When Smart Grid Meets Cloud Computing: Theory, Algorithm Development, and Evaluation

点击翻译按钮获取中文摘要

The smart grid is the next generation electric power grid. It aims to integrate with renewable sources and provide electricity in an efficient, reliable, clean and sustainable manner. Smart grid supports advanced metering infrastructure and real-time bidirectional communications between utility companies and their customers. This facilitates the use of demand side management, which allows the customers to schedule loads from on-peak to off-peak periods. The smart grid also supports the integration of microgrids, which have their own electricity generations and energy storage. For a successful deployment of smart grid, it is important for utility companies to (a) integrate large-scale distributed energy resources to the grid in a seamless manner; (b) develop efficient and reliable grid control and monitoring algorithms; (c) have sufficient computation capabilities and data storage to manage the large amounts of information from the smart meters. In the computer networking domain, cloud computing is the next generation paradigm for computing, information management, and data storage. Cloud computing involves many distributed data centers which are accessible through the Internet. On-demand computing resources can be rapidly provisioned to the customers. It is beneficial for the utility companies to offload some of the computational tasks and data storage to the cloud. However, data centers consume a significant amount of electricity to operate. It is important to develop energy-efficient algorithms to schedule the operations of servers in the data centers. In this proposed research program, we aim, in the long-term, to develop a cloud-based smart grid architecture, which (a) facilitates seamless integration of various types of distributed energy resources; and (b) supports real-time monitoring, forecasting, and control of distributed generators and distributed energy resources at the transmission and distribution level. We will focus on the following inter-related short-term objectives: (i) Design an energy-efficient cloud computing network architecture for smart grid to facilitate access, monitoring, data analytics, real-time control, load forecasting, and data storage; (ii) Develop cloud-based unit commitment, economic dispatch, and demand side management algorithms that can control various types of loads and heterogeneous distributed energy resources connected to the power grid in a scalable and reliable manner; (iii) Develop efficient cloud-based energy trading and pricing algorithms for energy systems and microgrids in different market types. We anticipate that our research will find practical novel solutions and applications for the benefit of the local and the global energy systems and telecommunications industries. They are expected to have a lasting impact in contributing towards the long-term competitiveness of Canada’s technological innovation.