Smart buildings and connected communities are the cornerstones of future sustainable power grids. Tomorrow’s energy communities and the resilient operation of the energy infrastructure hinges on accurate prediction of buildings’ temporal energy presumption (production and consumption). Traditional prediction models primarily leverage historical information (such as energy demand) at the grid or building level. This is mainly due to data sharing hesitance which originates from privacy concerns of building managers and occupants. This project intends to develop a novel distributed machine learning model for predicting the temporal energy needs of future connected communities. The proposed multi-agent approach leverages centralized oversight of a central aggregator to inform distributed collaboration among each building, taking advantage of additional private information that is not shared with the aggregator. The proposed solution lends itself well to a multi-agent structure of connected energy communities where agents are willing to collaborate to improve their prediction accuracy. [Press Release].

Climate goals under the Biden Administration include a 50 percent reduction in greenhouse gas (GHG) pollution from 2005 levels in 2030 and a 100 percent carbon pollution-free electricity by 2035. Achieving these objectives requires significant infrastructure investments as well as substantial technology and policy developments—and transportation electrification sits at the intersection of these changes. Supported by Carnegie Mellon University's Block Center for Technology and Society, this project focuses on addressing the policy needs of electrifying the U.S. transportation sector. This work intends to develop a policy analysis framework to measure the potential impacts of various federal investment scenarios to find efficiencies and maximize the carbon offset potential of new electric grid investments [Press Release].

This project is focused on helping REN, operator of the national electricity transmission grid and the national natural gas transportation grid in Portugal, to architect the Portuguese future grid. Portugal's renewable energy portfolio is expected to significantly grow over the coming years and the future grid should be able to accommodate high-level penetration of intermittent energy resources. This multi-year partnership is focused on developing long-term operational plans to expand and operate the grid as economically as possible according to reliability criteria. Expanding the current transfer capacity of the electric network is a challenging problem to solve given the many renewable generation scenarios and infrastructure expansion options that need to be taken into consideration. Our recent success in ARPA-E's Grid Optimization Competition has led to the inception of GRID ARMOR framework. The GRID ARMOR platform enables solving national scale power system security and optimization problems and the we intend to extend the capabilities of this platform to solve grid expansion planning for REN.

Grid ARMOR (which stands for Grid Analytics, Rapid Modeling, and Optimization Routines) platform is an award winning grid optimization platform that was ranked top 10 in all divisions of the ARPA-E Grid Optimization (GO) Competition. The competition results were announced by the Secretary of Energy and can be found [here]. Grid ARMOR is a lightweight optimization algorithm that could optimize power generation settings across large power networks while adhering to physical grid constraints and will enable solving national scale power system security and optimization problems. [News announcement]

With financial support of the State of Pennsylvania through the Engineering Research Accelerator, this project intends to investigate and mitigate the impact of communication disruption on the coordination and operation of Internet of Things (IoT) connected Distributed Energy Resource (DER)s and power systems operation. Affordable energy resources are shaping the future of power networks and their resilient and secure operation is of prominent importance for maintaining reliability of electric power networks. Understanding the dependencies in these complex cyberphysical systems (networked DERs and communication networks) and mitigating the potential risks are critical for the successful development and evolution of smart electric infrastructures. The results of this project will be implemented using the Carnegie Mellon's campus-wide energy IoT test-bed known as CarnegiePLUG.

A team comprised of faculty from Carnegie Mellon University and University of Colorado Boulder were one of ten university teams selected to receive funding to participate in the ARPA-E Grid Optimization Competition. This challenge seeks to find optimization algorithms which allow the electrical grid to be operated in an efficient and resilient way, even as the grid is pushed further and further towards its limits. In a collaboration with the University of Colorado Boulder, the Carnegie Mellon University team will design intelligent optimization and control algorithms within the next year which can provide efficient solutions to the security-constrained optimal power flow problem (SCOPF). Learn more about the project. [Project Webpage]

Funded by Pennsylvania Infrastructure Technology Alliance (PITA) program and in collaboration with Grid Fruit LLC, this project is focused on developing scalable coordination methods for thermal and electrochemical storage devices. Through real-time distributed coordination of geographically spread small scale storage devices, our solution can provide a meaningful load shifting capacity to help the electric grid. [Related paper]

Warehouse inventory management is the corner stone of supply chain management for retailers. Inaccurate on-hand inventory data can result in a great financial loss for retailers. In partnership with our industry partner in Thailand, we are working on real-time inventory analytics solutions to provide actionable intelligence to increase inventory turnover and improve product freshness.

This project is focused on helping one of the nation's leading retailers to achieve sustainability goals and reduce carbon emission through intelligent renewable integration and building energy management. [Related paper]

This project will develop and demonstrate a distributed, agent-based control system to integrate smart inverters, energy storage, and commercial off-the-shelf home automation controllers and smart thermostats. The system will optimize photovoltaic (PV) generation, storage, and load consumption behaviors using high-performance, distributed algorithms. Learn more about the project. [Project Webpage]