Abstract:With the development of the society, the cable has been widely used with many advantages. However, the cable damage caused by various types of violations has been endless, causing huge losses in the social economy. Aiming at the problem of power cable external force damage in distribution network, this paper designs an anti external invasion early warning system for middle and low voltage distribution network cable channel based on the principle of optical fiber interference. The system can realize vibration signal identification and synthetically evaluate the intrusion signals. It mainly consists of two parts of the hardware system and software system. The invading vibration signals are collected in the monitoring range, and their attribute properties are extracted respectively. The support vector machine SVM and BP neural network are used to identify the intrusion signals. Intrusion vibration signals can be classified into three levels: safety, warning, and danger. Once the dangerous intrusive source is near the cable channel, the early warning system will alarm the person on duty in time by SMS. The test results show that the system is reasonable in design and has high accuracy in vibration signal recognition, which can reach 80%. The system has simple working principle and convenient installation, and is suitable for monitoring external force invasion of cable passages in medium and low voltage distribution networks.