Skip to main content
Research Highlight | Energy Systems and Infrastructure Analysis Division

Project looks at power of big data to predict hydropower component failures

Developing new models that can use data generated by sensors on hydropower components to predict how the components will degrade over time and estimate a component’s remaining life.

Researchers from Argonne and Idaho national laboratories and Wayne State University are developing new models that can use data generated by sensors on hydropower components to predict how the components will degrade over time and estimate a component’s remaining life. Digitally transforming risk prediction can guide effective operations and maintenance (O&M) policies for hydropower facilities, enabling proactive mitigation, reducing forced outages, and lowering O&M costs. Researchers will leverage monitoring/sensor data from the Hydropower Research Institute, covering 44% of MW hydro-capacity in the U.S., and maintenance records and sensing data from industry partners. Researchers will leverage hydro prognostics capabilities developed in Department of Energy Office of Energy Efficiency and Renewable Energy-Water Power Technologies Office (WPTO) seedling projects and integrate their asset management models into an open-source tool that will be co-developed with industry partners. Feng Qiu, a principal computational scientist and group manager for advanced grid modeling – optimization and analytics at Argonne, is principal investigator for the project, which is funded with $500,000 from WPTO