The application discloses a UHPC anti-bending test initial
cracking strength determination method and
system based on an image method, and belongs to the technical field of
mechanical property testing, and comprises the following steps: spraying random speckles on the surface of a UHPC test piece, pre-embedding an
optical fiber Bragg
grating sensor array in the test piece, and performing X-
ray computed tomography on test pieces in the same batch to establish a
fiber-matrix interface transition zone damage evolution model; when three-point anti-bending loading is performed on the test piece, surface speckle images are collected in real time, the DIC
algorithm is used to calculate the full-field
strain distribution, and the surface initial
cracking time and the strain concentration area coordinates are determined in combination with a criterion; meanwhile, the
wavelength drift of the
optical fiber is monitored, the internal initial
cracking time is determined by calculating the
coefficient of variation; the interface debonding rate and the interface initial cracking point time are calculated according to the model and the real-time interface stress; a multi-source initial cracking characteristic
data set containing the
surface strain, the internal strain anomaly and the interface debonding rate is constructed, the weight coefficients are obtained by inputting the neural
network model, and the initial cracking anti-bending strength is calculated by weighted summation.