Model Parameter Transfer for Gear Fault Diagnosis under Varying Working Conditions

Abstract Gear fault diagnosis technologies have received rapid development and been effectively implemented in many engineering applications. However, the various working conditions would degrade diagnostic performance make gear (GFD) more challenging. In this paper, a novel model parameter transfer (NMPT) is proposed to boost of GFD under varying conditions. Based on previous strategy that controls empirical risk source domain, method further integrates superiorities multi-task learning with idea (TL) acquire transferable knowledge by minimizing discrepancies separating hyperplanes between one specific condition (target domain) another (source domain), then transferring both commonality specialty parameters over tasks use domain samples assist target task when sufficient labeled from are unavailable. For NMPT implementation, insufficient features abundant supervised information fed into train robust classifier for task. Related experiments prove expected be valuable technology practical The methods provides learning-based framework handle problem training caused variable operation