Fault Diagnosis of AC Transmission System via GoogleNet

The application of intelligent detection means in AC transmission systems is crucial to ensure the reliable operation of power systems and improve the quality of power supply. The traditional detection methods rely on human expertise and prior knowledge, which may not be efficient in complex transmission systems. Therefore, it is essential to develop an intelligent detection method that can automatically learn from the system's behavior and detect potential faults. To address this issue, we propose an intelligent fault detection method based on machine learning. This method relies on the abnormal operation database of AC transmission systems and combines it with the experience of first-line power grid technicians. It forms a set of reliable intelligent knowledge graph fault detection methods. Firstly, the substation fault recording data (single-phase grounding, relative phase short-circuit, valve short-circuit) is converted into two-dimensional time-frequency images using wavelet transform. Then, data enhancement technology is used to expand the fault sample data. Finally, the GoogleNet network is employed for AC transmission system fault diagnosis. To verify the effectiveness of the proposed method, we compared its performance with traditional methods such as KNN and decision tree. The test results demonstrate that the classification accuracy of GoogleNet on the training set is 93%, and the classification accuracy on the test set is 82%. These results suggest that the proposed method is an effective solution for AC transmission system fault diagnosis. The proposed intelligent fault detection method based on machine learning can effectively identify faults in AC transmission systems. It combines the experience of power grid technicians with an advanced machine learning algorithm to form a reliable and efficient fault detection system. The experimental results demonstrate the high reliability and good effectiveness of GoogleNet in AC transmission system fault diagnosis. Therefore, this method has the potential to improve power system reliability and enhance power quality.