Skip to main content
Press Release

GE Integrating AI to Enable Performance-Informed Gas Turbine Inverse Design

June 24, 2020
  • GE researchers developing an artificial intelligence (AI) and machine learning (ML)- enabled inverse design framework that allows performance metrics to create more optimized designs for industrial gas turbine (IGT) aerodynamic components  
  • Project aims to achieve a 30-50% reduction in design cycle times, or from 1 year to a few months
  • Partnered with University of Notre Dame and GE Gas Power on the project
  • Emerging digital toolset will help push combined cycle power plant efficiency to new heights

NISKAYUNA, NY – June 24, 2020 – Aiming to let new performance metrics be the principal driver in the design of cleaner, more efficient aerodynamic energy systems, GE Research, the technology development arm for GE, has been awarded Phase I of a two -year, $2.1 million project through ARPA-E’s DIFFERENTIATE (Design Intelligence Fostering Formidable Energy Reduction and Enabling Novel Totally Impactful Advanced Technology Enhancements) program to build an AI-driven invertible neural network that can direct translate these metrics into optimized designs.     

Today, complex aerodynamic energy components such as gas turbine blades have extremely long design cycle times of more than a year that require compromise between cost, performance and reliability.  GE researchers, together with GE’s Gas Power business and the University of Notre Dame, are aiming to develop and demonstrate a new AI and ML- enabled design framework that takes half the time and is dictated almost entirely by the desired performance metrics to take the design of aerodynamic energy components to a whole new level.

Sayan Ghosh, a Lead Engineer in Probabilistic Design and project leader, explained the team is building a probabilistic inverse design machine learning framework – Pro-ML IDeAS – which uses an AI-driven invertible neural network to overcome multiple design iterations and challenges that typically require engineering expertise across many complex functional spaces to solve. “This will essentially create a paradigm shift in gas turbine design by enabling us to explore and discover new learning curves not previously possible,” Ghosh says.  “We believe that the Pro-ML IDeAS, powered by AI and ML, will allow us to break free from the traditional design constraints and let us achieve more optimal designs in significantly less time versus the current state-of-the-art.


CFD image

Pictured is a computational fluid dynamics (CFD) model predicting the flow trajectory and resulting losses through the hub of a gas turbine blade. Design and performance metrics, such as aerodynamic loss for a turbine, will evaluated by the artificial intelligence (AI) and machine learning (ML)-enabled inverse design framework being developed by GE researchers as part of their ARPA-E DIFFERENTIATE project.

Ghosh added, “One of the chief reasons GE Gas Power has set world records in combined cycle gas turbine (CCGT) efficiency, is the design of more efficient aerodynamic parts and components.   With the integration of new AI-powered digital solutions like our invertible neural network being supported through ARPA-E’s DIFFERENTIATE program, we will be well on the path to achieving 65% efficiency and beyond.”

GE’s HA gas turbine technology, which includes some of the most highly advanced parts and components, has helped to deliver two world records - one for powering the world’s most efficient combined cycle power plant, based on achieving 63.08 percent gross efficiency at Chubu Electric Nishi-Nagoya Power Plant Block-1 in Japan and another for helping EDF’s Bouchain Power Plant achieve 62.22 percent net combined cycle efficiency in France.

Together with the GE Research and Gas Power teams, a team of researchers from the University of Notre Dame team led by Prof. Nicholas Zabaras will bring more than 30 years of experience solving tough inverse/design problems. Prof. Zabaras’s pioneering work in the area of regularization techniques, high-dimensional Bayesian inverse methods, Gaussian process models for inversion, and most recently the integration of deep learning and inversion tasks will further accelerate learnings on this project.

The end goal of the two- year project is to create an inverse design process to optimize the design of a gas turbine blade component and reduce the design cycle time.  In future, the framework will also be extended to other applications such as aviation turbine engines, aeroderivative engines, wind turbines, and hydro turbines.

About GE Research

GE Research is GE’s innovation powerhouse where research meets reality. We are a world-class team of scientific, engineering and marketing minds working at the intersection of physics and markets, physical and digital technologies, and across a broad set of industries to deliver world-changing innovations and capabilities for our customers. To learn more, visit our website at

business unit