This project designs algorithms for the integration of plug-in hybrid electric vehicles (PEVs) into the power grid. Specifically, the project will formulate and solve optimization problems critical to various entities in the PEV ecosystem -- PEV owners, commercial charging station owners, aggregators, and distribution companies -- at the distribution / retail level. Charging at both commercial charging stations and at residences will be considered, for both the case when PEVs only function as loads, and the case when they can also function as sources, equipped with vehicle-to-home (V2H) or vehicle-to-grid (V2G) energy reinjection capability. The focus of the project is on distributed decision making by various individual players to achieve analytical system-level performance guarantees. Electrification of the transportation market offers revenue growth for utility companies and automobile manufacturers, lower operational costs for consumers, and benefits to the environment. By addressing problems that will arise as PEVs impose extra load on the grid, and by solving challenges that currently impede the use of PEVs as distributed storage resources, this research will directly impact the society. The design principles gained will also be applicable to other cyber-physical infrastructural systems. A close collaboration with industrial partners will ground the research in real problems and ensure quick dissemination of results to the marketplace. A strong educational component will integrate the proposed research into the classroom to allow better training of both undergraduate and graduate students.

Inadequate system understanding and inadequate situational awareness have caused large-scale power outages in the past. With the increased reliance on variable energy supply sources, system understanding and situational awareness of a complex energy system become more challenging. This project leverages the power of big data analytics to directly improve system understanding and situational awareness. The research provides the methodology for detecting anomalous events in real-time, and therefore allow control centers to take appropriate control actions before minor events develop into major blackouts. The significance for the society and for the power industry is profound. Energy providers will be able to prevent large-scale power outages and reduce revenue losses, and customers will benefit from reliable energy delivery with service guarantees. Students, including women and underrepresented groups, will be trained for the future workforce in this area. The project includes four major thrusts: 1) real-time anomaly detection from measurement data; 2) real-time event diagnosis and interpretation of changes in the state of the network; 3) real-time optimal control of the power grid; 4) scientific foundations underpinning cyber-physical systems. The major outcome of this project is practical solutions to event or fault detection and diagnosis in the power grid, as well as prediction and prevention of large-scale power outages.

As information technology has transformed physical systems such as the power grid, the interface between these systems and their human users has become both richer and much more complex. For example, from the perspective of an electricity consumer, a whole host of devices and technologies are transforming how they interact with the grid: demand response programs; electric vehicles; "smart" thermostats and appliances; etc. These novel technologies are also forcing us to rethink how the grid interacts with its users, because critical objectives such as stability and robustness require effective integration among the many diverse users in the grid. This project studies the complex interweaving of humans and physical systems. Traditionally, a separation principle has been used to isolate humans from physical systems. This principle requires users to have preferences that are well-defined, stable, and quickly discoverable. These assumptions are increasingly violated in practice: users' preferences are often not well-defined; unstable over time; and take time to discover. Our project articulates a new framework for interactions between physical systems and their users, where users' preferences must be repeatedly learned over time while the system continually operates with respect to imperfect preference information. We focus on the area of power systems. Our project has three main thrusts. First, user models are rethought to reflect the fact this new dynamic view of user preferences, where even the users are learning over time. The second thrust focuses on developing a new system model that learns about users, since we cannot understand users in a "single-shot"; rather, repeated interaction with the user is required. We then focus on the integration of these two new models. How do we control and operate a physical system, in the presence of the interacting "learning loops", while mediating between many competing users? We apply ideas from mean field games and optimal power flow to capture, analyze, and transform the interaction between the system and the ongoing preference discovery process. Our methods will yield guidance for market design in power systems where user preferences are constantly evolving. If successful, our project will usher in a fundamental change in interfacing physical systems and users. For example, in the power grid, our project directly impacts how utilities design demand response programs; how smart devices learn from users; and how the smart grid operates. In support of this goal, the PIs intend to develop avenues for knowledge transfer through interactions with industry. The PIs will also change their education programs to reflect a greater entanglement between physical systems and users.

