Research article Special Issues

PSO-MCKD-MFFResnet based fault diagnosis algorithm for hydropower units


  • Due to the coupling effect of external environmental noise and vibration noise, the feature rate of the original hydroelectric unit fault signal is not prominent, which will affect the performance of fault diagnosis algorithms. To solve the above problems, this paper proposes a PSO-MCKD-MFFResnet algorithm for fault diagnosis of hydropower units (Particle swarm optimization, PSO; Maximum correlation kurtosis deconvolution, MCKD; Multi-scale feature fusion residual network, MFFResnet). In practical applications, the selection of key parameters in the traditional MCKD method is heavily dependent on prior knowledge. First, this paper proposes a PSO-MCKD enhancement algorithm for fault features, which uses the PSO algorithm to search for the influencing parameters of MCKD to enhance the features from the original fault signal. Second, a fault feature diagnosis algorithm based on MFFResnet is proposed to improve the utilization of local features. The multi-scale residual module is used to extract features at different scales and then put the enhanced signal into MFFResnet for training and classification. The experimental results show that our approach can accurately and effectively classify the fault types of hydropower units, with an accuracy rate of 98.85. It is superior to other representative algorithms in different indicators and has a good stability.

    Citation: Xu Li, Zhuofei Xu, Yimin Wang. PSO-MCKD-MFFResnet based fault diagnosis algorithm for hydropower units[J]. Mathematical Biosciences and Engineering, 2023, 20(8): 14117-14135. doi: 10.3934/mbe.2023631

    Related Papers:

    [1] Prateebha Ramroop, Hudaa Neetoo . Antilisterial activity of Cymbopogon citratus on crabsticks. AIMS Microbiology, 2018, 4(1): 67-84. doi: 10.3934/microbiol.2018.1.67
    [2] Afraa Said Al-Adawi, Christine C. Gaylarde, Jan Sunner, Iwona B. Beech . Transfer of bacteria between stainless steel and chicken meat: A CLSM and DGGE study of biofilms. AIMS Microbiology, 2016, 2(3): 340-358. doi: 10.3934/microbiol.2016.3.340
    [3] John Samelis, Athanasia Kakouri . Hurdle factors minimizing growth of Listeria monocytogenes while counteracting in situ antilisterial effects of a novel nisin A-producing Lactococcus lactis subsp. cremoris costarter in thermized cheese milks. AIMS Microbiology, 2018, 4(1): 19-41. doi: 10.3934/microbiol.2018.1.19
    [4] Elena Kotenkova, Dagmara Bataeva, Mikhail Minaev, Elena Zaiko . Application of EvaGreen for the assessment of Listeria monocytogenes АТСС 13932 cell viability using flow cytometry. AIMS Microbiology, 2019, 5(1): 39-47. doi: 10.3934/microbiol.2019.1.39
    [5] Thomas Bintsis . Foodborne pathogens. AIMS Microbiology, 2017, 3(3): 529-563. doi: 10.3934/microbiol.2017.3.529
    [6] Agni Hadjilouka, Konstantinos Nikolidakis, Spiros Paramithiotis, Eleftherios H. Drosinos . Effect of co-culture with enterocinogenic E. faecium on L. monocytogenes key virulence gene expression. AIMS Microbiology, 2016, 2(3): 359-371. doi: 10.3934/microbiol.2016.3.359
    [7] Stephen H. Kasper, Ryan Hart, Magnus Bergkvist, Rabi A. Musah, Nathaniel C. Cady . Zein nanocapsules as a tool for surface passivation, drug delivery and biofilm prevention. AIMS Microbiology, 2016, 2(4): 422-433. doi: 10.3934/microbiol.2016.4.422
    [8] Tatyana V. Polyudova, Daria V. Eroshenko, Vladimir P. Korobov . Plasma, serum, albumin, and divalent metal ions inhibit the adhesion and the biofilm formation of Cutibacterium (Propionibacterium) acnes. AIMS Microbiology, 2018, 4(1): 165-172. doi: 10.3934/microbiol.2018.1.165
    [9] Itziar Chapartegui-González, María Lázaro-Díez, Santiago Redondo-Salvo, Elena Amaro-Prellezo, Estefanía Esteban-Rodríguez, José Ramos-Vivas . Biofilm formation in Hafnia alvei HUMV-5920, a human isolate. AIMS Microbiology, 2016, 2(4): 412-421. doi: 10.3934/microbiol.2016.4.412
    [10] Joseph O. Falkinham . Mycobacterium avium complex: Adherence as a way of life. AIMS Microbiology, 2018, 4(3): 428-438. doi: 10.3934/microbiol.2018.3.428
  • Due to the coupling effect of external environmental noise and vibration noise, the feature rate of the original hydroelectric unit fault signal is not prominent, which will affect the performance of fault diagnosis algorithms. To solve the above problems, this paper proposes a PSO-MCKD-MFFResnet algorithm for fault diagnosis of hydropower units (Particle swarm optimization, PSO; Maximum correlation kurtosis deconvolution, MCKD; Multi-scale feature fusion residual network, MFFResnet). In practical applications, the selection of key parameters in the traditional MCKD method is heavily dependent on prior knowledge. First, this paper proposes a PSO-MCKD enhancement algorithm for fault features, which uses the PSO algorithm to search for the influencing parameters of MCKD to enhance the features from the original fault signal. Second, a fault feature diagnosis algorithm based on MFFResnet is proposed to improve the utilization of local features. The multi-scale residual module is used to extract features at different scales and then put the enhanced signal into MFFResnet for training and classification. The experimental results show that our approach can accurately and effectively classify the fault types of hydropower units, with an accuracy rate of 98.85. It is superior to other representative algorithms in different indicators and has a good stability.



