-time markets to compensate for intermittency of renewable generation. We demonstrate that the OPD minimizes the total expected charging cost with real pricing data.