Optimised ambient vibration testing of long span bridges

Vibration testing of long span bridges is becoming a commissioning requirement. Long span bridges represent the extreme end of experimental capability, with challenges for logistics and access (due to length and location), instrumentation (due to frequency range, resolution and physical separation of accelerometers) and system identification (because of the extreme low frequencies). Similar challenges apply to other extreme structures such as tall buildings, masts, offshore lighthouses and extended geotechnical structures stretching technology requirements for both instrumentation and signal interpretation. A solution for instrumentation is autonomous 'wireless' recorders. The problem with signal interpretation is the reliability of the modal parameter estimates that is particularly challenged with low frequency modes, which 'third generation' operational modal analysis procedures offer using Bayesian approaches. The paper describes a preliminary exercise combining both these technologies in readiness for testing two very large bridges, one in China and one in Scotland.