Maximum correlation Pearson correlation coefficient deconvolution and its application in fault diagnosis of rolling bearings

Qin, Limu; Yang, Gang; Sun, Qi

doi:10.1016/j.measurement.2022.112162

Cited by 29 publications

(10 citation statements)

References 60 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Based on the input variables, the ANN learns to predict the output variable, and the weights may be used to detect the strength and direction of the correlations between parameters. Pearson's correlation matrix is used in this study to conduct a correlation analysis between the predictive parameters of the network and the dependent parameters 45 . In other words, Correlation analysis employing Pearson correlation on a developed BRNN is used in this study to examine the connection between the BRNN model's forecasts and the actual target values.…”

Section: Methodsmentioning

confidence: 99%

A novel neural network-based framework to estimate oil and gas pipelines life with missing input parameters

Shaik,

Jongkittinarukorn,

Benjapolakul

et al. 2024

Sci Rep

View full text Add to dashboard Cite

Dry gas pipelines can encounter various operational, technical, and environmental issues, such as corrosion, leaks, spills, restrictions, and cyber threats. To address these difficulties, proactive maintenance and management and a new technological strategy are needed to increase safety, reliability, and efficiency. A novel neural network model for forecasting the life of a dry gas pipeline system and detecting the metal loss dimension class that is exposed to a harsh environment is presented in this study to handle the missing data. The proposed strategy blends the strength of deep learning techniques with industry-specific expertise. The main advantage of this study is to predict the pipeline life with a significant advantage of predicting the dimension classification of metal loss simultaneously employing a Bayesian regularization-based neural network framework when there are missing inputs in the datasets. The proposed intelligent model, trained on four pipeline datasets of a dry gas pipeline system, can predict the health condition of pipelines with high accuracy, even if there are missing parameters in the dataset. The proposed model using neural network technology generated satisfactory results in terms of numerical performance, with MSE and R2 values closer to 0 and 1, respectively. A few cases with missing input data are carried out, and the missing data is forecasted for each case. Then, a model is developed to predict the life condition of pipelines with the predicted missing input variables. The findings reveal that the model has the potential for real-world applications in the oil and gas sector for estimating the health condition of pipelines, even if there are missing input parameters. Additionally, multi-model comparative analysis and sensitivity analysis are incorporated, offering an extensive comprehension of multi-model prediction abilities and beneficial insights into the impact of various input variables on model outputs, thereby improving the interpretability and reliability of our results. The proposed framework could help business plans by lowering the chance of severe accidents and environmental harm with better safety and reliability.

show abstract

Section: Methodsmentioning

confidence: 99%

A novel neural network-based framework to estimate oil and gas pipelines life with missing input parameters

Shaik,

Jongkittinarukorn,

Benjapolakul

et al. 2024

Sci Rep

View full text Add to dashboard Cite

show abstract

“…MCPCCD is a new BD method under the MCKD framework, which solves the optimal filter by maximizing the objective function value of the reconstructed signal using the coupled equation of the correlation Pearson correlation coefficient and the signal fidelity term as the objective function [25]. The objective function of MCPCCD is…”

Section: Mcpccd Methodsmentioning

confidence: 99%

“…In the framework of MCKD and MED, McDonald and Zhao [21] further proposed the multipoint optimal minimum entropy deconvolution adjusted (MOMEDA) method without iterative solving. Based on the MCKD framework, Li et al [22] proposed the maximum correntropy criterion-based blind deconvolution (MCCBD) that can adaptively determine the fault pulse period, Liang et al [23] proposed the maximum average kurtosis deconvolution (MAKD) method that can detect non-uniform rotating mechanical faults, Miao et al [24] proposed the sparse maximum harmonic-to-noise ratio deconvolution method that can adaptively select the fault pulse period, and Qin et al [25] proposed the maximum correlation Pearson correlation coefficient deconvolution method (MCPCCD) based on a novel design norm, which does not change the phase frequency characteristics and waveform characteristics of the reconstructed signal. With the development of BD methods, Buzzoni et al [26] designed the maximum second-order cyclostationarity blind deconvolution (CYCBD), a new framework different from MCKD, by using the second-order cyclostationarity index and Rayleigh quotient.…”

Section: Introductionmentioning

confidence: 99%

Multi-objective adaptive guided differential evaluation blind deconvolution and its application in bearing fault detection

Qin

Yang

Sun

et al. 2023

Meas. Sci. Technol.

