Research Projects
Simone SilvestriAssistant Professor
Computer Science Department
University of Kentucky
Telephone:
Email: silvestri@cs.uky.edu
Current funded projects
|
Role: PI Funding agency: NSF CPS Period: 2020-2025 Amount awarded: $524,989 Project Description: While substantial progress has been made in the control of electric grid considering the cyber and physical characteristics, there has been a gap in the integration of smart grid research as it integrates with human behavior -- especially in interactions with energy management systems. For example residential energy consumption has been rapidly increasing during the last decades, especially in the U.S. where 2.6 trillion kilowatt-hours were consumed during 2015, and an additional 13.5% increase is expected by 2040 . Research efforts such as demand response have been made to reduce this consumption especially in smart residential environments. Concepts such as demand response have largely overlooked the complexity of human behaviors and perceptions, and recent research in the social-science domain and recent experience has challenged the effectiveness of this approach and in some instances led to an abandonment and avoidance of such concepts. The objective of this proposal is to overcome the limitations associated with state-of-the-art energy management systems by designing novel algorithms, machine learning models, and optimization techniques that specifically consider user behaviors, perceptions, and psychological processes. This revolutionary approach will unleash the full potential of smart residential environments in reducing residential energy consumption and has the potential to transform the way in which energy management systems are designed, implemented, and used by people. This project also supports innovative educational activities such as classes, real time demonstrations, coding challenges, and research experiences for high school students. The PI will also lead a cohort of students to the diversity-oriented Grace Hopper conference and teach seminars for Hispanic elementary students. Finally, a new class on Cyber-Physical-Human System will be designed and several graduate and undergraduate students will participate in the research activities. The proposed research combines novel algorithmic, machine learning, and optimization solutions that consider previously un-examined human behaviors, perceptions, and psychological processes. Specifically, in order to enable fine grained energy monitoring, we propose novel stream-based appliance recognition algorithms for smart outlets. These algorithms learn the appliance consumption signatures and the user engagement with the system to optimize the learning process. In addition, energy saving optimization strategies are designed by considering the user perception through social-behavioral well-being models. These models learned and refined through novel machine learning algorithms based on regressograms, interpolation, and regression using user feedback provided through a smartphone. In addition, we develop optimization algorithms for energy exchange in the context of smart residential environments equipped with renewable energy generation. These algorithms match the users' demand and production, by considering and learning also their availability and preferences in the energy exchange process. The proposed research is validated through real testbeds and large-scale simulations based on real traces. |
|
|
Role: PI Funding agency: NSF EPCN Period: 2019-2022 Amount awarded: $333,340 Project Description: The goal of this collaborative proposal is to develop novel machine learning based algorithms to address the problem of energy optimization at the building and district levels. These algorithms are integrated within a simulation framework that combines user behavior with the collaboration between buildings equipped with photovoltaic arrays, energy storage systems, and smart grid meters. The project proposes and integrates, within the same software tool, novel machine learning models for complex user behavior at the individual building level, for energy load prediction and energy storage systems scheduling at the district level, and for cost reduction via energy peak spreading. These models are used to formulate and construct algorithmic solutions based on reinforcement learning, recurrent and deep neural networks, and deep reinforcement learning suitable for implementation in the future generation Virtual Power Plants. The methodologies employed for energy reduction and cost minimization include: 1) alter user behavior through personalized recommendations regarding changes in the appliance states (e.g., heating and air conditioning settings), 2) district-level scheduling of energy storage systems among buildings equipped with photovoltaic arrays and smart grid meters, and 3) building-level scheduling of energy consumption events for smart appliances equipped with smart Internet-of-Things controllers to take benefit of different energy prices. |
|
|
