Investigation of AI-based Image Recognition Technology Combined with Sensor Technology for Power Grid Quality and Safety
Abstract
The power grid (PG) has the important task of transmitting, distributing, and supplying electricity, and is an indispensable infrastructure in modern society. Its quality and safety are directly related to productivity, economic development, and people’s everyday lives. The current traditional power grid quality and safety monitoring rely mainly on manual inspection, which has the problems of poor efficiency and high labor costs. In order to enhance the quality and safety of the power grid, improve the efficiency of power grid monitoring, and reduce energy consumption, this study combines image recognition technology with sensor technology based on artificial intelligence (AI) to conduct in-depth research on the quality and
safety of the power grid. This study uses images 1 and 2 of the sample PG route as infrared technology data, and performs noise reduction and feature extraction on the images, analyzing the role of sensor technology in PG job safety detection. To calculate the PG quality safety IR test results, this study sets the parameters to a total of 500 rounds every 50 times. The experimental results show that the node energy L of neural networks, genetic algorithms, and decision tree algorithms is totally consumed by the 600th iteration, while simulated annealing is completely consumed by the 550th iteration. This indicates that the combination of image recognition technology with sensor technology can efficiently monitor in real-time the quality and safety of the power grid, which helps to provide effective support for the safe and stable operation of the power grid.
Keywords: Image recognition technology, artificial intelligence, power grid system, sensor technology, power grid quality safety
Cite As
Z. Liu, Y. Bai, B. Hou, K. Ning, X. Liu, J. Zhang, "Investigation of AI-based Image Recognition
Technology Combined with Sensor Technology for Power Grid Quality and Safety", Engineering
Intelligent Systems, vol. 32 no. 5, pp. 533-542, 2024.