Prediction of ice thickness of Optical Fiber Composite Overhead Ground Wire (OPGW) based on multi-class support vector machine

Optical fiber composite overhead ground wire (OPGW), which consists of ground wire and optical fibers, plays an important role at electric power communication system. In daily operation, due to bad low temperature, OPGW will be covered by ice, which will lead to OPGW galloping and sleet jump of the wire. Meanwhile, OPGW icing can also make the wire with strain, which has the possibility of fiber fracture. Therefore, it is very important to judge and predict the state of OPGW icing. In this paper, a prediction model of ice coating thickness of OPGW based on multi-class Support Vector Machine (SVM) is proposed. In this model, the optical cable icing data measured in actual operating environment and laboratory simulation experiments are used as the data set to construct the prediction model.