Denoising algorithm for mechanical vibration signal using quantum Hadamard transformation

• Quantum theory is employed to analyze the mechanical fault signal.
• The signal is quantized, thus the signal is expressed by quantum bit.
• Noise is reduced by quantum Hadamrd probability with adaptive thresholding value.

Vibration signal processing has been an effective way of monitoring mechanical equipment for many years. However, mechanical vibration signal are usually masked by significant background noise, which have motivated many studies into developing denoising methods. In this paper, the quantum theory is introduced for proposing a quantum Hadamard transformation based denosing algorithm (QHTDA) used for mechanical vibration signal under fault condition. Compared with other signal processing methods, QHTDA offers the advantage of superposition analysis by effectively quantizing vibration signal and computing the quantum Hadamard probability (QHP) of the fault information and noise information. In doing so, denoising formula can be related to the dynamic information of nonlinear and nonstationary signal, rather than a purely mathematical operation. Therefore, the proposed algorithm has potential to greatly facilitate noise suppression of target signals. Experimental results demonstrate that QHTDA can successfully reduce noise of the measured signals during the bearing fault diagnosis task.