Although numerous damage identification methods have been proposed and applied to detect stiffness degradation location in superstructures of bridge

damages on bridge substructures are seldom covered.A damage identification approach based on fingerprint match for database of vibration curvature mode indicators is proposed to identify corrosion in pile steel casing of the cross ocean bridge.Marine environment may easily corrode the steel pipe and causes the peeling of pile surface.Refined multi-scale FE models were developed to simulate the damage of the towel base of Jintang Bridge in China.The damage location

axial damage length and damage depth of the corroded zone on the steel casing were extracted.Cubic-spline interpolation and central difference method were adopted to obtain the pile vibration curvature modes in all damage cases for a small number of displacement points.The corresponding damage identification fingerprint database was then established.Numerical experiments indicate that at least six interpolated points are necessary to obtain the accurate pile vibration mode

which is the base for deriving the curvature mode indicator.The location and axial length of the corroded zone in the steel pile casing can be accurately determined by matching the indicator database with a damage-unknown case.