Transitioning our aging power grid to a smart grid brings a number of benefits including increased efficiency and sustainability as renewable resources (e.g., solar and wind) are incorporated into the grid. Additionally, it will also improve reliability and resiliency as smart sensors will be deployed throughout the grid to detect, respond, and adapt to events such as faults and failures. Finally, through smart meters and other Internet-of-Things devices in smart homes, the smart grid will also bring economical benefits to both power producers and consumers, as producers can introduce real-time pricing to reduce peak power consumption, and home automation systems of consumers can adapt accordingly while satisfying the needs and constraints of the homeowners.
We are currently pursuing a number of parallel subprojects within this area:
- We are investigating appropriate formalisms and algorithms for the elicitation and scheduling problem of home automation systems. Home automation systems need to intelligently elicit preferences and constraints from homeowners regarding their use of smart devices as exhaustive elicitation is not feasible. Such systems also need to schedule the devices in an efficient manner while optimizing certain objectives (e.g., minimizing the monetary cost to the homeowner, minimizing the peak power demand) while satisfying the preferences and constraints elicited.
- We are working towards transitioning off-the-shelf scheduling and coordination algorithms towards a working prototype of a home automation system. The goal is to deploy this prototype system in our microgrid at NMSU. Such a system can also be a testbed for the algorithms developed above. More info on this effort can be found here.
- We are also participating in PowerTAC – a trading agent competition, where an autonomous agent plays the role of a broker in an energy marketplace. The goal of the agent is to buy energy from wholesalers and sell them to consumers in such a way that maximizes their profit.
- Huiping Cao, New Mexico State University
- Ferdinando Fioretto, University of Michigan
- Russell Folk, Sandia National Laboratories
- Christopher Kiekintveld, University of Texas, El Paso
- Tiep Le, New Mexico State University
- Francisco Natividad, Google
- Enrico Pontelli, New Mexico State University
- Satishkumar Ranade, New Mexico State University
- Tran Cao Son, New Mexico State University
- Atena M. Tabakhi, Washington University in St. Louis
iCREDITS: Interdisciplinary Center of Research Excellence in Design of Intelligent Technologies for Smartgrids.
National Science Foundation (2014 – 2019).
- Atena M. Tabakhi, Tiep Le, Ferdinando Fioretto, and William Yeoh. “Preference Elicitation for DCOPs.” In Proceedings of the International Conference on Principles and Practice of Constraint Programming (CP), to appear, 2017.
- Ferdinando Fioretto, William Yeoh, and Enrico Pontelli. “A Multiagent System Approach to Scheduling Devices in Smart Homes.” In Proceedings of the International Conference on Autonomous Agents and Multiagent Systems (AAMAS), to appear, 2017.
- Ferdinando Fioretto, William Yeoh, Enrico Pontelli, Ye Ma, and Satishkumar Ranade. “A Distributed Constraint Optimization (DCOP) Approach to the Economic Dispatch with Demand Response.” In Proceedings of the International Conference on Autonomous Agents and Multiagent Systems (AAMAS), to appear, 2017.
- Moinul Morshed Porag Chowdhury, Russell Y. Folk, Ferdinando Fioretto, Christopher Kiekintveld, and William Yeoh. “Investigation of Learning Strategies for the SPOT Broker in Power TAC.” In Agent-Mediated Electronic Commerce: Designing Trading Strategies and Mechanisms for Electronic Markets, Sofia Ceppi, Esther David, Chen Hajaj, Valentin Robu, and Ioannis A. Vetsikas (editors). Volume 271 of Lecture Notes in Business Information Processing, pages 96–111. Springer, 2017.
- Francisco Natividad, Russell Y. Folk, William Yeoh, and Huiping Cao. “On the Use of Off-the-Shelf Machine Learning Techniques to Predict Energy Demands of Power TAC Consumers.” In Agent-Mediated Electronic Commerce: Designing Trading Strategies and Mechanisms for Electronic Markets, Sofia Ceppi, Esther David, Chen Hajaj, Valentin Robu, and Ioannis A. Vetsikas (editors). Volume 271 of Lecture Notes in Business Information Processing, pages 112–126. Springer, 2017.