Artificial Neural Network Plan for Transmission Tower Structure Safety Assessment

• Published in: IOP conference series. Earth and environmental science 2021-05, Vol.772 (1), p.12010
• Main Authors: Sun, Mi, Xie, Jinghai, Jiang, Yu, Su, Dongyu, Lu, Shihua, Guo, Jia
• Format: Article
• Language: English
• Subjects: Artificial neural networks; Electric power grids; Evaluation; Feasibility; Finite element method; Mathematical models; Neural networks; Security; Training; Transmission towers
• ISSN: 1755-1307; 1755-1315
• DOI: 10.1088/1755-1315/772/1/012010

The security of transmission tower structure is the foundation to ensure the safe operation of power grid. The research of fuzzy qualitative assessment method for the structural health of transmission towers is the basic work of large-scale digital power grid engineering. In this paper, artificial neural network (ANN) is used as a generalized tool for structural health assessment of transmission towers. The feasibility of using artificial neural network to qualitatively evaluate the structural health of transmission towers is verified by numerical simulation data. In order to train the artificial neural network, the displacement, stress and other signals generated by the finite element model are calculated and used as the training input of the artificial neural network. The detailed stress results of the finite element model are calculated as the qualitative evaluation basis of the structural health, and the training output of the artificial neural network is generated. By the same method, the test set is generated, and it is found that the artificial neural network is feasible for the fuzzy qualitative assessment of the structural health of transmission tower, and the misstatement rate is less than 5%.