Browsing by Author "Wang, Tao"

Now showing 1 - 10 of 10
  • Thumbnail Image
    ItemOpen Access
    A chemical sensor based on a photonic-crystal L3 nanocavity defined in a silicon-nitride membrane
    (Royal Society of Chemistry, 2014-08-27) Deasy, Kieran; Sediq, Khalid N.; Brittle, Stuart; Wang, Tao; Davis, Frank; Richardson, Tim H.; Lidzey, David G.
    The application of a silicon-nitride based L3 optical nanocavity as a chemical sensor is explored. It is shown that by adjusting the thickness of an ultra-thin Lumogen Red film deposited onto the nanocavity surface, the fundamental optical mode undergoes a progressive red-shift as the layer-thickness increases, with the cavity being able to detect the presence of a single molecular monolayer. The optical properties of a nanocavity whose surface is coated with a thin layer of a porphyrin-based polymer are also explored. On exposure of the cavity to an acidic-vapour, it is shown that changes in the optical properties of the porphyrin-film (thickness and refractive index) can be detected through a reversible shift in the cavity mode wavelength. Such effects are described using a finite difference time-domain model.
  • Thumbnail Image
    ItemOpen Access
    Fault diagnosis of industrial robot based on dual-module attention convolutional neural network
    (Springer, 2022-06-01) Lu, Kaijie; Chen, Chong; Wang, Tao; Cheng, Lianglun; Qin, Jian
    Fault diagnosis plays a vital role in assessing the health management of industrial robots and improving maintenance schedules. In recent decades, artificial intelligence-based data-driven approaches have made significant progress in machine fault diagnosis using monitoring data. However, current methods pay less attention to correlations and internal differences in monitoring data, resulting in limited diagnostic performance. In this paper, a data-driven method is proposed for the fault diagnosis of industrial robot reducers, that is, a dual-module attention convolutional neural network (DMA-CNN). This method aims to diagnose the fault state of industrial robot reducer. It establishes two parallel convolutional neural networks with two different attentions to capture the different features related to the fault. Finally, the features are fused to obtain the fault diagnosis results (normal or abnormal). The fault diagnosis effect of the DMA-CNN method and other attention models are compared and analyzed. The effectiveness of the method is verified on a dataset of real industrial robots.
  • Thumbnail Image
    ItemOpen Access
    Label-free creatinine optical sensing using molecularly imprinted titanium dioxide-polycarboxylic acid hybrid thin films: a preliminary study for urine sample analysis
    (MDPI, 2021-07-17) Lee, Seung-Woo; Ahmed, Soad; Wang, Tao; Park, Yeawon; Matsuzaki, Sota; Tatsumi, Shinichi; Matsumoto, Shigekiyo; Korposh, Sergiy; James, Steve
    Creatinine (CR) is a representative metabolic byproduct of muscles, and its sensitive and selective detection has become critical in the diagnosis of kidney diseases. In this study, poly(acrylic acid) (PAA)-assisted molecularly imprinted (MI) TiO2 nanothin films fabricated via liquid phase deposition (LPD) were employed for CR detection. The molecular recognition properties of the fabricated films were evaluated using fiber optic long period grating (LPG) and quartz crystal microbalance sensors. Imprinting effects were examined compared with nonimprinted (NI) pure TiO2 and PAA-assisted TiO2 films fabricated similarly without a template. In addition, the surface modification of the optical fiber section containing the LPG with a mesoporous base coating of silica nanoparticles, which was conducted before LPD-based TiO2 film deposition, contributed to the improvement of the sensitivity of the MI LPG sensor. The sensitivity and selectivity of LPGs coated with MI films were tested using CR solutions dissolved in different pH waters and artificial urine (near pH 7). The CR binding constants of the MI and NI films, which were calculated from the Benesi–Hildebrand plots of the wavelength shifts of the second LPG band recorded in water at pH 4.6, were estimated to be 67 and 7.8 M−1, respectively, showing an almost ninefold higher sensitivity in the MI film. The mechanism of the interaction between the template and the TiO2 matrix and the film composition was investigated via ultraviolet–visible and attenuated total reflectance Fourier-transform infrared spectroscopy along with X-ray photoelectron spectroscopy analysis. In addition, morphological studies using a scanning electron microscope and atomic force microscope were conducted. The proposed system has the potential for practical use to determine CR levels in urine samples. This LPG-based label-free CR biosensor is innovative and expected to be a new tool to identify complex biomolecules in terms of its easy fabrication and simplicity in methodology.
