CTMA Project #: 140999
Problem: Inspection and repair of large, thick section welds in critical applications has seen little innovation in the last few decades, with the same costly, and relatively slow methods in use. Commercial companies involved in welding intensive fabrication enterprises, as well as those engaged in maintenance and sustainment industries will require new sensors, and sophisticated scalable databases coupled with artificial intelligence and deep learning tools in order to achieve major advances in reducing the time and cost associated with ensuring quality in critical thick section welds.
Benefit: A reliable real-time detection method that would allow defects to be identified as they form, would generate substantial costs savings, and improve productivity and quality making a major difference in the maintenance and sustainment of systems in which thick section welding is a key element.
Solution/Approach: The overall approach is to demonstrate a weld quality monitoring approach that can reliably identify when a defect is forming in a weld using data from acoustic sensors with narrow band optical sensors to improve weld diagnostic capability accuracy. It would allow maintainers to know when a weld defect is forming, so that welding can be stopped, and the problem addressed immediately prior to additional resources being expended. The solution would improve the quality and efficiency of thick section, critical application repair welding processes applicable to commercial and public business today.
Impact on Warfighter:
- Improve weld quality, inspection, and productivity
- Reduce maintenance costs
- Increase safety
- Enhance warfighter readiness
DOD Participation:
- U.S. Navy
- U.S. Fleet Forces Command
- U.S. Air Force (observer)
- U.S. Marine Corps (observer)
Industry Participation:
- Perisense, Inc.
- Microsoft Corporation
- NCMS
Benefit Area(s):
- Cost savings
- Repair turn-around time
- Safety
- Improved readiness
Focus Area:
- Enhanced inspection
