Statistical Characteristics Clustering Based Adaptive Debris Signature Identification Method for Inductive Sensors with Dual Phase

Abstract Wear monitoring plays an important role in the early warning of mechanical equipment failures and in predicting operational life. Inductive sensors provide data support for wear analysis by monitoring and extracting key features of oil debris online. However, the low identification accuracy of tiny metal particles under complex interference remains a critical factor that limits detection sensitivity. The sensors with dual probes utilize a double-induction structure to enhance noise reduction and debris perception through correlation analysis. Nevertheless, the peformance of debris signature identification still face challenges related to dependence on prior knowledge, insufficiant sensitivity, destruction to features and weak generalizability. In this study, we propose a novel debris signatures identification method, names GIM-SCC. By constructing a global independence metric (GIM), time series samples are transformed into characterization vectors and debris identification is then achieved by a statistical characteristic clustering (SCC). Through numerical simulations and experiments, we demonstrate the advantages of this method in terms of signature identification accuracy, robustness, feature protection ability and generalization capability through algorithm comparison. This contribution is expected to provide reliable technical support for the accurate extraction of debris sigantures via inductive sensors with dual probes.