Back to Search
Start Over
Research on the design and precise monitoring technology of intelligent early warning system for power tower settlement based on multi-dimensional information fusion technology
- Source :
- Applied Mathematics and Nonlinear Sciences, Vol 9, Iss 1 (2024)
- Publication Year :
- 2024
- Publisher :
- Sciendo, 2024.
-
Abstract
- Multi-dimensional information fusion technology helps to realize the synergy and complementarity between the information, which provides a guarantee for the settlement monitoring and early warning of electric power pole towers. The settlement monitoring algorithm for electric power pole towers is improved by this paper based on this technology and a settlement warning model is proposed. The traditional PDA algorithm can be enhanced by enhancing the pole tower image acquisition capacity with the wavelet transform algorithm and achieving the fusion and evaluation of multidimensional information through an isolated forest algorithm. The relationship functions of tilt angle, settlement wind speed size and other indicators are established, and the system of super-definite equations is used to solve the correlation coefficients, and the early warning system of pole tower settlement is constructed. The analysis results show that the monitoring results of the monitoring algorithm are fluctuating and stable; the absolute value of the error in the horizontal direction is not more than 9 mm, and the absolute value of the error in the vertical direction is not more than 14 mm. The predicted values of the maximum displacement and stress R2 are close to 1, and the MAPE is 0.436% and 1.123%, respectively. It indicates that the improved power pole tower settlement monitoring algorithm and early warning system in this paper have satisfactory performance.
Details
- Language :
- English
- ISSN :
- 24448656
- Volume :
- 9
- Issue :
- 1
- Database :
- Directory of Open Access Journals
- Journal :
- Applied Mathematics and Nonlinear Sciences
- Publication Type :
- Academic Journal
- Accession number :
- edsdoj.8b6c2f7e5e7045bca9e3bb5e4dcb8b6f
- Document Type :
- article
- Full Text :
- https://doi.org/10.2478/amns-2024-3124