Diagnosis of Transformer Faults Using the Multi-Class Adaboost Algorithm

Traditional shallow machine learning algorithms are incapable of exploring the link between fault data of oil-immersed transformers, resulting in low fault diagnostic accuracy. In answer to this challenge, this study provides a transformer defect diagnostic approach based on Multi-class AdaBoost Algorithms. First, the AdaBoost technique is merged with Support Vector Machines (SVM). The SVM is improved by the AdaBoost algorithm, and the transformer defect data is thoroughly investigated.The dynamic weight is then introduced into the Particle Swarm Optimization (PSO); by realtime updating of the particle inertia weight, the particle swarm optimization algorithm's search accuracy and optimization speed are improved, and the improved particle swarm optimization algorithm (IPSO) is used to optimize the parameters of the SVM. Finally, the uncoded ratio technique creates a new gas group collaboration by studying the link between the dissolved gas in the transformer oil and the fault type.As the input feature vector, the better ratio technique is built. Based on simulations of 117 sets of IECTC10 standard data and 419 sets of transformer fault data collected in China, the diagnosis method proposed in this paper has a strong search ability, a fast convergence speed, and a significant improvement in diagnostic accuracy when compared to traditional methods.