Search

The effective fault feature extraction is the core of rolling bearing fault diagnosis. However, rolling bearings usually operate in normal state and fault duration is very short, which will cause imbalance in fault diagnosis data, thus leading to difficulty in fault feature extraction and low diagnosis accuracy. Meanwhile, mutual interference between multiple fault responses will also lead to poor diagnosis performance. To solve these issues, a novel compound fault diagnosis method with imbalanced data based on frequency-domain Gramian angular field (FGAF) and convolutional neural networks optimized by instance normalization and efficient channel attention (IECNN) is proposed. Firstly, FGAF is adopted to map frequency-domain features of fault signals to the polar coordinate to obtain 2D FGAF feature spectrum. Secondly, an instance normalization module is established to reduce internal covariant shift caused by data distribution discrepancy and improve generalization ability. An efficient channel attention module is constructed to further excavate fault features and improve anti-interference ability. Finally, experiments are conducted under imbalanced dataset and imbalance intensified dataset, and the average accuracy of 99.91% and 99.92% were obtained, respectively, which shows the proposed method has better resistance to data imbalance. [ABSTRACT FROM AUTHOR]

The remaining useful life prediction of rolling bearing is vital in safety and reliability guarantee. In engineering scenarios, only a small amount of bearing performance degradation data can be obtained through accelerated life testing. In the absence of lifetime data, the hidden long-term correlation between performance degradation data is challenging to mine effectively, which is the main factor that restricts the prediction precision and engineering application of the residual life prediction method. To address this problem, a novel method based on the multi-layer perception neural network and bidirectional long short-term memory network is proposed. Firstly, a nonlinear health indicator (HI) calculation method based on kernel principal component analysis (KPCA) and exponential weighted moving average (EWMA) is designed. Then, using the raw vibration data and HI, a multi-layer perceptron (MLP) neural network is trained to further calculate the HI of the online bearing in real time. Furthermore, The bidirectional long short-term memory model (BiLSTM) optimized by particle swarm optimization (PSO) is used to mine the time series features of HI and predict the remaining service life. Performance verification experiments and comparative experiments are carried out on the XJTU-SY bearing open dataset. The research results indicate that this method has an excellent ability to predict future HI and remaining life. [ABSTRACT FROM AUTHOR]

Strain monitoring of high-speed train crossbeams is essential because as a vital load-bearing structure carrying several tonnes of under-vehicle equipment for high-speed trains, they are likely to be damaged under prolonged loading. This paper proposed a strain field reconstruction and monitoring method based on load-strain linear superposition. Firstly, finite element analysis was performed with two loads and three loads showing that the algorithm reconstructed the strain field of the crossbeam on a high-speed train with an average relative error of less than 0.01%. Then a sensing network with 24 fiber Bragg grating sensors was established. Train line operation tests were conducted and strain measurement information was obtained. On this basis, the strain field of the crossbeam was reconstructed with an average fitting error of less than 2%. This research shows the power of the load-strain linear superposition method in the strain field monitoring of the load-bearing structure for the operation and maintenance of the train. [ABSTRACT FROM AUTHOR]

A damage identification system of carbon fiber reinforced plastics (CFRP) structures is investigated using fiber Bragg grating (FBG) sensors and back propagation (BP) neural network. FBG sensors are applied to construct the sensing network to detect the structural dynamic response signals generated by active actuation. The damage identification model is built based on the BP neural network. The dynamic signal characteristics extracted by the Fourier transform are the inputs, and the damage states are the outputs of the model. Besides, damages are simulated by placing lumped masses with different weights instead of inducing real damages, which is confirmed to be feasible by finite element analysis (FEA). At last, the damage identification system is verified on a CFRP plate with 300 mm × 300 mm experimental area, with the accurate identification of varied damage states. The system provides a practical way for CFRP structural damage identification. [ABSTRACT FROM AUTHOR]

A label-free immunosensor based on optical fiber Fabry–Perot interferometer (FPI) is proposed. The FPI is fabricated by splicing both ends of a short section of hollow-core photonic crystal fiber to single mode fibers (SMFs) and cleaving the SMF pigtail to proper length. Then, the end surface of sensor is silanization modified and immobilized with goat anti-rabbit immunoglobulin G (IgG) by covalent binding to detect the specific rabbit IgG. The antibody immobilization and antigen binding will change the thickness of sensing layer, leading to the interference spectrum change of the FPI sensor. The responses of the sensor to immune reaction are analyzed theoretically and demonstrated experimentally. Results show that after antibody immobilization, the wavelength increases by 60 pm and the fringe contrast decreases by about 1 dB. And for antigen binding, the wavelength shows an obvious increase by 190 pm and the fringe contrast decreases by 2.15 dB. The wavelength shift during the antibody-antigen binding process conforms to the immune reaction kinetics. With the advantages of compact structure, good performance, capacity for real-time and label-free detection, easy fabrication, and low cost, the proposed optical fiber FPI sensor shows wide application prospect in label-free immunosensing. [ABSTRACT FROM PUBLISHER]

An optical fibre Fabry–Perot interferometer (FPI) sensor for relative humidity (RH) measurement is proposed. The FPI is formed by splicing a short section of hollow-core photonic crystal fibre(HCPCF) to single mode fibre and covering a chitosan film at the end of HCPCF. The refractive index of chitosan and film thickness will change with ambient RH, leading to the change in the reflected interference spectrum of FPI. RH response of the FPI sensor is analysed theoretically and demonstrated experimentally. It shows nonlinear response to RH values from 35 to 95%RH. The interference fringe shifts to shorter wavelength as RH increases with a maximum sensitivity of 0.28 nm/%RH at high RH level. And the fringe contrast also decreases as RH increases with an available maximum sensitivity of 0.5 dB/%RH. The sensor shows good stability and fast response time less than 1 min. With its advantages of compact structure, good performance, simple and safe fabrication, the proposed optical fibre FPI sensor has great potential for RH sensing. [ABSTRACT FROM PUBLISHER]

Some relations between the Hausdorff-dimensions and the entropies for expansive homeomorphisms on compact metric spaces are obtained. [ABSTRACT FROM AUTHOR]