Hybrid GA and Improved CNN Algorithm for Power Plant Transformer Condition Monitoring Model

Under the general trend of smart grid development in China, it has especially importance to maintain stability power generation, safety operation and reliability supply. However, most plants need participate frequency regulation market spot market, resulting frequent load fluctuations often unstable operating conditions generation equipment. In this study, a real-time monitoring method based on hybrid Genetic Algorithm (GA) Convolutional Neural Networks (CNN) algorithm is utilized monitor status transformers real time. The GA-CNN model proposed by analyzing advantages disadvantages CNN GA. It proved that accuracy greatly improved compared with CNN. recognition results, error rate only 1.86%, while 4%; random matrix predicted actual output values 98.11%, three factors affecting equipment, namely temperature humidity external environment daily plant, are also acceptable. selected for study able detect abnormalities state provide timely feedback changes