The invention discloses a method and a
system for detecting fracture
impact elastic
waves of a prestressed
tendon of a prestressed composite
steel cylinder concrete
pipe. The method comprises the following steps: S1, obtaining the
surface wave velocity of
pipe core concrete; s2, obtaining the predominant frequency, the
P wave arrival time and the
P wave velocity of the
signal; s3, calculating the component thickness of the measuring point; s4, executing S1-S3 on all the measuring points in the measuring area to obtain P-
wave velocity, predominant frequency, P-wave
arrival time and component thickness of all the measuring points to form a
data set; and S5, reconstructing the P-
wave velocity, the predominant frequency, the P-wave time, the component thickness and whether the
label is broken into a
time sequence, designing a double-layer LSTM network, and training an LSTM model. According to the invention, the
machine learning technology is utilized to comprehensively consider the influence of the P-
wave velocity change of the
pipe core concrete caused by the breakage of the prestressed
tendon and the stripping of an external
mortar protective layer on the thickness of the component, and the breakage position of the prestressed
tendon of the prestressed composite
steel cylinder concrete pipe can be detected.