Multi-meteorological-factor prediction method for power transmission line

Abstract

The invention discloses a multi-meteorological-factor prediction method for a power transmission line, relates to the technical field of machine learning, and solves the technical problem of low meteorological prediction accuracy. The technical main points are that a TCN (Temporal Convolutional Network) network and a DenseNet network are combined for feature extraction, the TCN network can process time sequence information in parallel, and the method has a flexible receptive field and a stable gradient, occupies a lower memory compared with other methods, can perform good fitting on a sequential relationship, and improves the prediction accuracy of each meteorological factor; secondly, the DenseNet network is combined on the basis of the TCN network, so that over-fitting can be resisted, and the accuracy of weather prediction is improved; and finally, when each meteorological factor is predicted, historical micro-meteorological information is utilized, and weather forecast information is combined, so that the accuracy of meteorological prediction is improved.

Description

technical field [0001] The present disclosure relates to the technical field of machine learning, and in particular to a multi-meteorological factor prediction method for power transmission lines. Background technique [0002] The meteorological conditions around the transmission line have always been very important to the safety of the power grid system. There have been many natural disasters such as wind disasters and ice disasters in different degrees at home and abroad, and the power grid system has also experienced disconnections, tower collapses, and flashovers. And with the gradual warming of global temperature and the increasing number of extreme weather, the frequency of such disasters is on the rise. [0003] Transmission line disasters caused by high wind speed generally occur under severe weather conditions. When strong winds or hurricanes act on the wind pressure surface of wires and insulators, the wires will be deflected and displaced to a certain extent. Whe...

AI Technical Summary

Problems solved by technology

At present, the existing meteorological factor prediction models of transmission lines mainly use physical models to predict, such as WRF (The Weather Research and Forecasting Model, weather forecasting model), this type of model is generally difficult to fit the time series relationship of meteorological information, and cannot more accurate weather forecasts

Images