  • Thumbnail Image
    ItemOpen Access
    A long period grating optical fiber sensor with nano-assembled porphyrin layers for detecting ammonia gas
    (Elsevier, 2016-01-19) Wang, Tao; Yasukochi, Wataru; Korposh, Sergiy; James, Stephen W.; Tatam, Ralph P.; Lee, Seung-Woo
    In this study, the ability of a long period grating (LPG) optical fiber sensor coated with a multilayer film of poly(diallyldimethyammonium chloride) (PDDA) and tetrakis(4-sulfophenyl)porphine (TSPP) to detect ammonia gas was demonstrated. The nano-assembled thin film was prepared using a layer-by-layer deposition technique. This combination of an LPG and TSPP could allow highly sensitive optical sensors that specifically bind ammonia to be fabricated. An LPG with a period of 100 μm and a PDDA/TSPP coating produced by depositing fifteen layers afforded a limit of detection of 0.67 ppm for ammonia gas. TSPP molecules in the film acted as ammonia receptors because the TSPP morphologically changed from J- or H-aggregates to free base monomers when it was deprotonated by being exposed to ammonia. Interestingly, HCl vapor could be used to increase the relative amount of J-aggregation in the TSPP and to restore the sensor response. The reversibility of the morphological change in the TSPP allowed reversible changes to occur in the properties of the coating, including the refractive index, film thickness and density, and electrostatic interactions; these influenced the transmission spectrum of the LPG.
  • Thumbnail Image
    ItemOpen Access
    Long-period grating fiber-optic sensors exploiting molecularly imprinted TiO2 nanothin films with photocatalytic self-cleaning ability
    (Springer, 2020-11-17) Wang, Tao; Korposh, Sergiy; James, Stephen; Lee, Seung-Woo
    Highly sensitive and selective long-period grating (LPG) fiber-optic sensors modified with molecularly imprinted TiO2 nanothin films were fabricated. The films were deposited onto the surface of the optical fiber via liquid-phase deposition (LPD), using tetrakis(N-methylpyridinium-4-yl)porphyrin (TMPyP) as a template. Three LPG resonance bands were monitored during film deposition, which was of duration 4.5 h. Prior to template removal, heat treatment at 60 °C under high-humidity conditions led to an increase in refractive index of the TiO2 film, evidenced by changes in the central wavelengths of the attenuation bands. After template removal using HCl solution (0.01 M), the TMPyP-imprinted film-modified LPG sensor showed higher sensitivity to the template molecule than to structurally related guest molecules. This was measured at the 1st and 2nd resonance bands, with wavelengths ranging from 690 to 738 nm and 815 to 905 nm, respectively. No selective binding of the template was observed with a non-imprinted TiO2 film prepared in the same manner. Furthermore, the heat-treated imprinted films exhibited a substantial enhancement of photocatalytic activity for template irradiation. In particular, the self-cleaning property of the imprinted film-modified LPG sensor under ultraviolet irradiation led to highly efficient and selective binding to the template. The mechanism of the interaction between the template and the TiO2 matrix was investigated by UV–vis and Fourier-transform infrared (FTIR) spectroscopies. Additionally, morphological studies using scanning electron microscopy (SEM) were conducted.
  • Thumbnail Image
    ItemOpen Access
    A machine learning-based approach for elevator door system fault diagnosis
    (IEEE, 2022-10-28) Liang, Taiwang; Chen, Chong; Wang, Tao; Zhang, Ao; Qin, Jian
    The door system is the core part of the elevator. An accurate diagnosis of the door system can aid engineers in troubleshooting and reduce maintenance costs. However, the research of fault diagnosis based on elevator operation and maintenance data is still in its infancy. With the development of the industrial Internet-of-things, real-time monitoring data of elevator can be collected and used for fault diagnosis modeling. This paper investigates a machine learning-based approach to achieve accurate elevator door fault diagnosis. An experimental study was conducted based on the monitoring data collected from the real-world elevator door system. The experimental results revealed that XGBoost algorithm can accurately identify the fault type of the elevator door.
