DIAGNOSIS OF RAILWAY GEARBOX FAULTS BASED ON WAVELET DE-NOISING - CHARACTERISTIC-SCALE DECOMPOSITION - PARTICLE SWARM OPTIMIZATION - LEAST SQUARES SUPPORT VECTOR MACHINE

摘要

This paper proposes a combined method involving wavelet de-noising, local characteristic-scale decomposition, particle swarm optimization, and least squares support vector machine for diagnosis of railway gearbox faults. In this method, the original signal is first de-noised by wavelet de-noising, and local characteristic scale decomposition is applied to decompose the de-noised signals to intrinsic scale components. Secondly, the intrinsic scale component energy-torques are extracted as feature parameters. Finally, the extracted features are used as inputs for classification by a least squares support vector machine using particle swarm optimization to optimize its parameters. The vibration signals from a gearbox test rig are used to experimentally verify the effectiveness of the proposed method. The results show that the novel method is accurate for the diagnosis of railway gearbox faults.