LASERNET Optical Oil Debris Monitor
Abstract
The LASERNET optical oil debris monitor has been developed for real time online identification of fault type and severity through detection of size, shape and rate of production of failure related debris in critical applications such as engines and gearboxes of helicopters. Detection of failure related debris without sampling has required the development of a high resolution high speed imaging and processing capable of recording and analyzing images at rates up to 500 frames per second. We have designed and constructed such a system based on parallel/series CCD technology, high speed dedicated image processors and neural net classifiers. The system has been tested on the T700 engine at the helicopter power train test cell at Naval Air Warfare Center, Trenton, NJ. Qualitative performance evaluation has demonstrated the ability to detect debris in real time and to distinguish and classify air bubble patterns in varying degrees of complexity. The overall false alarm rate depends on the strategy adopted in the image processor. For a dual processor architecture the results indicate that LASERNET will be capable of operating with a false alarm (defined as an incorrect identification of a rejectable gear box) rate of less than one every 2000 operating hours.
Document Details
- Document Type
- Technical Report
- Publication Date
- Jan 01, 1998
- Accession Number
- ADA347454
Entities
People
- A. Schultz
- J. Reintjes
- L. L. Tankersley
- M. D. Duncan
- T. L. Mcclelland
Organizations
- United States Naval Research Laboratory