Improved Processing and Analysis of Test and Operating Data from Rotating Machines

Project ID: 22044
Principal Investigator: Stephen Agee
Research Topic: Improved Power Generation
Funded Fiscal Years: 2022, 2023 and 2024
Keywords: None

Research Question

How can the data acquired from large synchronous machines be processed and analyzed to reduce one of Reclamation's largest expenses: Operation and Maintenance (O&M)? Simultaneously, how can this data be used to enhance sustainability, reliability, and operations of Reclamation hydropower and pumping facilities? Every day, Reclamation collects millions of data points from its fleet of rotating machines. If analyzed properly, these data are extremely valuable and can be used to determine maintenance needs and outages for synchronous machines. In addition, data gathered from various online data acquisition systems and routine offline tests can be combined to provide a more complete picture of machine dynamics and overall equipment health.

As Reclamation continues to install more continuous monitoring systems to rotating machines, the amount of underutilized data continues to grow. If the collected data are not processed more thoroughly, much of the information that can be used to apply CBM and important machine diagnostics is lost. Large amounts of unprocessed data are commonly known as the "big data" problem and exist because of increased data collection without an increase in data processing and analysis tools. Many of these processing tools exist for other industries and would need to be adapted for use on hydro units.

In addition to historic operating data, Reclamation also has offline test data which can be used to create a complete picture of machine health. DCR tests are required to be performed every 6 years (FIST 4.1b) and provide a detailed record of generator stator insulation health over time. In addition, Reclamation has 30 plus years of experience performing and analyzing these tests and is the leading expert in the field in interpreting the results. For this reason, DCR test data have been chosen as the single source of offline test data for this research effort.

Need and Benefit

This research project focusses on development of new data science tools for existing data. The overall goal of this project is to use existing data to produce useful information that can help refine future maintenance activities. This project will work in conjunction with the regions, power managers, and PRO in a coordinated effort to implement improved maintenance techniques.

Contributing Partners

Contact the Principal Investigator for information about partners.

Research Products

Contact the Principal Investigator for information about these documents.

Return to Research Projects

Last Updated: 6/22/20