Diagnosis of unbalanced rolling bearing fault sample based on adaptive sparse contrative Auto-encoder and IGWO-USELM

Song, Haiquan; Ma, Wengang; Han, Zhonghe; Zhu, Xiaofan

doi:10.1016/j.measurement.2022.111353

Cited by 13 publications

(2 citation statements)

References 57 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Yi et al (2021) equalized the damage on the inner raceway to the isosceles trapezoid groove and studied the vibration signal in the double rotor system with damaged bearing. Song et al (2022) proposed an algorithm to extract the bearing damage signal from the measured data and proved that the algorithm has the ability to extract the time–frequency characteristics of the damage. Miao et al (2021) measured the vibration displacement signal of the bearing system and analyzed the amplitude frequency characteristic and nonlinear response.…”

Section: Introductionmentioning

confidence: 99%

Vibration analysis of shaft-cylindrical roller-bearing-pedestal system considering damage by Runge–Kutta algorithm

Haozhe

et al. 2023

Journal of Vibration and Control

View full text Add to dashboard Cite

The raceway damage of the bearing restricts the operation stability of the system. Therefore, in the case of damage, simulating vibration response and predicting the performance of the shaft-cylindrical roller-bearing-pedestal system have high research value. According to the structural characteristics and the force transfer relationship, a dynamic model of the shaft-cylindrical roller-bearing-pedestal system is established in this paper. Then, the vibration response of the system is simulated by Runge–Kutta algorithm and measured by experiment. Finally, the effects of load and speed, damage size, and radial clearance on the dynamic performance are investigated. The simulation and measurement results indicate that the load F and the speed N can promote the enhancement of the vibration acceleration signal. As the damage length continues to increase, the ring in the Poincaré map begins to break, and the system transits from quasi-periodic to chaotic motion. The vibration signal is more affected by the damage size than by the damage location. When the radial clearance increases from 5 × 10−6 m to 15 × 10−6 m, the PTP value increases from 23.83 m·s−2 to 61.36 m·s−2.

show abstract

Section: Introductionmentioning

confidence: 99%

Vibration analysis of shaft-cylindrical roller-bearing-pedestal system considering damage by Runge–Kutta algorithm

Haozhe

et al. 2023

Journal of Vibration and Control

View full text Add to dashboard Cite

show abstract

“…Qian et al [24] used the genetic algorithm (GA) to seek the hyper-parameters of a long shortterm memory network. Song et al [25] proposed an adaptive sparse shrinkage automatic encoder and an improved grey wolf optimized unsupervised ELM model to realize real-time fault diagnosis of unbalanced bearing data sets.…”

Section: Introductionmentioning

confidence: 99%

A method for diagnosing rolling bearing faults based on SDAE-ADHKELM

Li¹,

Jin²

2022

Meas. Sci. Technol.

View full text Add to dashboard Cite

Aiming to extract useful features from bearing signal for fault classification, an intelligent fault diagnosis method is proposed with stacked denoising auto-encoder (SDAE) and adaptive deep hybrid kernel extreme learning machine (ADHKELM). The deep network architecture of SDAE is used automatically extract deeply important features, a new hybrid kernel extreme learning machine (HKELM) is constructed by combining polynomial with wavelet kernel function to overcome a single kernel function being not universal. After that a deep hybrid kernel extreme learning machine (DHKELM) from stacking multiple HKELMs, and the sparrow search algorithm (SSA) is introduced to iteratively determine the optimal value of core hyper-parameters combinations of the DHKELM to generate the final fault classifier ADHKELM to enhance the performance of the model. Two experimental verification results show that SDAE-ADHKELM has better fault classification precision, robustness and generalizability than other related methods.

show abstract

Corrosion fatigue life prediction method of aluminum alloys based on back-propagation neural network optimized by Improved Grey Wolf optimization algorithm

Ji,

Li,

et al. 2024

J Mater Sci

View full text Add to dashboard Cite

Diagnosis of unbalanced rolling bearing fault sample based on adaptive sparse contrative Auto-encoder and IGWO-USELM

Cited by 13 publications

References 57 publications

Vibration analysis of shaft-cylindrical roller-bearing-pedestal system considering damage by Runge–Kutta algorithm

Vibration analysis of shaft-cylindrical roller-bearing-pedestal system considering damage by Runge–Kutta algorithm

A method for diagnosing rolling bearing faults based on SDAE-ADHKELM

Corrosion fatigue life prediction method of aluminum alloys based on back-propagation neural network optimized by Improved Grey Wolf optimization algorithm

Contact Info

Product

Resources

About