Search

Machine learning has shown advantages in assessing wheel degradation in metro vehicles for fault prognostic and health management (PHM). However, practical implementation faces challenges due to disturbances in wheel vibration signals caused by factors like load, road conditions, and temperature, introducing significant noise that masks evaluation tendencies. This article develops a novel unsupervised approach to evaluate wheel degradation under strong noise disturbance. It utilizes a second-generation wavelet transform for time–frequency analysis of noisy signals and introduces a second-generation wavelet deep autoencoder (SWDAE) network to extract adaptive feature representations in different frequency bands. The training algorithm alternately optimizes the wavelet transform and DAE. With frequency-saliency interpretability, health indicators (HIs) for the degradation process are constructed using principal component analysis (PCA) on the obtained features in each frequency band, selecting the most representative frequency component based on the monotonicity of the HIs. State changes are automatically determined using a second-order derivative-based assessment method aligned with the first/second warning strategy. Comparative experiments using Beijing Subway wheel data demonstrate the monotonicity and physical significance of the constructed HIs, with warning locations accurately matching changes in wheel diameter recorded during repairs.

This article tries to solve the problem of online early fault detection (EFD) with unlabeled streaming data by addressing the following challenges: 1) no state information is available for online data; 2) online working condition is not determined in advance; and 3) false alarm should be avoided. This article first proposes a deep tensor multitask anomaly detection (Tensor-MAD) model with rule adaptation for the online EFD. Running on multitask learning architecture, Tensor-MAD builds a new tensorized pooling filter to keep the essential information of each task’s operating status from noisy data. With hypersphere-based one-class detection rule representation, Tensor-MAD further constructs a new rule adaptation mechanism to transfer the detection rule from offline labeled data to the online unlabeled data. A training algorithm with an alternating minimization scheme is also provided to update tensor decomposition and rule adaptation. Then, the optimal information filter level and the rule adaptation degree can be determined. Based on the obtained anomalies, this article proposes a nonparametric alarm threshold setting method based on the sequential accumulation of anomaly probability. This threshold can be adaptively chosen once an expected false alarm rate (FAR) is given. A rationality proof is also provided. Experimental results on the IEEE PHM Challenge 2012 bearing dataset demonstrate that the proposed approach can adaptively and accurately evaluate the early fault occurrence from unlabeled streaming data. More importantly, the proposed approach has a much lower FAR and a faster convergence speed, providing an easy-to-deploy and reliable solution for online EFD.

With deep transfer learning techniques, this article focuses on the online remaining useful life (RUL) prediction problem across different machines and tries to address the following concerns: 1) the effect of transfer learning decreases significantly due to considerable divergence of degradation characteristic; 2) a high computational cost is raised by retraining whole model with online data; and 3) error accumulation occurs because of lacking label information of online data. In this article, a self-supervised deep tensor domain-adversarial regression adaptation approach is proposed. In the pretraining stage, a novel tensor domain-adversarial network, with a tensorized domain discriminator, is constructed using the offline whole-life degradation data and early fault data of the target machine. A new training algorithm with an alternating minimization scheme is then built to seek the optimal core tensor and domain-invariant feature representation. In the online stage, a new self-supervised fine-tuning strategy is designed for the target network initialized from the pretrained network. The core tensor-formed self-supervised information, extracted from the monotonicity of online degradation process, and the pseudo-supervised information from the pretrained network are integrated to realize fast and adaptive RUL prediction. This article takes rolling bearing as an example and runs both cross-condition and cross-machine experiments on three rolling bearings datasets, i.e., IEEE PHM Challenge 2012, XJTU-SY, and our test rig. The results verify that the use of tensor representation can facilitate regression adversarial training and demonstrate that the proposed approach can effectively improve predictive accuracy and stability under unknown online conditions.

Recent research has shown that creative self‐concept might be not only a path to people's creativity but also a path to their positive affect. This study proposed and examined the underlying mechanisms of how creative self‐concept is associated with positive affect through the mediating role of people's daily creative behavior and creative self‐assessment. We recruited 101 Chinese undergraduate students to report their creative behavior, creative self‐assessment, and experiences of positive emotion states each day for 14 days. A multilevel structural equation model was conducted. Results confirmed the mediating role of creative self‐assessment and the chain‐mediating role of creative behavior and creative self‐assessment between creative self‐concept and positive affect. The direct effect of creative behavior on positive affect was not verified. These findings highlight the role that creative self‐concept plays in everyday activities and everyday well‐being and show the importance of creative self‐assessment in creativity education and mental health care.

To tolerate high processing temperature during the fabrication of low‐temperature polycrystalline silicon thin‐film transistors (LTPS–TFT) in flexible OLED devices, the polyimide (PI) films, which are used as substrate, should have ultra‐high glass transition temperature (Tg> 450°C) and ultra‐low coefficient of thermal expansion (CTE at 0–5 ppm K−1). In this paper, two novel heterocyclic monomers, namely, N,N'‐(xanthone‐2,7‐diyl)bis(4‐aminobenzamide) (p‐DAXBA) and N,N'‐(xanthone‐2,7‐diyl)bis(3‐aminobenzamide) (m‐DAXBA), which contain a xanthone moiety, are prepared and polycondensed with pyromellitic dianhydride (PMDA), respectively. PI films (PIa and PIb) with intrinsic high Tgand low CTE are designed from the perspective of rigid conjugate xanthone structure and hydrogen bonding interaction. It is found that the PIa films prepared by p‐DAXBA have better linear structure of molecular chains and show relatively higher Tgand lower CTE. The Tgof PIa‐40 is greater than 450°C, and CTE can reach as low as 2.7 ppm K−1, tensile strength of 179 MPa, modulus of 5.67 GPa, indicating potential application prospect as a flexible OLED substrate.

