Across the hall from the main stage is the CMU Research Exhibition, highlighting our most exciting energy research and innovations in a more intimate format. Attendees will have the opportunity to interact with CMU researchers and view their findings and/or innovative technologies.
When: Thursday, March 19, from 12:00 - 1:15 PM
Where: McKenna Peter Wright (across the hall from Rangos, the Energy Week main stage)
Research exhibition booths
| From the research group of: | Title | Description |
|---|---|---|
Trustworthy Forecasting for Severe Weather Events |
Trustworthy probabilistic weather forecasts are critical for managing grid operations, particularly during severe weather that can disrupt infrastructure and shift electricity demand. This work develops machine learning methods with the goal of making AI-based forecasting models more reliable and practical by providing clear diagnostic tools, transparent uncertainty measures, and systematic recalibration methods that support informed operational decision-making. |
|
How Regulation Affects Construction in the U.S. Electricity System |
The U.S. needs to double its regional transmission capacity to decarbonize, but regulatory frameworks and incentives often favor local upgrades over larger-scale transmission expansion. This research tests whether a major change in federal transmission regulation creates an incentive for utilities to invest locally and helps policymakers understand what barriers to transmission construction exist. |
|
Quantifying Minimum Energy Needs for Household Comfort and Safety |
Traditional ways of measuring energy poverty focus on what people spend, which often overlooks families who dangerously limit their heating or cooling to stay within their budget. This research uses advanced computer simulations of various home types to establish a "minimum energy baseline," determining the actual amount of electricity a household needs to maintain a healthy and safe indoor environment. By defining these health-based standards, we can ensure the power grid reliably meets the essential survival needs of every household as we transition to an all-electric future. |
|
Advanced Methodology for a More Reliable Modern Power Grid |
As power systems incorporate increasing levels of renewable and inverter-based resources, detailed time-based simulations are essential for reliable planning and grid security. However, traditional methods for studying long-term operating conditions use simplified assumptions that do not align with these detailed models. As a result, it is difficult to determine both where the system will ultimately settle and whether it will successfully reach that point. This research introduces a method that adapts high-fidelity electromagnetic models for evaluating long-term operating conditions, supporting clearer and more consistent stability assessments. |
|
Microgrids for AI Datacenters |
New AI data centers are driving growth in electricity demand, and these new facilities can increase local utility costs. We evaluated microgrids to power these facilities, where data centers would purchase and operate their own power plants without connecting to the local electric grid. We found that these microgrids can power AI facilities without affecting local ratepayers, and that solar and battery-based systems are preferred in most US locations when valuing deployment speed. |
|
Data-Driven Methods for Safe and Resilient Grid Planning and Operations |
Four distinct projects using AI/ML approaches will be featured, ranging from novel long-term resilience planning, models that provide planners resiliency/cost tradeoffs to make more informed decisions, a tool that provides early-warning checks for grid safety to prevent instability and outages, and a planning framework for making distribution upgrades to minimize wildfire risks with minimal service impacts. |