Challenges in Implementation of SHM

SHM is being implemented in more and more practical situations and the gains are fairly visible. Users are learning form the experiences of the existing SHM systems. The technology will mature with time and increased use.

Following are the major challenges in its implementation:

• Improved and cost effective SHM sensors for stable and reliable performance during the useful life of the host structure.
• Proven capability of the SHM systems in detecting undesired changes in the structures in percentage terms with minimal false alarms.
• Effective integration of the SHM system elements for improved structural maintenance strategies such as integration of sensors along with data transmission, computational power and processing ability in the structures.
• Effective filtering of data acquired from sensors and adequate computational models of the structure.
• Optimally define the number and location of the sensors; identification of the features sensitive to small damage levels; the ability to discriminate changes in these features caused by damage from those caused by changing environmental and / or test conditions.
• Development of statistical methods to discriminate features from undamaged and damaged structures and comparative studies of different damage identification methods applied to common data sets.
• Conceiving systems for data reduction and diagnostic formulations.
• In many situations feature selection and damage identification must be performed in an unsupervised learning mode where data from the damaged systems are not available.