A novel transmission line icing prediction method

Abstract

The invention discloses a novel transmission line icing prediction method. The transmission line icing prediction method comprises the following steps: step 1, inputting selected indexes and carryingout standardization treatment on each input index; 2, constructing a least square support vector machine model; 3, improving the least square support vector machine to form a transmission line icing thickness prediction model based on the improved least square support vector machine. The invention adopts wavelet kernel function instead of traditional radial basis function to improve least square support vector machine.

Description

technical field [0001] The invention relates to the technical field of icing forecasting of power transmission lines, in particular to a novel method for predicting icing of power transmission lines. Background technique [0002] The icing of transmission lines will cause a series of accidents such as icing flashover of insulators, disconnection of ground wires, and collapse of towers, which will seriously affect the safe and stable operation of the power system. Since the record of icing accidents, there have been reports at home and abroad that the icing of transmission lines caused tower collapse and disconnection of high-voltage transmission lines, and some of the disasters were more serious. In January 1998, Canada was hit by a week-long ice storm that left 1 million customers without power. In January 2008, there were four large-scale rainy and snowy weathers in southern China. The power grid was severely covered with ice, and power lines were disconnected and towers ...

AI Technical Summary

Problems solved by technology

However, the ice thickness prediction model obtained by this data-based statistical method cannot be extended to other transmission lines with large geographical differences, so the expected effect of this model is not very ideal

Images