Fault Diagnosis of Transmission Lines Based on INGO- VMD and ZOA-SVM
Yao Ma, Yaosong Xu · 2024
Aiming at the problem of the low diagnosis and identification rate of short-circuit failure in transmission lines, this paper presents an innovative transmission line fault diagnosis technique that incorporates a modified Northern Eagle optimization algorithm, Variable Divided Modal Decomposition (VMD), and Support Vector Machines, which work together to improve diagnostic accuracy. A solution to the problem that VMD parameters need to be pre-set is proposed in this paper, and the improved Northern Pale Eagle optimization algorithm is used to optimize and derive the best combination of parameters. Then the optimized VMD is used to decompose the fault voltage signal of the transmission line, and the envelope entropy value of the IMF component is obtained, which is composed of fault feature vectors. Then the support vector machine fault identification model is optimized using the improved Northern Eagle optimization algorithm. Finally, the experimental comparison with CNN, LSTM, and SVM diagnostic models shows that the identification precision of the fault diagnostic matrix suggested in the file is higher than that of the previous three, and the training speed is faster, with an accuracy rate as high as 95%, reflecting the advantageous of the proposed method.