    1. Introduction

    Biofilms are living microbial communities attached to a solid surface or flowing in aqueous systems. Mature biofilms are highly organized ecosystems in which microbial cells are embedded in extracellular matrices with water channels that provide passages for the exchange of nutrients, metabolites and waste products [1]. Such biofilm structures can confer protection to bacterial cells and decrease the efficiency of cleaning and disinfection procedures [2]. Multispecies biofilms are the predominant forms of presence for microorganisms in the natural environments, and play important roles in the survival and persistence of microorganisms, including foodborne pathogens [3]. In such microbial communities, microorganisms interact in various ways that can be described as competition, antagonism, or synergism [4].

    Listeria monocytogenes is a Gram-positive foodborne pathogen that is responsible for serious infections in immunocompromised individuals and pregnant women [5]. It is frequently found in various food processing environments and has been isolated from a range of food products including meat, milk, cheese, and vegetables [6,7,8,9]. Over the last few decades, L. monocytogenes continued to be a major public health threat [10]. A recent large scale outbreak was caused by contaminated cantaloupes grown and packed in one farm in Colorado which caused 147 infections and at least 33 deaths in the USA [11]. Biofilm formation has been suggested an important factor for the persistence of L. monocytogenes in the environment and the contaminations of food processing facilities. However, L. monocytogenes cultivated alone in classic laboratory media, such as tryptone soy broth (TSB) or brain-heart infusion (BHI) broth, exhibit relatively low potential for forming biofilm [12,13].

    Ralstonia insidiosa is a Gram-negative, nonsporulating, aerobic, nonfermentative, motile rod. It has been isolated from the respiratory secretions of cystic fibrosis patients, river and pond water, soil, activated sludge [14] and has also been detected in water distribution systems [15]. R. insidiosa FC 1138 was isolated from a fresh-cut produce processing plant and has been shown a strong biofilm producer. When co-cultured with R. insidiosa, the presence E. coli O157:H7 in the biofilms significantly increase under various tested conditions [16]. In this study we evaluated the interactions of L. monocytogenes and R. insidiosa in biofilm formation during 24 h co-incubation. At this time period, although the biofilm is still considered in early stage, the interspecies interactions can be readily discerned.


    2. Materials and Methods


    2.1. Bacteria stains and growth conditions

    L. monocytogenes and R. insidiosa strains (Table 1) originally isolated from various food and other sources were from our laboratory collections. They were maintained at –80 ℃ in TSB (BD, Franklin Lakes, NJ) with 25% glycerol. Before all experiments, frozen cells were subcultured on tryptic soy agar plates (TSA; BD) for 24 h, and then one colony was transferred into TSB and grown to stationary phase. L. monocytogenes strain FS2025 was used as a representative in experiments of biofilm formation on stainless steel surface and in compartmentalized cultures.

    Table 1. Bacterial strains used in this study.
    Strain Serotype Isolation Source Collections
    R. insidiosa
    FC1138 NA Fresh produce facility EMFSL
    L. monocytogenes
    FS2005 1/2a Milk EMFSL
    FS2006 3b Milk EMFSL
    FS2007 1/2b Milk EMFSL
    FS2008 4b Milk EMFSL
    FS2009 4b Spinal fluid of child with meningitis ATCC
    FS2017 4b Beef and pork sausage EMFSL
    FS2018 1/2b Hard salami EMFSL
    FS2019 1/2a Frankfurter EMFSL
    FS2025 1/2b Cantaloupe NRRL
    FS2026 1/2a Cantaloupe NRRL
    EMFSL: Environmental Microbial and Food Safety Laboratory, USDA ARS; ATCC: American Type Culture Collections; NRRL: Northern Regional Research Laboratory, USDA ARS.
     | Show Table
    DownLoad: CSV

    2.2. Biomass quantification

    A static biofilm formation assay was performed in 96-well polystyrene plates (BD) as previously reported [17] with some modifications. Briefly, cells were cultured overnight in TSB. They were centrifuged at 4,500 g and re-suspended in 10% TSB. The OD600 value was adjusted to 0.5 and then diluted 100 fold in 10% TSB. Aliquots of 200 μL of this inoculated culture, containing either a single strain or a mixture of R. insidiosa and L. monocytogenes, were transferred to individual wells for biofilm formation. The plates were incubated at 30 ℃ for 24 h without shaking. After removal of the liquid culture, attached biofilms were stained with crystal violet. Excessive dye was removed and rinsed three times with PBS. The retained dye was dissolved in 33% acetic acid, and absorbance was measured at 570 nm to quantify total biomass.


