Skip to Main Content
Adaptive immune algorithm based data mining (AIA-data mining) is presented for fault diagnosis of power transformer. The information entropy is used for the production of the initial population, which leads to convergence speed of the algorithm to be faster than that of the initial population produced by random. On the basis of that, the bi-level search mechanism of the AIA further speeds up extraction of the decision-making table for the transformer fault diagnosis from the samples. Results from examples show that the method proposed is effective and feasible.