The electric power grid experiences disturbances all the time that are routinely controlled, managed, or eliminated by system protection measures- designed by careful engineering studies and fine-tuned by condensing years of operational experience. Despite this, the grid sometimes experiences disruptive events that can quickly, and somewhat unstoppably catapult the system towards a blackout. Arresting such blackouts has remained elusive - mainly because relays (protection devices) operate on local data, and are prone to hidden faults that are impossible to detect until they manifest, resulting in misoperations that have sometime been precipitators or contributors to blackouts. Inspired by the Presidential policy directive on resilience -- meaning the ability to anticipate, prepare, withstand, and recover from disruptive events, this project proposes "WARP: Wide Area assisted Resilient Protection", a paradigm that adds a layer of finer (supervisory) intelligence to supplement conventional protection wisdom - which we call resilient protection. Exploiting high fidelity measurements and computation to calculate and analyze energy function components of power systems to identify disturbances, WARP would allow relays to be supervised - correct operations would be corroborated, and misoperations will be remedied by judiciously reversing the relay operation in a rational time-frame. The project also envisions predicting instability using advanced estimation techniques, thus being proactive. This will provide power grid the ability to auto-correct and bounce back from misoperations, curtailing the size, scale and progression of blackouts and improving the robustness and resilience of the electric grid -- our nation's most critical infrastructure. In WARP, disruptive events are deciphered by using synchrophasor data, energy functions, and dynamic state information via particle filtering. The information is fused to provide a global data set and intelligence signal that supervises relays, and also to predict system stability. Resilience is achieved when the supervisory signal rectifies the misoperation of relays, or endorses their action when valid. This endows relays with post-event-auto-correct abilities 
- a feature that never been explored/understood in the protection-stability nexus. Architectures to study the effect of latency and bad data are proposed. WARP introduces new notions: global detectability and distinguishability for power system events, stability prediction based on
the sensitivity of the energy function components and uses a novel factorization method: (CUR) preserving data interpretability to reduce data dimensionality. All the proposed
tools will be wrapped into a simulation framework to assess scalability and accuracy-runtime tradeoffs, and quantify the degree of resilience achieved. The effectiveness of the proposed scheme during extreme events will be measured by reenacting two well-documented blackout sequences. In addition, simulations on benchmarked systems will be performed to assess scalability and accuracy-runtime tradeoffs, and quantify the degree of resilience achieved.

Electricity usage of buildings (including offices, malls and residential apartments) represents a significant portion of a nation's energy expenditure and carbon footprint. Buildings are estimated to consume 72% of the total electricity production in the US. Unfortunately, however, 30% of this energy consumption is wasted. Virtual energy assessment is an approach that can optimize building energy efficiency and minimize waste at a low cost with minimal expert intervention. A virtual energy audit includes a thorough and near real time analysis of different sources of building energy usage, individualized energy footprints of load appliances and devices, and proactive identification of energy holes and air leakages. This project builds a low cost solution that combines the use of non-intrusive single point energy monitoring and low cost IR cameras to provide continuous energy audits. The system will provide continuous virtual audit reports to the landlords or residential users. The system will be deployed in low-income neighborhoods in Baltimore City, Maryland, where poor insulation problems are assumed to be fiscally insurmountable and low cost solutions to determining these issues is important for the landlords. To develop a scalable low cost virtual energy auditing system, this breakthrough research pursues the interfaces of smart building sensing, computing and actuation. The project will be executed under three main research thrust areas. First, it utilizes an autonomous discovery, profiling and rule-based predictive model to capture the relationship between disaggregated power measures and a device's actual usage patterns to pinpoint any abnormal consumption. Second, the PIs develop zero-energy far-infrared imaging sensors for low cost low frequency heat map scanning and air leakage detection. Third, the project engineers and evaluates cyber-physical building sensing system with a control level design perspective for virtual energy auditing that drives the realization of deep energy savings and building efficiency. Additionally, the PIs with collaboration from Constellation will host building energy education projects and workshop where undergraduate, high school, and underrepresented group of students would understand how to design and program energy meters and smart plugs.