    2.3. Compartmentalized cultures

    Cell culture plates (12-well) and matching insert compartments with 0.4 μm pore size polycarbonate membrane (Corning, NY, USA) that supported free diffusion of culture medium and metabolites were used to physically separate R. insidiosa and L. monocytogenes strains during biofilm formation. A bacterial suspension (1.5 ml in 10% TSB) of L. monocytogenes strain was added to each well of the plate, and an equal volume of the same medium with or without inoculum of R. insidiosa was added in the insert compartment on top of the wells. After incubation at 30 ℃ for 24 h, the inserts were removed. Biofilms in the base compartments were determined by crystal violet stain. A reciprocal format with L. monocytogenes in the insert compartment was similarly tested.


    2.4. Scanning electron microscope (SEM) of biofilms

    Type 304 stainless steel coupons with #4 brush finishing (30 mm × 15 mm × 2 mm, Remaly Manufacturing, Tamaqua, PA) were used for biofilm formation. Coupons were placed in a sterile deep-well micro-dilution block with 2 ml of bacteria suspension in each well, which allows the submersion of half of the coupon. They were incubated at 30 ℃ for 24 h without shaking. Then the coupons were washed three times with PBS and fixed with formaldehyde for 1 h. Biofilms structures were observed using a Hitachi S-4700 low temperature scanning electron microscope (Hitachi High Technologies America, Inc., Schaumburg, IL) after coating with platinum.


    2.5. Bacterial enumeration

    Enumeration of bacteria in biofilms formed on stainless steel coupons (75 mm × 25 mm × 2 mm, Remaly Manufacturing) was carried out as follows: The coupon was half submerged in 20 ml growth medium in a 50 ml Falcon tube and allowed static incubation at 30 ℃ for 24 h. The coupons were rinsed three times with 3 ml of sterile PBS. Cells remaining attached to the coupon surface were harvested by uni-directional scraping with a sterile cotton-headed stick and released into 10 ml sterile PBS by rigorous twirling and squeezing against vial side. The resulted samples were then vortexed vigorously for 1 min to facilitate cell dispersion. Cell suspension was 10-fold serially diluted and plated on TSA containing 1 g/L of 5-Bromo-4-chloro-3-indolyl β-D-glucopyranoside for enumerating L. monocytogenes (blue colonies) and R. insidiosa (white colonies).


    2.6. Statistical analysis

    A one-way analysis of variance (ANOVA) was performed for statistical analysis using the SPSS software version 19.0. Data were presented as the mean values ± SD (n = 3). Differences were considered significant when P < 0.05.


    3. Results and Discussion


    3.1. Enhanced biofilm formation in mixed culture

    Biofilm formation by 10 L. monocytogenes strains of varying serotypes and isolation origins were examined as monoculture and in mixed culture with R. insidiosa strain FC1138 (Figure 1). When cultured alone, each of the L. monocytogenes strains exhibited weak ability of biofilm formation in 10% TSB, as indicated by the minimal biomass accumulation. No difference was observed among the tested L. monocytogenes strains. Previous studies by Harvey et al [12] showed that L. monocytogenes strains had varying biofilm formations, although the majority was weak biofilm producers. This was also consistent with observations in our previous study [18] which showed that all L. monocytogenes strains tested were weak in biofilm formation in monocultures. In contrast, biomass accumulation in the mixed cultures of R. insidiosa with individual L. monocytogenes strains was significantly higher, indicating strong biofilm formation. Since R. insidiosa alone is a strong biofilm former, it could be expected that the high biomass accumulation was due to the strong biofilm formation by R. insidiosa. However, the biomass accumulation in the mixed culture greatly exceeded the sum of the two individual strains, indicating a synergistic interaction between L. monocytogenes and R. insidiosa in forming biofilms. This synergistic interaction was also previously observed between R. insidiosa and E. coli O157:H7 strain [18].

    Figure 1. Biofilm formation of L. monocytogenes strains in monocultures and in mixed cultures with R. insidiosa measured by the crystal violet staining assay.

