In order to solve the problem of frequent disconnection of carbon fiber conductors for large-capacity and long-distance transmission due to undetectable hidden defects, a detection method for hidden defects in multi-strand carbon fiber composite core conductors based on optical fiber sensing technology was proposed. The operating environment of the wire was established and the operating conditions were simulated. The time-domain reflection technology based on distributed fiber Brillouin scattering was adopted to detect the temperature and strain distribution of carbon fiber wires, and the optical time-domain reflection technology was combined to detect the loss of optical fibers in carbon fiber wires. Comprehensive comparative analysis was conducted to obtain signal feature quantities such as fiber temperature, strain and loss that could characterize hidden defects in multi-strand carbon fibers, and a neural network model was constructed with each signal feature quantity as input to the model. Through model training, the various weight coefficients within the model were determined, enabling it to effectively detect hidden defects in multi-strand carbon fiber composite core wires. The experimental results show that this method can effectively obtain signal feature quantities of various types of fiber optic and can accurately detect various hidden defects in wires, which has important practical application values.