Cable defect positioning method and device fusing broadband impedance spectrum and attention mechanism

The invention relates to the technical field of cable detection, in particular to a cable defect positioning method and device fusing a broadband impedance spectrum and an attention mechanism. The construction method comprises the following steps: acquiring sample data by adopting a broadband impedance spectrum; and training a model adopting channel attention, space attention and a convolutional network based on the sample data and a preset loss function to obtain a cable defect positioning model. In the embodiment of the invention, based on the sample data acquired by the broadband impedance spectrum, the model constructed by the channel attention, the space attention and the convolutional network is utilized to carry out defect detection on the acquired data, and the distance positioning is completed, so that the local defect identification and positioning of the high-voltage cable are met, and the manual misjudgment rate is reduced while the manual labor amount is reduced. 本发明涉及电缆检测技术领域，具体涉及一种融合宽频阻抗谱与注意力机制