    3.2. Cell contact dependent interactions

    Quorum sensing is a common mechanism of intra-and inter-species communications in microbial communities. This mechanism relies on small diffusible molecules, such as acy-homoserine lactone (AHL), oligopeptides, and autoinducer 2, as signal molecules [19]. To examine the possibility that the synergistic interaction in biofilm formation was affected by a mechanism akin to quorum sensing, R. insidiosa and L. monocytogenes cells were inoculated in two separate compartments linked through 0.4 µm pore size filter membrane that supported free diffusion of culture medium and metabolites. Biomass production were determined (Figure 2) after incubation at 30 ℃ for 24 h. No significant difference was observed in the L. monocytogenes monoculture biofilms formed with or without the presence of R. insidiosa in a semi-permeable membrane-linked compartment. This observation does not support the notion that R. insidiosa metabolites or secreted signal molecules play significant roles in promoting the incorporation of listeria strains into dual species biofilms. In a reciprocal setting, R. insidiosa biofilm formation was not affected by the presence of L. monocytogenes in a connected semipermeable compartment. We also examined the biofilm formation of L. monocytogenes in the presence or absence of potential R. insidiosa diffusible signal molecules by supernatant replacement. Replacing supernatant from L. monocytogenes overnight culture with that from R. insidiosa growth did not increase the biofilm formation by L. monocytogenes (Data not shown). Taken together, these observations indicated that the synergism of these two species in biofilm formation was dependent on direct cell to cell contacts. However, these observations do not preclude the possibility that diffusible signal molecules are also required for this synergistic interaction.

    Figure 2. Biofilm formation of L. monocytogenes and R. insidiosa in physically separated cultures. Biofilm formation was measured by crystal violet staining of biomass in the base compartment. Letter in the parentheses indicate Listeria or Ralstonia cells in the insert compartment. Neg: Negative control, L: biofilm formation of L. monocytogenes, L(R): effect of R. insidiosa (gown in the insert compartment) on biofilm formation of L. monocytogenes (Grown in the base compartment), R: biofilm formation of R. insidiosa, R(L): effect of L. monocytogenes (Insert compartment) on biofilm formation of R. insidiosa (Base compartment), LR: dual-species biofilm formation in mixed culture in the base compartment.

    3.3. Biofilm formation on stainless steel and SEM observations

    Polystyrene microplates and glasses are the most commonly used substrata for biofilm studies. However, these types of surfaces are scarcely used for food processing. We have previously shown that R. insidiosa promoted the incorporation of E. coli O157:H7 cells into dual species biofilms, using substrata including microplates, glass slides, and glass bottomed petri dishes to facilitate biofilm quantification and microscopic observations [16,20]. In this study, we examined the synergism between L. monocytogenes and R. insidiosa in biofilm formation on stainless steel, which is the surface that is most commonly encountered in food processing environments. When L. monocytogenes and R. insidiosa were gown in 10% TSB and allowed to form biofilms on stainless steel coupons, visual inspection indicated that L. monocytogenes had minimal biofilm formation in monoculture, and that R. insidiosa formed thick biofilms both in monoculture and in the mixed culture with L. monocytogenes. These observations are similar to those made when the mono-and mixed cultures were grown in tissue culture plates, suggesting the difference in these substrata might not be critical for biofilm formation. After three consecutive rinses, cells remaining attached to the SS coupons were recovered and enumerated (Table 2). While comparable populations of R. insidiosa cells (~8 log CFU/coupon) were recovered for both the mono and mixed cultures, L. monocytogenes cells recovered from the coupons in mixed culture (7.3 log CFU/coupon) were nearly 1.9 logs higher than those from the monoculture (5.4 log CFU/coupon). This nearly 100-fold differential in L. monocytogenes from the mono-and dual-species biofilms indicated that R. insidiosa strongly promoted the incorporation of L. monocytogenes cells into the dual species biofilms.

    Table 2. Cell counts of bacteria in mono-culture and dual-species biofilms grown on stainless steel surface (log CFU/coupon).
    Biofilm L. monocytogenes R. insidiosa
    Mono culture 5.42 ± 0.31a 8.12 ± 0.05c
    Mixed culture 7.32 ± 0.11b 7.99 ± 0.15c
    The data represent the means and standard deviations of three independent experiments. Values followed by different letters in superscript are significantly different (P < 0.05).
     | Show Table
    DownLoad: CSV

    The mono-and mixed culture biofilms formed on stainless steel coupons were directly visualized using scanning electron microscopy (SEM) after rinsing and fixation (Figure 3). Consistent with above enumeration, L. monocytogenes cells in pure culture sparsely attached to the stainless steel surface without forming continuous biofilms (Figure 3A). In contrast, the SS coupons from R. insidiosa mono-and the mixed cultures were colonized by bacterial cells at much higher density which exhibited continuous biofilm characteristics (Figure 3B and Figure 3C). We have been unable to distinguish L. monocytogenes cells from those of R. insidiosa in the mixed culture biofilms by cell morphology under SEM, although it seemed the cells of these two species had slight differences in cell size and shapes, and the cells in the mixed culture biofilms seemed more heterologous. In our previous studies involving biofilms formed on glass fiber filter paper using SEM, R. insidiosa exhibited extensive web-like networking of extra-cellular polymeric substances (EPS), to which E. coli O157:H7 cells seemed to interact. In the current study of biofilm formation on stainless steel surface, such extensive EPS networking was not observed. Nevertheless, cells in the mixed culture biofilms were frequently observed with short EPS filaments (arrows in Figure 3C) connected to neighboring cells. Such cell connecting EPS filaments were rarely observed with the cells from the R. insidiosa monoculture biofilms, and their importance in cellular interactions is not clear. Dubey and Ben-Yehuda [21] hypothesized that such filaments play critical roles in intraspecies communications.

