Quantitative estimation and prediction method for icing load of power transmission line

Abstract

The invention relates to a quantitative estimation and prediction method for icing load of a power transmission line, and belongs to the technical field of online monitoring of overhead power transmission lines. The method includes establishing a chaos time sequence model of a power transmission line icing process by using icing process historical data of a monitoring point and based on the phase space reconstruction theory; and establishing a quantitative estimation and prediction model of the icing load of the power transmission line based on a machine learning method of a support vector machine (SVM). According to the models, icing online estimation results based on a mechanical model are amended, or online estimated values are replaced when a mechanical sensing device fails, and trend prediction is performed on the icing process of the power transmission line based on micrometeorological information known in advance. The method has the advantages of being capable of amending online estimation results of the icing load of the power transmission line, provided with the capacity of estimating the icing load of the power transmission line based on micrometeorological data when the mechanical sensing device fails, and capable of predicting the icing load trend of the power transmission line according to the micrometeorological information of monitoring points known in advance.

Description

Technical field: [0001] The invention belongs to the technical field of on-line monitoring of overhead transmission lines, and relates to a method for quantitatively estimating and predicting the icing load of transmission lines, in particular to a method for quantitatively estimating and predicting the icing load of transmission lines in low-latitude and high-altitude areas based on micro-meteorological information. Background technique: [0002] In the low-latitude and high-altitude areas of Yunnan, Guizhou, and Sichuan in my country, due to the complex topography and changing climate and environment, overhead transmission lines often pass through areas with criss-cross mountains, rivers, and altitudes. Every winter, the northern cold current and the southern warm and humid air flow meet in the above-mentioned areas, and are affected by the micro-topography and micro-climate to form intermittent icing phenomena. The severe icing process will lead to mechanical accidents su...

AI Technical Summary

Problems solved by technology

The Makkonen model and the wind tunnel test model are icing prediction methods based on analytical expressions, which require the accuracy of the analytical model and cannot predict the abrupt change in the deicing process.

[0009] On the other hand, the geographical and climatic characteristics of different monitoring points cannot be satisfied, and the robustness of the model obviously has problems

The generalization ability of the intelligent computing model based on the neural network is not strong, and it is more constrained by the data; the only predictive model based on fuzzy reasoning can only perform qualitative estimation and prediction of the ice covering process, but cannot do quantitative estimation and prediction. Therefore, the gross errors in the icing estimation results of the mechanical model cannot be identified and replaced

Images