Research has demonstrated that implicit theories of creativity are crucial in shaping an individual’s behavior and real‐life decisions toward being creative. The present study proposed and examined the underlying mechanisms of how two kinds of implicit theories—the growth mindset of the creative self and the stereotype of creative others—are associated with creative achievements through the mediating role of creativity motivation. Participants were 606 undergraduate students who were enrolled in an education major in two universities in China. Overall, the study found that Chinese students held a positive image toward a creative student, regarding him or her as highly competent, warm, and popular. Student perceptions of a creative other were positively related to their growth mindset of creativity. Moreover, results verified both the mediating role of creativity motivation on growth mindset, as well as the effect of positive stereotyping of the creative other on students’ creative achievement. These findings point to promising creativity motivation strategies including the cultivation of a malleable view of creativity and of creative role models, that may, in turn, promote creative achievement by encouraging students to do, learn, and accomplish new things.

Semiconductor photocatalysis has two important applications: environmental remediation and H2production. Bismuth vanadate has attracted great interests as an ideal candidate for use as high-performance photocatalysts for visible light-driven water splitting and photoanode in photoelectrochemical cells. Single crystal film is invaluable to deepen the fundamental understanding of materials. The object of this article was to provide a brief review on the advances in the preparation of epitaxial BiVO4thin film, the physical and photoelectrochemical properties. The epitaxial BiVO4thin film could be fabricated on (001) yttria-stabilized cubic zirconia or SrTiO3(001) by molecular beam epitaxy, chemical vapor deposition, and, pulsed laser deposition. Three crystal structures, namely, monoclinic scheelite, monoclinic clinobisvanite, and orthorhombic, were mentioned in the literatures. The optical band gap was reported to be 2.5~2.7 eV for a direct transition. The highest photocurrent of 4.83 mA cm−2at 1.23 V vs. the RHE was obtained at the sample with Lu2O3as a hole blocking layer.

Plants are exposed to different abiotic stresses that affect growth, development, and productivity. Previous studies have shown that late embryogenesis abundant (LEA) proteins play important roles in plant adaptation to abiotic stresses. However, reports that demonstrate their biological functions, especially in Caragana korshinskiiKom., are still very limited. In this study, 26 LEAgenes were identified from dehydration-treated suppressive subtractive hybridizations cDNA library and transcriptome sequencing data of C. korshinskii, and were classified into seven groups according to their structural features. Quantitative real-time PCR analysis revealed that the CkLEAswere induced by abscisic acid (ABA) and diverse abiotic stresses and widely expressed in various tissues. Moreover, overexpression of CkLEA2-3in Arabidopsis thalianaL. Heynh resulted in enhanced tolerance to ABA treatment, and osmotic and salt stresses during seed germination. CkLEA2-3overexpression lines also exhibited resistance to drought stress during seedling development. Taken together, our results indicate that CkLEA2-3 plays positive roles in conferring abiotic stress tolerance in Arabidopsis.

Objectives:We aimed to determine the efficacy and toxicity of treating locally advanced cervical cancer (LACC) with a neoadjuvant chemotherapy (NAC) regimen combining docetaxel and cisplatin followed by radical surgery.Methods:We retrospectively reviewed the clinical records of patients with stage IB2 – IIB (tumor diameter ≥ 4 cm) disease admitted between January 2007 and July 2009 who, before radical hysterectomy and pelvic lymph node dissection, received two to three courses of an NAC regimen comprising docetaxel (75 mg/m2) and cisplatin (70 – 75 mg/m2).Results:Fifty-two patients with LACC received 109 cycles of NAC. The objective response rate was 86.5% (26.9% CR and 17.3% pathological CR). Stage IB2 disease had a more favorable response to NAC (95.7%, p 0.019). Deep stromal invasion and lymph-vascular space metastasis rates were significantly lower in NAC responders (p 0.033) than in nonresponders (p 0.012). Most side effects of NAC were mild or moderate. Log-rank test showed the 2-year overall survival and progression-free survival rates were 100 and 90.3% for NAC responders, compared with only 57.1% (p 0.000) and 68.6% for nonresponders (p 0.012), respectively.Conclusion:Neoadjuvant docetaxel combined with cisplatin yielded a high response rate with well tolerable toxicity for LACC and could decrease pathological risk factors in NAC responders.

Supervised word segmentation heavily relies on large-scale and high quality labeled data. However, building such a corpus is difficult, especially with respect to domain specific data. In this paper, we propose a novel semi-supervised Chinese word segmentation (CWS) method. Specifically, we seek to select more useful sample sentences from the large-scale unlabeled sentences to extend the training data, by means of a sampling strategy that uses character-based semantic similarity. The presented similarity algorithm is used to calculate the similarity between unlabeled sentences and the training data, which can help select helpful sample sentences from unlabeled data. In addition, we integrate an attention mechanism into our word segmentation model to focus on available contextual information. Experiments on PKU, MSR and Weibo benchmark data sets show that our method outperforms the previous neural network models and state-of-the-art methods.