Transient feature extraction of encoder signal for condition assessment of planetary gearboxes with variable rotational speed

This paper proposes a transient feature extraction algorithm of encoder signal for the condition assessment of planetary gearboxes under variable speed condition. In the proposed method, local polynomial fitting and sparsity based algorithm (LSA) is first constructed to extract time domain transient features from raw encoder signal and then order tracking (OT) technique is applied to convert non-stationary transient features in time domain into stationary ones in angular domain. Meanwhile, to solve the formulated local polynomial fitting and sparsity based optimization problem, a fast convergence algorithm is derived based on alternating direction method of multipliers and majorization-minimization. Additionally, an adaptive parameter selection strategy is introduced for automatically selecting appropriate parameters of the proposed method LSA-OT. With the proposed method LSA-OT, the transient features can be effectively extracted and the condition assessment of planetary gearboxes can be easily achieved. The effectiveness of LSA-OT is confirmed by simulation evaluation and case study.