Self Cite

View full text Add to dashboard Cite

Blind deconvolution (BD) methods applied to bearing fault detection often cause inferior performance due to inaccurate input parameters. Moreover, the optimal parameters of BD vary for different speeds and fault types of bearings, which seriously undermines the applicability of BD in practical industries. In this scenario, this paper proposes a parameter-adaptive BD method (MOBD) based on the multi-objective adaptive guided differential evaluation algorithm (MOAGDE). Firstly, based on the linear discriminant analysis, the quotient of inter-class distance and intra-class distance is used to determine the superiority of common bearing fault characteristic indicators to establish the multi-objective function of MOAGDE. Then, the optimal parameters of BD are searched by MOAGDE improved by dynamic switched crowding method (DSC-MOAGDE). Finally, the bearing is judged whether or what kind of fault has occurred according to the fault type locating index proposed in this paper. The main advantage of MOBD is that only bearing speed and type priories are required to achieve online detection of bearing faults. The results of simulation and experimental signals demonstrate that MOBD significantly outperforms the traditional BD method.

show abstract

“…Although this method ensures that the reconstructed signal can retain the characteristics of the original signal, the reconstruction accuracy will be disturbed by noise. Therefore, IMCPCCD is used to preprocess the input signal in a timely manner to improve the fidelity of the fault characteristic signal [49]. Furthermore, it can be seen from equation (9) that k and N are the key parameters affecting the validity of G-log.…”

Section: Adaptive Generalized Logarithm Sparse Regularization Based O...mentioning

confidence: 99%

“…MCPCCD is a new BD method with the correlation Pearson correlation coefficient as the objective function [49], whose main idea is to preserve the original signal characteristics as much as possible when separating the fault feature components; its objective function is as follows,…”

Section: Improved Maximum Correlation Pearson Correlation Coefficient...mentioning

confidence: 99%

An adaptive generalized logarithm sparse regularization method and its application in rolling bearing fault diagnosis

Qin¹,

Yang²,

Lv³

et al. 2022

Meas. Sci. Technol.

Self Cite

View full text Add to dashboard Cite

The generalized logarithm sparse regularization method (G-log) for fault diagnosis of rotating devices can effectively reconstruct repetitive transient shocks from noise-disturbed signals, but its reconstruction accuracy frequently becomes inferior due to unsuitable regularization parameters. Moreover, the conventional sparse regularization methods perform nothing on the input signals to guarantee that the impulse characteristics remain constant during the entire iteration process, which exacerbates the influence of noise on the reconstruction accuracy. To overcome these challenges, an adaptive generalized logarithm sparse regularization method (AG-log) based on the second order cyclostationary indicator (ICS2) and the improved maximum correlation Pearson correlation coefficient deconvolution method (IMCPCCD) is proposed in this paper. Firstly, the optimal threshold parameter k for each iteration of AG-log is determined based on the ICS2 criterion to ensure the optimal reconstruction accuracy, while the optimal combination of iterations number N and k is established. Secondly, the original signal and the IMCPCCD filtered signal are alternately used as the input signal of AG-log according to the parity of the iterative steps to reduce the interference of noise. Finally, the application on simulated and two engineering case signals demonstrates AG-log has better reconstruction accuracy compared with the conventional nonconvex sparse regularization methods.

show abstract

Maximum correlation Pearson correlation coefficient deconvolution and its application in fault diagnosis of rolling bearings

Cited by 29 publications

References 60 publications

A novel neural network-based framework to estimate oil and gas pipelines life with missing input parameters

A novel neural network-based framework to estimate oil and gas pipelines life with missing input parameters

Multi-objective adaptive guided differential evaluation blind deconvolution and its application in bearing fault detection

An adaptive generalized logarithm sparse regularization method and its application in rolling bearing fault diagnosis

Contact Info

Product

Resources

About