    Figure 3. SEM observations of biofilm structure on stainless steel coupons. (A) L. monocytogenes, (B) R. insidiosa, and (C) dual-species. Arrows points to the observed short EPS filaments in the biofilms of the mixed culture.

    4. Conclusions

    Overall, we demonstrated that co-culturing L. monocytogenes and R. insidiosa significantly enhanced biofilm formation. L. monocytogenes cells incorporated into dual-species biofilms formed on stainless steel surface increased by nearly 100-fold compared to monoculture. This synergistic interaction in biofilm formation was dependent upon cell-cell contact. The nature of the interaction has yet to be determined.


    Acknowledgments

    We thank Dr. Ward (ARS National Center for Agricultural Utilization Research, Peoria, Il.) for providing some L. monocytogenes strains used in this study. Y. Xu is a pre-doctoral visiting student to EMFSL USDA ARS supported by China Scholarship Council.


    Conflict of Interest

    All authors declare no conflicts of interest in this paper.




    [1] R. Santis, M. A. Costa, Extended isolation forests for fault detection in small hydroelectric plants, Sustainability, 12 (2020), 6421. https://doi.org/10.3390/su12166421 doi: 10.3390/su12166421
    [2] W. Liu, Y. Zheng, Z. Ma, B. Tian, Q. Chen, An intelligent fault diagnosis scheme for hydropower units based on the pattern recognition of axis orbits, Meas. Sci. Technol., 34 (2022). https://doi.org/10.1088/1361-6501/ac97ff doi: 10.1088/1361-6501/ac97ff
    [3] Y. Lei, B. Yang, X. Jiang, F. Jia, N. Li, A. K. Nandi, Applications of machine learning to machine fault diagnosis:a review and roadmap, Mech. Syst. Signal Process., 138 (2022), 106587. https://doi.org/10.1016/j.ymssp.2019.106587 doi: 10.1016/j.ymssp.2019.106587
    [4] K. Xu, Y. Li, C. Liu, X. Liu, X. Hao, J. Gao, Advanced data collection and analysis in data-driven manufacturing process, Chin. J. Mech. Eng., 33 (2020), 40–60. https://doi.org/10.1186/s10033-020-00459-x doi: 10.1186/s10033-020-00459-x
    [5] A. R. Al-Obaidi, Detection of cavitation phenomenon within a centrifugal Pump based on vibration analysis technique in both time and frequency domains, Exp. Tech., 44 (2020), 329–347. https://doi.org/10.1007/s40799-020-00362-z doi: 10.1007/s40799-020-00362-z
    [6] J. Lin, C. Dou, Q. Wang, Comparisons of MFDFA, EMD and WT by neural network, mahalanobis distance and SVM in fault diagnosis of gearboxes, Sound Vib., 52 (2018), 11–15.
    [7] L. Bai, W. Xi, Early fault diagnosis of rolling bearing based empirical wavelet transform and spectral kurtosis, in 2018 IEEE International Conference on Prognostics and Health Management (ICPHM), (2018), 1–6. https://doi.org/10.1109/ICPHM.2018.8448997
    [8] J. S. Cheng, D. Yu, J. Tang, Y. Yang, Application of SVM and SVD technique based on EMD to the fault diagnosis of the rotating machinery, Shock Vib., 16 (2019), 89–98. https://doi.org/10.3233/SAV-2009-0457 doi: 10.3233/SAV-2009-0457
    [9] Y. Hu, Q. Li, An adjustable envelope based EMD method for rollingbearing fault diagnosis, in IOP Conference Series: Materials Science and Engineering, 1043 (2021), 032017. https://doi.org/10.1088/1757-899X/1043/3/032017
    [10] Y. He, H. Wang, H. Xue, T. Zhang, Research on unknown fault diagnosis of rolling bearings based on parameter-adaptive maximum correlation kurtosis deconvolution, Rev. Sci. Instrum., 92 (2021), 055103. https://doi.org/10.1063/5.0046113 doi: 10.1063/5.0046113
    [11] Z. Li, A. Ming, W. Zhang, T. Liu, F. Chu, Y. Li, Fault feature extraction and enhancement of rolling element bearings based on maximum correlated kurtosis deconvolution and improved empirical wavelet transform, Appl. Sci., 9 (2019), 1876. https://doi.org/10.3390/app9091876 doi: 10.3390/app9091876
