Back to Top
Skip to main content
NETL-Supported Project Addresses Boiler Impingement with High Performance Computing
Animated diagram depicting the beneficial factors and research thrusts of high-performing CFD computing

With NETL support, through the Lab’s University Training and Research program, researchers at the University of California, Riverside used advanced computing models that harness machine learning to efficiently reduce impingement in boilers — an innovation that can ensure longer and more efficient service life for power plants and even potentially extend the lives of helicopter rotor blades or aircraft engine components.

Erosion from particle impingement results in irreversible material degradation due to repeated impact of high-velocity particles on surfaces. This process causes perpetual wear of critical components in various energy and technological industries, like those used in petroleum refining and pipelines and in power plants. Mitigating these effects is particularly crucial because the financial loss from erosion is estimated to cost 1 – 4% of the gross national product in industrialized nations.

The effects of erosion from particle impingement continues to be concerning to energy and technology industries. Brute force computational fluid dynamics (CFD) approaches allow accurate predictions of complex erosion processes; however, these large-scale calculations can be very computationally expensive.

By harnessing Convolutional Neural Network (CNN) and Long and Short-Term Memory (LSTM) machine learning in combination with CFD the project developed a hybrid approach that can accurately predict entire particle trajectories and surface erosion profiles 600 times faster than conventional CFD calculations.

“The development out of University of California, Riverside is promising because it can extend the life of existing energy facilities in a faster and cheaper method than what’s currently available, which means cost savings for all involved stakeholders,” said Andrew Downs, an NETL federal project manager. “Furthermore, because impingement affects several other industries, this machine learning hybrid analysis could be used to help extend the lives of helicopter rotor blades or aircraft engine components, to name a few.”

The primary goal of this project was equipping underrepresented, diverse student groups with cutting-edge, translatable skillsets through research and development opportunities that will help them sustain successful STEM (science, technology, engineering and mathematics) careers. NETL’s University Training and Research program supports the Historically Black Colleges and Universities and Minority Serving Institutions (HBCU–MSI) such as University of California, Riverside. Projects such as these are pivotal for investing in the next generation of scientists and engineers to address the nation’s greatest energy and environmental challenges.

Technical goals of the project were to calculate and analyze particle impingement within boilers, quantify effects of particulates in boilers, and predict damage rates of boilers under different cycling modes. Collectively, these initiatives develop insight into existing power plant challenges using advanced modeling tools, particularly those leveraging high-performance computing resources.

NETL is a U.S. Department of Energy national laboratory that drives innovation and delivers technological solutions for an environmentally sustainable and prosperous energy future. By leveraging its world-class talent and research facilities, NETL is ensuring affordable, abundant and reliable energy that drives a robust economy and national security, while developing technologies to manage carbon across the full life cycle, enabling environmental sustainability for all Americans.