An Adaptive Incremental Learning Algorithm based on Shared Nearest Neighbors in Fault Detection

Incremental learning technology has been widely used in the field of fault diagnosis. Traditional incremental learning model frequent update to avoid losing knowledge of new samples. In this paper, we propose a adaptive incremental learning algorithm based on shared nearest neighbors (AIL-SNN), which controls the incremental update by detecting the concept drift. The algorithm in this paper is verified by experiments in CWRU dataset. The experimental results show that the algorithm can effectively detect the sample concept drift in the sample, and can use a small amount of training time to obtain better fault diagnosis accuracy.