Abstract: Cable stripping is a key process in the process of power engineering construction and cable recycling, and its operation efficiency is directly related to the stripping quality. However, the traditional cable stripping operation is difficult to realize the real-time optimization and control of operation parameters. Therefore, this paper designs a cable stripping efficiency online evaluation and operation parameter optimization technology based on improved support vector machine. In this paper, the particle swarm optimization algorithm is introduced into the traditional support vector machine, which significantly improves the classification accuracy and robustness of the model. On this basis, the peeling efficiency evaluation model is established, and the real-time online judgment of the peeling quality grade is realized. The experimental results show that the accuracy rate of peeling efficiency evaluation of the proposed technology is 96.3%, which provides theoretical support for the development of intelligent equipment for cable peeling.