Panoramic Intelligent Monitoring of Power Equipment Using Dynamic Black Hole-Driven DCGAN Under New Power Systems

Traditional ways of monitoring power systems do not offer sufficient real-time information on equipment status and do not sufficiently address various operational scenarios and parameters. To address these problems, a new method referred to as Dynamic Black Hole-driven Deep Convolutional Generative Adversarial Network (DBH-DCGAN) has been developed. This method utilizes the dynamic Black Hole mechanism that can adjust the flexibility and stability of the DCGAN model according to the power condition. The purpose of this study is to present and assess the novel DBH-DCGAN approach and its impact on improving the accuracy and efficiency of power plant monitoring. A large set of power equipment images was gathered that contains data regarding all the equipment. The images were then pre-processed using Histogram Equalization to improve the contrast of the images. To enhance the monitoring accuracy and flexibility in different power system situations, the proposed Dynamic Black Hole-driven Deep Convolutional Generative Adversarial Network (DBH-DCGAN) method was applied. Experimental results demonstrate that DBH-DCGAN effectively monitors power plants across different operating conditions, achieving performance metrics of recall (95.4%), accuracy (94.2%), and F1-score (96.3%). The study concludes that the DBH-DCGAN method significantly improves reliability and efficiency in power system management, thereby advancing intelligent monitoring technologies within the power grid.