FOR IMMEDIATE RELEASE | August 15, 2024 – Carnegie, PA
Department of Energy SBIR Phase I project will support development of EmiFor, an AI-driven forecasting and mitigation tool for amine emissions in carbon capture operations.
VISIMO Wins DOE SBIR Phase I Award to Develop EmiFor, an AI-Driven Amine Emissions Forecasting Tool for Carbon Capture
CARNEGIE, PA. – VISIMO, a Carnegie-based artificial intelligence and software engineering company, has been awarded a Department of Energy Small Business Innovation Research (SBIR) Phase I project to develop EmiFor, an AI-driven amine emissions forecasting and mitigation tool designed to support safer and more effective carbon capture operations.
Carbon capture technologies are an important part of efforts to reduce carbon dioxide emissions, but solvent-based post-combustion capture systems must be carefully managed to limit harmful byproduct emissions. During challenging operating states such as plant startup, shutdown, ramping, and partial-load operation, amine emissions can vary significantly, making accurate forecasting essential for operators seeking to manage risk, maintain compliance, and reduce environmental impact.
EmiFor is designed to help plant managers forecast future amine emissions in real time, understand causal relationships between operational variables and emissions, and evaluate “what-if” scenarios before changes are made in the plant environment. The goal is to move emissions management from a reactive process toward a more proactive, data-informed approach.
“Carbon capture will only reach its full potential if operators have the tools to manage the complete emissions profile of these systems,” said James Julius, Founder and CEO of VISIMO. “EmiFor applies VISIMO’s strengths in machine learning, forecasting, and decision-support software to a problem that matters for energy security, environmental protection, and the responsible deployment of carbon capture technology.”
Advancing AI for Carbon Capture Operations
The Phase I project will focus on developing and testing a machine learning framework for high-temporal-resolution emissions forecasting. EmiFor will be designed to predict amine emissions from post-combustion carbon capture operations while accounting for operating conditions and the physical processes that influence emissions.
VISIMO will evaluate statistical time-series forecasting methods alongside advanced machine learning approaches that can incorporate physical constraints into the modeling process. This approach is intended to improve forecasting performance while reducing the risk of models learning patterns that do not reflect the underlying carbon capture process.
In addition to forecasting, EmiFor will support causal impact analysis. This capability is designed to help users estimate how a specific operational change may have affected amine emissions by comparing observed outcomes against a modeled baseline. EmiFor will also support counterfactual analysis, allowing users to evaluate hypothetical operating scenarios and identify potential emissions mitigation strategies before implementation.
Supporting Safer and More Effective Emissions Management
Effective carbon capture requires more than capturing carbon dioxide. It also requires careful management of solvents, degradation products, and related emissions that may affect human health, ecosystems, and surrounding communities. By improving visibility into how amine emissions may change over time and under different operating states, EmiFor can help plant managers make more informed decisions about carbon capture operations.
The project has potential benefits for utilities, industrial facilities, carbon capture developers, federal stakeholders, and communities near large-scale energy infrastructure. Better emissions forecasting and mitigation planning can support environmental compliance, reduce operational uncertainty, improve solvent management, and strengthen confidence in carbon capture as part of a broader energy transition strategy.
Phase I Technical Objectives
During Phase I, VISIMO will evaluate the technical feasibility of EmiFor through a structured research and prototyping effort. The work will include acquiring and preparing relevant test data, developing technical infrastructure for forecasting and emissions mitigation analysis, implementing methods for data quality challenges, and testing the framework’s ability to support real-time forecasting, causal impact analysis, and counterfactual evaluation.
The Phase I effort is intended to lay the foundation for a future prototype that integrates the forecasting framework into a unified user interface. Later phases would focus on expanding the system, incorporating additional data sources, and preparing EmiFor for deployment in operational carbon capture environments.
“EmiFor is focused on helping operators understand not just what emissions may look like, but why they may change and what can be done about it,” said Dr. Tyler Gaona, Principal Investigator and Data Scientist at VISIMO. “That combination of forecasting, causal analysis, and scenario evaluation is what makes this work valuable for real-world carbon capture decision-making.”
About VISIMO
VISIMO is a Carnegie, Pennsylvania-based AI and software engineering company founded in 2015. VISIMO builds applied machine learning, data science, and mission software solutions for federal agencies and commercial partners, with a focus on rapid prototyping, defensible AI, and decision-support systems. VISIMO specializes in transforming complex data and technical uncertainty into practical tools that support better decisions in high-consequence environments.
Media Contact:
VISIMO Media Inquiries
VISIMO
412-699-6900
media@visimo.ai
Government Disclaimer:
This material is based upon work supported by the U.S. Department of Energy. Any opinions, findings, conclusions, or recommendations expressed in this material are those of VISIMO and do not necessarily reflect the views of the Department of Energy.