    [12] W. Hua, C. Luo, J. Leng, Z. Wang, Mine gearbox fault diagnosis based on multiwavelets and maximum correlated kurtosis deconvolution, J. Vibroeng., 19 (2017), 4185–4197. https://doi.org/10.21595/jve.2017.17497 doi: 10.21595/jve.2017.17497
    [13] F. Wang, C. Liu, W. Su, Z. Xue, Q. Han, H. Li, Combined failure diagnosis of slewing bearings based on MCKD-CEEMD-ApEn, Shock Vib., 2018 (2018), 1070–9622. https://doi.org/10.1155/2018/6321785 doi: 10.1155/2018/6321785
    [14] P. Wang, Y. Miao, Multi classification ERT flow pattern recognition method based on deep learning, J. Phys. Conf. Ser., 2181 (2022). https://doi.org/10.1088/1742-6596/2181/1/012010 doi: 10.1088/1742-6596/2181/1/012010
    [15] D. T. Hoang, H. J. Kang, A survey on deep learning based bearing fault diagnosis, Neurocomputing, 335 (2019), 327–335. https://doi.org/10.1016/j.neucom.2018.06.078 doi: 10.1016/j.neucom.2018.06.078
    [16] D. Yao, H. Liu, J. Yang, J. Zhang, Implementation of a novel algorithm of wheelset and axle box concurrent fault identification based on an efficient neural network with the attention mechanism, J. Intell. Manuf., 32 (2021), 0956–5515. https://doi.org/10.1007/s10845-020-01701-y doi: 10.1007/s10845-020-01701-y
    [17] B. Peng, Y. Bi, B. Xue, M. Zhang, S. Wan, A survey on fault diagnosis of rolling bearings, Algorithms, 15 (2022), 357. https://doi.org/10.3390/a15100347 doi: 10.3390/a15100347
    [18] H. Shao, H. Jiang, X. Li, T. Liang, Rolling bearing fault detection using continuous deep belief network with locally linear embedding, Comput. Ind., 96 (2018), 27–39. https://doi.org/10.1016/j.compind.2018.01.005 doi: 10.1016/j.compind.2018.01.005
    [19] B. Liu, C. Liu, Y. Zhou, D. Wang, Y. Dun, An unsupervised chatter detection method based on AE and merging GMM and K-mean, Mech. Syst. Signal Process., 186 (2023), 109861. https://doi.org/10.1016/j.ymssp.2022.109861 doi: 10.1016/j.ymssp.2022.109861
    [20] B. Ma, W. Cai, Y. Han, G. Yu, A novel probability confidence CNN model and its application in mechanical fault diagnosis, IEEE Trans. Instrum. Meas., 70 (2021), 1–11. https://doi.org/10.1109/TIM.2021.3077965 doi: 10.1109/TIM.2021.3077965
    [21] M. Sun, H. Wang, P. Liu, S. Huang, P. Fan, A sparse stacked denoising autoencoder with optimized transfer learning applied to the fault diagnosis of rolling bearings, Measurement, 146 (2019), 305–314. https://doi.org/10.1016/j.measurement.2019.06.029 doi: 10.1016/j.measurement.2019.06.029
    [22] D. Hoang, H. Kang, Rolling element bearing fault diagnosis using convolutional neural network and vibration image, Cognit. Syst. Res., 53 (2019), 42–50. https://doi.org/10.1016/j.cogsys.2018.03.002 doi: 10.1016/j.cogsys.2018.03.002
    [23] G. Liao, W. Gao, G. Yang, M. Guo, Hydroelectric generating unit fault diagnosis using 1-D convolutional neural network and gated recurrent unit in small hydro, IEEE Sens. J., 19 (2019), 9352–9363. https://doi.org/10.1109/JSEN.2019.2926095 doi: 10.1109/JSEN.2019.2926095
    [24] X. Wang, D. Mao, X. Li, Bearing fault diagnosis based on vibro-acoustic data fusion and 1D-CNN network, Measurement, 173 (2021), 108518. https://doi.org/10.1016/j.measurement.2020.108518 doi: 10.1016/j.measurement.2020.108518
    [25] X. Song, Y. Cong, Y Song, Y. Chen, P. Liang, A bearing fault diagnosis model based on CNN with wide convolution kernels, J. Ambient Intell. Hum. Comput., 13 (2022), 4041–4056. https://doi.org/10.1007/s12652-021-03177-x doi: 10.1007/s12652-021-03177-x
    [26] L. Jia, T. W. S. Chow, Y. Yuan, GTFE-Net: A Gramian time frequency enhancement CNN for bearing fault diagnosis, Eng. Appl. Artif. Intell., 119 (2023), 105794. https://doi.org/10.1016/j.engappai.2022.105794 doi: 10.1016/j.engappai.2022.105794
    [27] N. Sakli, H. Ghabri, B. O. Soufiene, F. Almalki, H. Sakli, O. Ali, et al., Resnet-50 for 12-Lead electrocardiogram automated diagnosis, Comput. Intell. Neurosci., 2022 (2022), 1–16. https://doi.org/10.1155/2022/7617551 doi: 10.1155/2022/7617551
