1. Renewable integration by means of Smart Buildings
The envisioned future power grid will rely on deep penetration of renewable sources such as wind and solar energies. However, these resources are volatile, intermittent and uncontrollable, which are very challenging to regulate to ensure reliability of the current grid. Buildings are the biggest electricity consumers in the US, who account for 70% of the total consumption. Furthermore, heating, ventilating, air-conditioning (HVAC) and lighting consist of the majority of the electricity consumption. Due to the large energy capacity of buildings as well as automated control feature of HVAC by building energy management system, it is feasible to adjust power consumption without noticeably impacting the building environment. This project investigates the technologies and control algorithms to integrate renewables by control of building energy consumptions.

2. Distributed Control of large-scale Multi-Agent Systems
A typical issue in distributed control is that as the number of agents increases, the performance of the system degrades. A PDE (Partial Differential Equation) model is proposed to study the scaling laws of stability margin and robustness to external disturbances for large-scale vehicular formations. Exact formulae for the stability margin and certain H-inf norm with respect to the number of vehicles are derived. Based on the PDE model, a novel control algorithm is proposed to improve the stability margin and ameliorate the error amplification and disturbance propagation. It is shown that with the proposed design, these performances can be made independent of the size of the formation.

3. Robust Synchronization of Coupled Nonlinear oscillators with application to power networks
The current US electric grid is over one hundred years old, it is operating at its capacity limit and is vulnerable to large disturbances which can lead to power failures. Reasons such as volatility of renewable energy input, seasonal load change and break down of some generators due to overload in the power network will result in loss of synchronization of generator rotors, which leads to widespread electricity blackouts. Thus, there is a strong need to design a robust controller that can ensure synchronization when the system is subject to large disturbances. This project studies a distributed synchronization method for a class of heterogeneous coupled nonlinear oscillators, which is robust in the sense that the network will achieve asymptotic synchronization when subject to arbitrary large but bounded and smooth exogenous disturbances.