Advanced signal processing of turbine rotor bore waveforms. Final report
EPRI Research Project 502-8 was established to investigate advanced signal processing methods as applied to the ultrasonic inspection of turbine rotor bores. The rf waveform responses from near bore defects contain much information which can be used to characterize these flaws. This information is not used in conventional UT systems. The purpose of this effort has been to (1) determine which signal processing operations could be used to enhance the quality and interpretation of rotor bore inspection data, and (2) to investigate advanced methods of characterizing the defects using the rf waveforms to provide the inspector with additional knowledge as to defect type. An outcome of this effort was to develop an Adaptive Learning Network (ALN) model tested using data recorded on the (retired) Kanawha and Connors Creek rotors. Subsequent destructive results showed that the ALN was able to discriminate accurately between inclusions and crack-like flaws in the Kanawha rotor.
- Research Organization:
- Adaptronics, Inc., McLean, VA (USA)
- OSTI ID:
- 5333837
- Report Number(s):
- EPRI-NP-2203; ON: DE82902202; TRN: 82-013331
- Country of Publication:
- United States
- Language:
- English
Similar Records
Reliability of steam-turbine rotors. Task 1. Lifetime prediction analysis system. Final report. [Using STRAP and SAFER computer codes and boresonic data]
Phased Array Ultrasonic Examination of Reactor Coolant System (Carbon Steel-to-CASS) Dissimilar Metal Weld Mockup Specimen
Related Subjects
20 FOSSIL-FUELED POWER PLANTS
FOSSIL-FUEL POWER PLANTS
STEAM TURBINES
NUCLEAR POWER PLANTS
ULTRASONIC TESTING
DATA PROCESSING
PERFORMANCE
ACOUSTIC TESTING
MACHINERY
MATERIALS TESTING
NONDESTRUCTIVE TESTING
NUCLEAR FACILITIES
POWER PLANTS
PROCESSING
TESTING
THERMAL POWER PLANTS
TURBINES
TURBOMACHINERY
220200* - Nuclear Reactor Technology- Components & Accessories
200104 - Fossil-Fueled Power Plants- Components