Fault Feature Extraction Method of Reciprocating Compressor Valve Based on SPA-MF

Li, Ying; Pan, Yunjie; Ba, Peng; Wu, Shihu; Chen, Jiawen

doi:10.1109/access.2022.3226512

Cited by 4 publications

(3 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Additionally, factors such as oil film, gaps, and friction during signal transmission contribute to the non-stationary and nonlinear characteristics observed in collected vibration signals [5]. Consequently, the characteristic frequency and amplitude of vibration signals related to reciprocating mechanical faults exhibit time-varying behavior, making the extraction of fault characteristics challenging [6,7].…”

Section: Introductionmentioning

confidence: 99%

Improved lightweight federated learning network for fault feature extraction of reciprocating machinery

Zhang,

Duan,

et al. 2024

Meas. Sci. Technol.

View full text Add to dashboard Cite

he working environment of reciprocating machinery is complex, characterized by nonlinear and non-stationary signals. Deep learning can solve the above problems, but it has its own problems such as complex model and large amount of parameters. Additionally, privacy considerations among enterprises prevent data sharing, leading to the emergence of "data islands" and inadequate training of deep learning models. Based on the above analysis, this paper proposes a reciprocating mechanical feature extraction method based on an improved federated lightweight network. A lightweight network SqueezeNet model is used to solve the problems such as long training time of deep learning. By establishing a federated learning framework, the reciprocating mechanical data can be collectively diagnosed across various enterprises, thereby addressing the problem of limited model training caused by insufficient data. Furthermore, to enhance the accuracy of network training and diagnosis, modifications are made to the SqueezeNet network to reduce the number of model parameters while increasing the number and variety of feature extractions. Experimental results demonstrate that when the number of 1×1 and 3×3 channels is 1 to 7, the fault diagnosis accuracy is the highest, about 97.96%, which enriches the categories of feature extraction. The number of parameters in In-SqueezeNet is 56% of that in SqueezeNet network model, and the training time is reduced by nearly 15%. The fault diagnosis accuracy is increased from 95.1% to 97.3%, and the diversity of extracted features is increased. Compared with other network models such as ResNet, the improved lightweight federated learning network has a fault diagnosis accuracy of 96.6%, an improvement of 10.6%. At the same time, the training time was reduced to 1982s, a reduction of about 41.5%. The validity of the proposed model is further verified.

show abstract

Section: Introductionmentioning

confidence: 99%

Improved lightweight federated learning network for fault feature extraction of reciprocating machinery

Zhang,

Duan,

et al. 2024

Meas. Sci. Technol.

View full text Add to dashboard Cite

show abstract

“…Compared with the reciprocating compressor, the PMSM based single rolling rotor compressor is commonly used in the air conditioner system due to low vibration and fewer piston parts. In general, the compressor is sealed inside the compression cylinder filled with refrigerant and the hall sensors or encoders will fail to acquire the position or speed information [4]- [6]. Influenced by the eccentric placement of the rotor and the as shown in FIGURE 1, where the friction always exists between the slider and rotor during the compressor operation process.…”

Section: Introductionmentioning

confidence: 99%

Zero-Pole Cancellation Based Iterative Learning Control Method for Speed Ripple Suppression of Single Rolling Rotor Compressor

Wang,

Deng,

2024

IEEE Access

View full text Add to dashboard Cite

Owing to the severe load torque fluctuation, the speed ripple is a significant concern for the compressor drive system, especially within low frequency range. To address this issue, a zero-pole cancellation (ZPC) based iterative learning control (ILC) method is proposed for speed ripple suppression in this paper. The load torque characteristic is obtained based on the estimated speed harmonics. The essence of the speed ripple is investigated utilizing the compressor mechanical model, which attributes to the conjugate poles effect of the load torque term at the specific mechanical frequency. The zero point of the conventional ILC method is fixed at s=0, which can hardly reduce the dominant poles effect of the torque term. Afterwards, the poles' matching function is analytically derived to cancel out the dominant poles of the torque term, and then the specific order harmonic can be effectively suppressed. Considering the non-periodic disturbance caused by the operating frequency and temperature variation, the error correction law is designed to avoid the error accumulation and ensure the performance of proposed ILC method. Finally, the effectiveness of the method is validated by experiments through a 750W compressor drive system. Experimental results indicate that the total speed ripples reduce by nearly 60% and each harmonic component of the speed ripple can be effectively suppressed according to the FFT results.INDEX TERMS PMSM-compressor drive, Torque ripple, Iterative learning control, Speed harmonics.

show abstract

“…Deep learning techniques such as CNN and LSTM are used for dynamic characteristic prediction [7,8], along with digital image processing and artificial neural networks for fault detection [9,10]. However, as a research focus, adaptive signal extraction has the advantages of data adaptation, mode independence, high accuracy and easy implementation and is the current research focus of signal fault feature extraction [11].…”

Section: Introductionmentioning

confidence: 99%

A Double Interpolation and Mutation Interval Reconstruction LMD and Its Application in Fault Diagnosis of Reciprocating Compressor

Zhao

Liu³

et al. 2023

Applied Sciences

View full text Add to dashboard Cite

The accuracy and stability of the envelope estimation function are enduring issues throughout the research process of LMD. This paper presents double interpolation and mutation interval reconstruction local mean decomposition (DIMIRLMD) to improve the stability of the demodulation process and the accuracy of PF components. DIMIRLMD first proposes a mutation interval reconstruction envelope algorithm using extreme symmetry points to suppress the demodulation mutation phenomenon, which disturbs the stability of the demodulation process, and then selects the optimal PF component from a double interpolation PF component library based on the index of orthogonality (IO) for a better hierarchical property. DIMIRLMD was employed to analyze the simulation signal and vibration signal of a reciprocating compressor in an oversized bearing clearance state, and the results illustrate its performances are more excellent than those of three other LMD methods. Furthermore, the envelope frequency spectrum obtained from the proposed LMD presents a clear double rotation fault frequency and lower noise disturbance.

show abstract

Fault Feature Extraction Method of Reciprocating Compressor Valve Based on SPA-MF

Cited by 4 publications

References 26 publications

Improved lightweight federated learning network for fault feature extraction of reciprocating machinery

Improved lightweight federated learning network for fault feature extraction of reciprocating machinery

Zero-Pole Cancellation Based Iterative Learning Control Method for Speed Ripple Suppression of Single Rolling Rotor Compressor

A Double Interpolation and Mutation Interval Reconstruction LMD and Its Application in Fault Diagnosis of Reciprocating Compressor

Contact Info

Product

Resources

About