Support Vector Machine-based Denoising Technique for Removal of White Noise in Partial Discharge Signal

Velayutham, Maheswari Ramasamy; Perumal, S. Arumuga; Basharan, Vigneshwaran; Silluvairaj, Willjuice Iruthayarajan Maria

doi:10.1080/15325008.2014.943826

Cited by 7 publications

(1 citation statement)

References 31 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Many signal processing techniques have been developed and applied in the context of PD signal denoising [7]. For example, the wavelet transform (WT) algorithm [8,9], support vector machine algorithm [10], spatial correlation denoising arithmetic [4] and so on have been employed to reduce the white noise. In addition, some well-known algorithms have also been used to suppress the narrow-band interference, such as wavelet packet transform [11], ensemble empirical mode decomposition (EEMD) [12].…”

Section: Introductionmentioning

confidence: 99%

Denoising of UHF PD signals based on optimised VMD and wavelet transform

Long

Wang

Dai

et al. 2017

IET Science, Measurement & Technology

View full text Add to dashboard Cite

Section: Introductionmentioning

confidence: 99%

Denoising of UHF PD signals based on optimised VMD and wavelet transform

Long

Wang

Dai

et al. 2017

IET Science, Measurement & Technology

View full text Add to dashboard Cite

Review of UHF-Based Signal Processing Approaches for Partial Discharge Detection

Schubert

Palo

Schlechter

2018

Computer Aided Systems Theory – EUROCAST 2017

View full text Add to dashboard Cite

Genetic algorithm based automated threshold estimation in translation invariant wavelet transform for denoising PD signal

Maheswari¹,

Vigneshwaran²,

Kalaivani³

2015

COMPEL: The International Journal for Computation and Mathematics in Electrical and Electronic Engineering

View full text Add to dashboard Cite

Purpose – The purpose of this paper is to investigate the condition of insulation in high-voltage equipments using partial discharge (PD) measurements. It proposes the methods to eliminate several noises like white noise, random noise and discrete spectral interferences which severely pollutes the PD signals. The study aims to remove these noises from the PD signal effectively by preserving the signal features. Design/methodology/approach – This paper employs fast Fourier transform, discrete wavelet transform and translational invariant wavelet transform (TIWT) for denoising the PD signals. The simulated damped exponential pulse and damped oscillatory pulse with low- and high-level noises and a measured PD signal are considered for this analysis. The conventional wavelet denoising approach is also improved by estimating the automated global optimum threshold value using genetic algorithm (GA). The statistical parameters are evaluated and compared. Among these methods, GA-based TIWT approach provides robustness and reduces computational complexity. Findings – This paper provides effective condition monitoring of power apparatus using GA-based TIWT approach. This method provides the low value of mean square error, pulse amplitude distortion and also high reduction in noise level due to its robustness and reduced computational complexity. It suggests that this approach works well for both signals immersed in noise as well as for noise immersed in signals. Research limitations/implications – Because of the chosen PD signals, the research results may lack for multiple discharges. Therefore, researchers are encouraged to test the proposed propositions further. Practical implications – The paper includes implication for the development of online testing for equipment analysis and diagnostics during normal operating condition. Corrective actions can be planned and implemented, resulting in reduced unscheduled downtime. Social implications – This PD-based analysis often present well in advance of insulation failure, asset managers can monitor it over time and make informed strategic decisions regarding the repair or replacement of the equipment. These predictive diagnostics help society to prioritize investments before an unexpected outage occurs. Originality/value – This paper provides an enhanced study of condition monitoring of HV power apparatus by which life time of insulation can be increased by taking preventive measures.

show abstract

Support Vector Machine-based Denoising Technique for Removal of White Noise in Partial Discharge Signal

Cited by 7 publications

References 31 publications

Denoising of UHF PD signals based on optimised VMD and wavelet transform

Denoising of UHF PD signals based on optimised VMD and wavelet transform

Review of UHF-Based Signal Processing Approaches for Partial Discharge Detection

Genetic algorithm based automated threshold estimation in translation invariant wavelet transform for denoising PD signal

Contact Info

Product

Resources

About