    [28] W. Zhang, X. Li, Q. Ding, Deep residual learning-based fault diagnosis method for rotating machinery, ISA Trans., 95 (2019), 295–305. https://doi.org/10.1016/j.isatra.2018.12.025 doi: 10.1016/j.isatra.2018.12.025
    [29] Y. Wang, J. Liang, X. Gu, D. Ling, H. Yu, Multi-scale attention mechanism residual neural network for fault diagnosis of rolling bearings, Proc. Inst. Mech. Eng., Part C: J. Mech. Eng. Sci., 236 (2022), 10615–10629. https://doi.org/10.1177/09544062221104598 doi: 10.1177/09544062221104598
    [30] Y. Jin, C. Qin, Y. Huang, C. Liu, Actual bearing compound fault diagnosis based on active learning and decoupling attentional residual network, Measurement, 173 (2021), 108500. https://doi.org/10.1016/j.measurement.2020.108500 doi: 10.1016/j.measurement.2020.108500
    [31] Y. Chen, D. Zhang, H. Ni, J. Cheng, H. R. Karimi, Multi-scale split dual calibration network with periodic information for interpretable fault diagnosis of rotating machinery, Eng. Appl. Artif. Intell., 123 (2023), 106181. https://doi.org/10.1016/j.engappai.2023.106181 doi: 10.1016/j.engappai.2023.106181
    [32] Y. Chen, D. Zhang, H. Zhang, Q. Wang, Dual-path mixed-domain residual threshold networks for bearing fault diagnosis, IEEE Trans. Ind. Electron., 69 (2022), 13462–13472. https://doi.org/10.1109/TIE.2022.3144572 doi: 10.1109/TIE.2022.3144572
    [33] Y. Chen, D. Zhang, K. Zhu, R. Yan, An adaptive activation transfer learning approach for fault diagnosis, IEEE/ASME Trans. Mechatron., 2023 (2023), 1–12. https://doi.org/10.1109/TMECH.2023.3243533 doi: 10.1109/TMECH.2023.3243533
    [34] N. H. Phong, A. Santos, B. Ribeiro, PSO-convolutional neural networks with heterogeneous learning rate, IEEE Access, 10 (2022), 89970–89988. https://doi.org/10.1016/10.1109/ACCESS.2022.3201142 doi: 10.1016/10.1109/ACCESS.2022.3201142
    [35] Z. Jiang, J. Zheng, H. Pan, Sigmoid-based refined composite multiscale fuzzy entropy and t-SNE based fault diagnosis approach for rolling bearing, Measurement, 129 (2018), 332–342. https://doi.org/10.1016/j.measurement.2018.07.045 doi: 10.1016/j.measurement.2018.07.045
    [36] J. Yu, C. Xiao, T. Hu, Y. Gao, Selective weighted multi-scale morphological filter for fault feature extraction of rolling bearings, ISA Trans., 132 (2022), 544–556. https://doi.org/10.1016/j.isatra.2022.06.003 doi: 10.1016/j.isatra.2022.06.003
    [37] X. Dong, G. Li, Y. Jia, K. Xu, Multiscale feature extraction from the perspective of graph for hob fault diagnosis using spectral graph wavelet transform combined with improved random forest, Measurement, 176 (2021), 109178. https://doi.org/10.1016/j.measurement.2021.109178 doi: 10.1016/j.measurement.2021.109178
    [38] W. Huang, J. Chen, Y. Yang, G. Guo, An improved deep convolutional neural network with multi-scale information for bearing fault diagnosis, Neurocomputing, 395 (2019), 77–92. https://doi.org/10.1016/j.neucom.2019.05.052 doi: 10.1016/j.neucom.2019.05.052
    [39] B. Zhao, X. Zhang, Z. Zhan, Q. Wu, Deep multi-scale adversarial network with attention:a novel domain adaptation method for intelligent fault diagnosis, J. Manuf. Syst., 59 (2021), 565–576. https://doi.org/10.1016/j.jmsy.2021.03.024 doi: 10.1016/j.jmsy.2021.03.024
    [40] R. Fezai, K. Dhibi, M. Mansouri, M. Trabelsi, M. Hajji, K. Bouzrara, et al., Effective random forest-based fault detection and diagnosis for wind energy conversion systems, IEEE Sens. J., 21 (2021), 6914–6921. https://doi.org/10.1109/JSEN.2020.3037237 doi: 10.1109/JSEN.2020.3037237
    [41] J. Xu, S. Liang, X. Ding, R. Yan, A zero-shot fault semantics learning model for compound fault diagnosis, Expert Syst. Appl., 221 (2023), 119642. https://doi.org/10.1016/j.eswa.2023.119642 doi: 10.1016/j.eswa.2023.119642
  • This article has been cited by:

