Robots could improve safety of power plant inspection, repair

Colorado School of Mines has been awarded funding from the U.S. Department of Energy to develop an artificial intelligence-enabled robot capable of inspecting and repairing power plant boilers.

Hao Zhang, assistant professor of computer science, is the primary investigator on the project, which will receive $400,000 over three years from the DOE’s Office of Fossil Energy.

“Our goal is to develop an integrated autonomous robot that is equipped with advanced onboard sensors to perform real-time inspection of threats on boiler furnace walls, operate a repair device to achieve live repair, and use artificial intelligence to enable smart data gathering and autonomy,” Zhang said. “Our robot will be designed to bridge the technological gaps in live inspecting and repairing of power plant boilers – reducing the risk for human operators and increasing boiler reliability, usability and efficiency.”

More specifically, Zhang and his team plan to develop live Non-Destructive Evaluation (NDE) sensors with signal processing techniques and reliability analysis, as well as design and evaluate live repair devices for robots based on fusion and solid-state technologies. The robot itself will be able to attach to and navigate vertical boiler furnace walls using magnetic drive tracks. 

“Boilers are the largest and one of the most critical components of a thermal power plant, which convert energy contained in fuel, like coal, into high temperature steam. Damage can occur in the boiler furnace chamber – damage that if left unchecked can cause catastrophic failures, causing loss of life and other serious safety issues,” Zhang said. “Boiler maintenance is challenging and dangerous for inspectors working on scaffolding in the confined space inside a boiler.”

An automated robot that can both inspect and repair boiler walls could eliminate the need to send operators into those difficult-to-access or hazardous areas, as well as the time-consuming and expensive scaffolding installation and de-installation, he said. 

The interdisciplinary research team includes Zhang as the AI robotics lead; Andrew Petruska, assistant professor of mechanical engineering, as the robotics platform lead; Zhenzhen Yu, assistant professor of metallurgical and materials engineering, as the repair lead; and Yiming Deng of Michigan State University as the NDE lead. Stephen Liu, professor of metallurgical and materials engineering and American Bureau of Shipping Endowed Chair at Mines, and Lalita Udpa of MSU will serve as the senior personnel on the project. The team is also closely collaborating with industry partners, including Xcel Energy and EnergynTech Inc.

Emilie Rusch, Public Information Specialist, Communications and Marketing | 303-273-3361 |
Ashley Spurgeon, Editor, Mines Magazine | 303-273-3959 |

About Mines
Colorado School of Mines is a public R1 research university focused on applied science and engineering, producing the talent, knowledge and innovations to serve industry and benefit society – all to create a more prosperous future.