Abstract: In this paper, an intelligent inspection system for underground cable passage of substation based on PLC is designed to improve the inspection efficiency and safety of cable passage. The system uses high-precision sensors for data acquisition, integrates deep learning algorithms to achieve fault diagnosis, and combines remote monitoring technology for all-weather supervision of cable channels. The experimental results show that the intelligent inspection system is significantly superior to the traditional manual inspection in terms of fault detection rate, false alarm rate and inspection efficiency, which provides an effective guarantee for the safe and stable operation of the power grid.