Fault Diagnosis of Rotating Machine Using an Indirect Observer and Machine Learning

TayebiHaghighi, Shahnaz; Koo, Insoo

doi:10.1109/ictc49870.2020.9289590

Cited by 6 publications

(8 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Thus, the main challenge of nonlinear modern control-based observers (such as the sliding mode observer and the feedback linearization observer) is complexity. To reduce complexity, modern linear control-based observers, such as the proportional integral (PI) observer [22] and the proportional multi-integral (PMI) observer [34], have been suggested. Although PI observers and PMI observers have good performance in terms of accuracy and complexity, they have some problems related to robustness about uncertainties [34].…”

Section: Introductionmentioning

confidence: 99%

“…To reduce complexity, modern linear control-based observers, such as the proportional integral (PI) observer [22] and the proportional multi-integral (PMI) observer [34], have been suggested. Although PI observers and PMI observers have good performance in terms of accuracy and complexity, they have some problems related to robustness about uncertainties [34]. To address the robustness issue, a combination of PMI observer and sliding mode approach was used in [34].…”

Section: Introductionmentioning

confidence: 99%

“…Although PI observers and PMI observers have good performance in terms of accuracy and complexity, they have some problems related to robustness about uncertainties [34]. To address the robustness issue, a combination of PMI observer and sliding mode approach was used in [34]. The main limitation of this algorithm is to increase the amount of chattering and signal fluctuations in the presence of variation in torque load and motor speed.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

SVM-Based Bearing Anomaly Identification with Self-Tuning Network-Fuzzy Robust Proportional Multi Integral and Smart Autoregressive Model

TayebiHaghighi

Koo

2021

Applied Sciences

Self Cite

View full text Add to dashboard Cite

In this paper, the combination of an indirect self-tuning observer, smart signal modeling, and machine learning-based classification is proposed for rolling element bearing (REB) anomaly identification. The proposed scheme has three main stages. In the first stage, the original signal is resampled, and the root mean square (RMS) signal is extracted from it. In the second stage, the normal resampled RMS signal is approximated using the AutoRegressive with eXternal Uncertainty (ARXU) technique. Moreover, the nonlinearity of the bearing signal is solved using the combination of the ARXU and the machine learning-based regression, which is called AMRXU. After signal modeling by AMRXU, the RMS resampled signal is estimated using a combination of the proportional multi-integral (PMI) technique, the variable structure (VS) Lyapunov technique, and a self-tuning network-fuzzy system (SNFS). Finally, in the third stage, the difference between the original signal and the estimated one is calculated to generate the residual signal. A machine learning-based classification technique is utilized to classify the residual signal. The Case Western Reserve University (CWRU) dataset is used to evaluate anomaly identification performance of the proposed scheme. Regarding the experimental results, the average accuracy for REB crack identification is 98.65%, 97.7%, 97.35%, and 97.67%, respectively, when the motor torque loads are 0-hp, 1-hp, 2-hp, and 3-hp.

show abstract

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

SVM-Based Bearing Anomaly Identification with Self-Tuning Network-Fuzzy Robust Proportional Multi Integral and Smart Autoregressive Model

TayebiHaghighi

Koo

2021

Applied Sciences

Self Cite

View full text Add to dashboard Cite

show abstract

“…To bearing crack discrimination via the hybrid method, the base step is the computation of residual signals. According to the difference between original signals and estimated signals, residual signals will be calculated [5,6]. Among different algorithms for estimation the main signal, an observation-based technique has been suggested in this research.…”

Section: Introductionmentioning

confidence: 99%

“…For signal estimation, various techniques have been applied by researchers that can be classified into two main groups: linear-based signal estimation and nonlinear-based signal estimation. In this work, linear-based observers such as proportional integral observer and proportional multi-integral observer are suggested for signal estimation [5,6].…”

Section: Introductionmentioning

confidence: 99%

New Fuzzy Observer Fault Pattern Detection by NARX-Laguerre Model Applied to the Rotating Machine

TayebiHaghighi

Koo

2021

Lecture Notes in Networks and Systems

Self Cite

View full text Add to dashboard Cite

A rotating machine is a ubiquitous product in industries. Inner, outer, and roller cracking in bearings can cause structural failure of the rotating machines and possibly decrease the service life of different industrial systems such as motors. The nonlinearity and complexity associated with the uncertain and unknown behavior inherent in rotating machines lead to difficulty in the detection and identification of bearings' cracks in real-time. To address these issues, the combination of indirect Proportional-Integral-Derivative (PID) observer and fuzzy logic approach is proposed in the paper to detect the fault pattern in the bearing. To do this, firstly the indirect fuzzy observer is modeled by the nonlinear autoregressive with Laguerre filter that is improved by the fuzzy technique. After signal estimation using indirect fuzzy observer, the difference (residual signals) between original and estimated signals are computed. Finally, the machine learning technique is utilized for residual signal pattern detection. The Case Western Reverse University Bearing Dataset (CWRUBD) is used to test the proposed scheme.

show abstract

Cross Diversity Entropy-Based Feature Extraction for Fault Diagnosis of Rotor System

Li,

Jiao,

Wang

et al. 2024

IEEE/ASME Trans. Mechatron.

View full text Add to dashboard Cite

Fault Diagnosis of Rotating Machine Using an Indirect Observer and Machine Learning

Cited by 6 publications

References 16 publications

SVM-Based Bearing Anomaly Identification with Self-Tuning Network-Fuzzy Robust Proportional Multi Integral and Smart Autoregressive Model

SVM-Based Bearing Anomaly Identification with Self-Tuning Network-Fuzzy Robust Proportional Multi Integral and Smart Autoregressive Model

New Fuzzy Observer Fault Pattern Detection by NARX-Laguerre Model Applied to the Rotating Machine

Cross Diversity Entropy-Based Feature Extraction for Fault Diagnosis of Rotor System

Contact Info

Product

Resources

About