Transmission tower structure state evaluation method and system

Abstract

The invention provides a method and a system for evaluating the structural state of a power transmission tower. The method comprises the following steps: acquiring a to-be-evaluated power transmissiontower image through satellite remote sensing; utilizing a plurality of pre-constructed classification models to predict the occurrence probability of the to-be-evaluated transmission tower image in each set structure state; and determining the structural state of the power transmission tower based on the prediction result obtained by each classification model and the weight set for the predictionresult of each classification model. Based on the satellite remote sensing technology, rapid and full-coverage acquisition of the power transmission tower image in various complex environments can berealized; a multi-classifier fusion mode is adopted to identify the structural state of the power transmission tower, so that the limitation of using a single classifier to identify a target is overcome, and the accuracy of identifying the structural state of the power transmission tower is improved; and thus, efficient, timely and accurate evaluation of all transmission tower body structure states under various complex environment conditions is realized on the whole.

Description

technical field [0001] The invention belongs to the technical field of power transmission tower monitoring, and in particular relates to a method and system for evaluating the structural state of power transmission towers. Background technique [0002] At present, in the power grid system with UHV transmission channels as the backbone, high-voltage transmission towers are one of the most important infrastructures in the power grid system. Affected by various natural and man-made disasters such as strong winds, freezing, lightning strikes, floods, and external damage, serious tower structural failures such as tower collapse, large deformation of the tower head, tower tilt, loss of tower materials, and bending of main materials have seriously threatened The safe and stable operation of the entire power grid, therefore, how to efficiently, timely and accurately evaluate the structural status of the transmission tower body is an urgent need for the relevant management and operat...

AI Technical Summary

Problems solved by technology

[0005] From the perspective of related fields, there are relatively many studies on the recognition of military targets such as tanks, aircraft, and ships using satellite remote sensing data. Researchers at home and abroad have proposed different deep learning models, but these models are difficult to directly transplant into the multi-state evaluation of transmission towers. , the reason is that the technical nature of rapid assessment of five different types of towers, tower collapse, large deformation of tower head, tower tilt, loss of tower material and bending of main material, belongs to multi-type classification, and there are few samples of transmission towers, especially the deformation of tower head The number of abnormal transmission towers including , collapsed towers, etc. is very small (that is, there are few negative samples), it is difficult to support the training of the deep convolutional neural network model, and it cannot be directly used for the post-disaster state assessment of transmission towers; however, if only using Traditional shallow features (such as brightness, texture, G0 distribution features, etc.), existing research has proved that the robustness and accuracy of the algorithm cannot be guaranteed, and the portability is not enough.

Images