A smart grid is an electrical power grid that is enhanced with communications and networking, computing, control, and signal processing technologies. Although considerable advances have been achieved, the smart grid is still in its infancy stage and many obstacles need to be overcome. For example, it would be risky to open up such a critical infrastructure to allow distributed control. A fault or malicious attack may lead to huge loss in our society. The key to the success of the smart grid lies in developing effective techniques to make it more secure, with respect to detecting anomalies and attacks, and more profitable, with respect to more efficient energy management. To fully harvest its potential, various theoretical tools have been developed and applied to smart grid problems. However, statistical techniques for smart grid research are rather sparse, and have begun to emerge as important and practical tools for smart grid modeling and analysis, but with many challenges. The successful completion of this project will significantly improve the state-of-the-art of the smart grid, provide a significant step forward to fully harvest its potential, and bring the smart grid into practice. The project's education plan includes developing and enhancing various undergraduate and graduate-level courses. Outcomes of this project will be disseminated through technical publications, conference presentations, a project website. Algorithms and software implementations will be disseminated in the public domain as open-source tools. The team is fully committed to promoting participation from under-represented groups in research, and will continue to greatly further such efforts via outreach, in particular, through the NSF REU and RET programs.
This project aims to address the above critical challenges by exploiting new statistics and learning techniques including survey sampling techniques, functional time series models, robust statistics and LASSO based approaches such as online network LASSO and group LASSO. The PIs propose a novel statistics and learning based approach to tackle the several key problems in the modern power grid, i.e., (i) bad data injection detection to make it secure, (ii) load and generation forecasting to enable more effective power management, and (iii) LASSO based approaches to enable cooperative microgrids. The focus is on the modern power system, i.e., the smart grid, with distributed, renewable energy sources and microgrids. In light of the high potential of the smart grid, the goal of this project is to gain a deep understanding of the intrinsic properties and structures, and to develop effective algorithms to make the smart grid more efficient and secure. The proposed research intends to significantly reduce the overhead and improve the energy efficiency of the smart grid with intelligent exploration of advanced statistics and learning theory.