Role: PI Funding agency: National Institute of Food and Agriculture Period: 02/2017-02/2021 Amount awarded: $802,981 Project Description: Smart energy management is at the core of future smart cities, since energy profoundly impacts the city's livablity, workability and sustainability. Key building blocks for smart energy management are intelligent residential environments, generally termed smart homes. These homes will include a plethora of smart interconnected appliances, realized through the Internet of Things paradigm, which can improve residential energy efficiency by controlling the energy usage. This research aims at designing previously unexamined social behavioral models involved in the human interaction with both smart appliances and smart energy management systems. Based on these models, we make use of graph theory to design formal user models that enable algorithm design and optimization. In addition, we propose machine learning techniques to correlate social behavioral dimensions to quantitative metrics observable by smart devices as well as algorithms that use this correlation to refine the user model. The formal models are used to design social-behavioral aware efficient algorithms for energy optimization for individual smart homes, as well as for communities of multiple homes in a microgrid. |
|
|
Role: Co-PI Funding agency: NSF CPS Period: 09/2015-08/2020 Collaborators: Sajal Das (MS&T), Mariesa Crow (MS&T) Amount awarded: $310,476 Project Description: Smart grid includes two interdependent infrastructures: power transmission and distribution network, and the supporting telecommunications network. Complex interactions among these infrastructures lead to new pathways for attack and failure propagation that are currently not well understood. This innovative project takes a holistic multilevel approach to understand and characterize the interdependencies between these two infrastructures, and devise mechanisms to enhance their robustness. Specifically, the project has four goals. The first goal is to understand the standardized smart grid communications protocols in depth and examine mechanisms to harden them. This is essential since the current protocols are notoriously easy to attack. The second goal is to ensure robustness in state estimation techniques since they form the basis for much of the analysis of smart grid. In particular, the project shall exploit a steganography-based approach to detect bad data and compromised devices. The third goal is to explore trust-based attack detection strategies that combine the secure state estimation with power flow models and software attestation to detect and isolate compromised components. The final goal is to study reconfiguration strategies that combine light-weight prediction models, stochastic decision processes, intentional islanding, and game theory techniques to mitigate the spreading of failures and the loss of load. A unique aspect of smart grid security that will be studied in this project is the critical importance of timeliness, and thus a tradeoff between effectiveness of the mechanisms and the overhead introduced. The project is expected to provide practical techniques for making the smart grid more robust against failures and attacks, and enable it to recover from large scale failures with less loss of capacity. The project will also train students in the multidisciplinary areas of power systems operation and design, networking protocols, and cyber-physical security. |
|
Past projects
|
Role: Co-Director Funding agency: NATO Science for Peace and Security (SPS) Programme Period: 2015-2019 Collaborators: Novella Bartolini (Sapienza University of Rome), Ala' Khalifah (Jordan German University) Amount awarded: 400'000 Euro Project Description: This project considers Hybrid Sensor Networks, composed by static, terrestrial an aerial sensors, to address on-demand, event-driven deployments, where complete and prompt coverage of the event area is required. In particular, the project designs new algorithms and protocols for event driven deployment. In addition, it defines new mission assignment algorithms to let devices autonomously coordinate with each other to perform several sensing tasks. Furthermore, the project investigates path planning strategies for unmanned aerial vehicles to recharge static sensors using radio waves. Finally, it designs algorithms for context assessment, situation awareness, and prediction of environmental changes. |
|
|
Role: PI Funding agency: NSF EPSCoR funded Missouri Transect Period: 2015-2016 Amount awarded: $71,055 Project Description: This proposal investigates the use of Unmanned Aerial Vehicles (UAVs) to monitor how crops are impacted by climate change and drought. The use of this technology is often prevented by its high cost. In this proposal, for the first time, we propose a framework to optimize the tradeoff between the monitoring accuracy provided by an UAV network, and its cost. The results of this project will enable the wide spread adoption of UAVs to autonomously and accurately monitor large scale crop fields. |
|
|
Role: PI Funding agency: Defense Threat Reduction Agency (DTRA), subcontract Pennsylvania State University Period: 2014-2015 Collaborators: Thomas La Porta (Pennsylvania State University), Ananthram Swami (Army Research Lab) Amount awarded: $162,540 Project Description: This project proposes general models and tools to analyze and predict the spreading of large scale failures in interdependent networks. The proposed models are mapped to real network instances such as the Internet and the Smart grid. The analysis is used to design network recovery strategies which target relevant objectives such as minimal recovery time, maximum network utility and minimum recovery cost. |
|