With the advancement of sensing technologies, various structures, infrastructure assets and also industrial manufacturing processes can be monitored real-time today. This helps asset owners and operators improve both operational efficiency and future designs but comes with the challenge of big data management.
Acquisition of large data sets is a nice problem to have as long as meaningful conclusions can be drawn via advanced analytics based on sound physical principles.
TWI engineers have been working on monitoring of joining processes such as arc welding, static structures like bridges or pipelines and rotary equipment like aircraft engines or rotors in wind turbines.
The aim of structural health monitoring is to use sensors that detect damage on a structure in order that timely action is taken preventing failure, and managing maintenance regimes. Whereas periodic inspection and non-destructive testing is well established, continuous monitoring is less so.
One issue is the amount and complexity of the data that is produced. Acoustic emission and guided wave ultrasonics offer global methods for monitoring structural health, but only if specific failure modes, such as fatigue cracking and corrosion and their locations on the structure have been identified beforehand. A risk-based approach to determining inspection or monitoring requirements becomes necessary.
TWI engineers and mathematicians have been applying methods of signal processing and artificial intelligence in order to discriminate indications like corrosion and cracks. Following damage classification and its localisation, non-destructive testing (NDT) techniques are employed for a detailed characterisation. NDT results are subsequently shared with structural analysts for remaining life assessments.
If you are involved in asset integrity management, contact us by emailing firstname.lastname@example.org to see how we can help you reduce operational costs.