    1. Manyun Yang, Xiaobo Liu, Yaguang Luo, Arne J. Pearlstein, Shilong Wang, Hayden Dillow, Kevin Reed, Zhen Jia, Arnav Sharma, Bin Zhou, Dan Pearlstein, Hengyong Yu, Boce Zhang, Machine learning-enabled non-destructive paper chromogenic array detection of multiplexed viable pathogens on food, 2021, 2, 2662-1355, 110, 10.1038/s43016-021-00229-5
    2. Dennis Nurjadi, Sébastien Boutin, Katja Schmidt, Melinda Ahmels, Daniel Hasche, Identification and Elimination of the Clinically Relevant Multi-Resistant Environmental Bacteria Ralstonia insidiosa in Primary Cell Culture, 2020, 8, 2076-2607, 1599, 10.3390/microorganisms8101599
    3. Andrea Foote, Kristin Schutz, Zirui Zhao, Pauline DiGianivittorio, Bethany R. Korwin-Mihavics, John J. LiPuma, Matthew J. Wargo, Olaya Rendueles, Characterizing Biofilm Interactions between Ralstonia insidiosa and Chryseobacterium gleum , 2023, 2165-0497, 10.1128/spectrum.04105-22
    4. Faizan Ahmed Sadiq, Koen De Reu, Mette Burmølle, Sharon Maes, Marc Heyndrickx, Synergistic interactions in multispecies biofilm combinations of bacterial isolates recovered from diverse food processing industries, 2023, 14, 1664-302X, 10.3389/fmicb.2023.1159434
    5. Qingying Chen, Xingguo Zhang, Qingqing Wang, Jingxian Yang, Qingping Zhong, The mixed biofilm formed by Listeria monocytogenes and other bacteria: Formation, interaction and control strategies, 2023, 1040-8398, 1, 10.1080/10408398.2023.2200861
    6. Tessa Tuytschaever, Katleen Raes, Imca Sampers, Listeria monocytogenes in food businesses: From persistence strategies to intervention/prevention strategies—A review, 2023, 1541-4337, 10.1111/1541-4337.13219
    7. Tingting Gu, Yaguang Luo, Zhen Jia, Apisak Meesrison, Sophia Lin, Isabella J. Ventresca, Sarah J. Brooks, Arnav Sharma, Sitara Sriram, Manyun Yang, Arne J. Pearlstein, Patricia D. Millner, Keith R. Schneider, Boce Zhang, Surface topography and chemistry of food contact substances, and microbial nutrition affect pathogen persistence and symbiosis in cocktail Listeria monocytogenes biofilms, 2024, 09567135, 110391, 10.1016/j.foodcont.2024.110391
    8. Grishma S. Prabhukhot, Charles D. Eggleton, Moon Kim, Jitendra Patel, Impact of surface topography and hydrodynamic flow conditions on single and multispecies biofilm formation by Escherichia coli O157:H7 and Listeria monocytogenes in presence of promotor bacteria, 2024, 201, 00236438, 116240, 10.1016/j.lwt.2024.116240
    9. Tessa Tuytschaever, Christine Faille, Katleen Raes, Imca Sampers, Influence of slope, material, and temperature on Listeria monocytogenes and Pseudomonas aeruginosa mono- and dual-species biofilms , 2024, 0892-7014, 1, 10.1080/08927014.2024.2380410
    10. Yi Wang, Yihang Feng, Xinhao Wang, Chenyang Ji, Abhinav Upadhyay, Zhenlei Xiao, Yangchao Luo, A short review on Salmonella spp. involved mixed-species biofilm on food processing surface: Interactions, disinfectant resistance and its biocontrol, 2025, 19, 26661543, 101660, 10.1016/j.jafr.2025.101660
    11. Julia Burzyńska, Aleksandra Tukendorf, Marta Fangrat, Katarzyna Dzierżanowska-Fangrat, Unveiling Ralstonia spp. in the Neonatal Intensive Care Unit: Clinical Impacts and Antibiotic Resistance, 2025, 14, 2079-6382, 259, 10.3390/antibiotics14030259
  • Reader Comments
  • © 2023 the Author(s), licensee AIMS Press. This is an open access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0)
通讯作者: 陈斌, bchen63@163.com
  • 1. 

    沈阳化工大学材料科学与工程学院 沈阳 110142

  1. 本站搜索
  2. 百度学术搜索
  3. 万方数据库搜索
  4. CNKI搜索

Metrics

Article views(12061) PDF downloads(85) Cited by(8)

Article outline

Figures and Tables

Figures(8)  /  Tables(7)

Other Articles By Authors

/

DownLoad:  Full-Size Img  PowerPoint
Return
Return

Catalog