  • Thumbnail Image
    ItemOpen Access
    Model-agnostic meta-learning for fault diagnosis of industrial robots
    (IEEE, 2023-10-16) Liu, Yuxin; Chen, Chong; Wang, Tao; Cheng, Lianglun; Qin, Jian
    The success of deep learning in the field of fault diagnosis depends on a large number of training data, but it is a challenge to achieve fault diagnosis of multi-axis industrial robots in the case of few-shot. To address this issue, this paper proposes a method called Model-Agnostic Meta-Learning (MAML) for fault diagnosis of industrial robots. Its goal is to train an effective industrial robot fault classifier using minimal training data. Additionally, it can learn to recognize faults in new scenarios with high accuracy based on the training data. Experimental results based on a six-axis industrial robot dataset show that the proposed method is superior to traditional convolutional neural network (CNN) and transfer learning, and that the diagnostic results with the same amount of data in few-shot cases are better than existing intelligent fault diagnosis methods.
  • Thumbnail Image
    ItemOpen Access
    Optical gas sensing using gold nanoparticles/polyelectrolyte layer-by-layer films: a case study with poly(acrylic acid) for ammonia detection
    (Scientific Publishing Division Myu, 2016-03-30) Wang, Tao; Okada, Takuya; Hayashi, Kenshi; James, Stephen W.; Lee, Seung-Woo
    this study, we demonstrated a new approach for gas sensing using multilayer films selfassembled with cationic gold nanoparticles (AuNPs+) and anionic polyelectrolytes via layerby- layer (LbL) deposition. The surface of AuNPs+ was modified with a cationic disulfide, [–S(CH2)2NHCO(CH2)10(CH3)3N+Br−]2. Two types of LbL film using AuNPs+ were selfassembled with anionic polyelectrolytes, poly(acrylic acid) (PAA), and poly(sodium 4-styrene sulfonate) (PSS) on quartz substrates, and their film growth was confirmed by UV–vis measurements. The 10-cycle AuNPs+/PAA film showed reproducible responses after exposure to ammonia gas (3000 ppm) and HCl treatment (0.1 M), showing absorbance changes of 0.0428 ± 0.0033 at 299 nm and 0.0053 ± 0.0013 at 544 nm for five repeated tests. On the other hand, no significant changes were observed for the 10-cycle AuNPs+/PSS film. Ammonia binding was based on the acid–base interaction between ammonia and the free carboxylic acid groups of PAA. The film's composition and morphology were also investigated by Fourier transform infrared spectroscopy (FTIR) and atomic force microscopy (AFM) measurements to clarify the sensing mechanism.
  • Thumbnail Image
    ItemOpen Access
    Recognition of speed signs in uncertain and dynamic environments
    (IOP Publishing: Conference Series, 2019-05-08) Zhu, Zhilong; Xu, Gang; He, Hongmei; Jiang, Juanjuan; Wang, Tao
    The speed limit signs recognition directly affects the safety of autonomous vehicles. Vehicles are usually running in an uncertain and dynamic environment. The performance of the recognition system is affected by various factors such as the different sizes of pictures, illumination condition and position circumstances, which can lead to misclassification. This makes the speed sign recognition challengeable. To improve the recognition rate of the speed signs in such environments, this work firstly applies the method of the saliency target detection based on the background-absorbing Markov chain, to extract the node in an image, then uses SPP-CNN to classify the extracted nodes with ten-folder validation. The recognition rate is up to 9.32%, higher than that obtained directly by SPP-CNN working on raw dataset.
  • Thumbnail Image
    ItemOpen Access
    Spatial attention-based convolutional transformer for bearing remaining useful life prediction
    (IOP Publishing, 2022-08-02) Chen, Chong; Wang, Tao; Liu, Ying; Cheng, Lianglun; Qin, Jian
    The remaining useful life (RUL) prediction is of significance to the health management of bearings. Recently, deep learning has been widely investigated for bearing RUL prediction due to its great success in sequence learning. However, the improvement of the prediction accuracy of existing deep learning algorithms heavily relies on feature engineering such as handcrafted feature generation and time–frequency transformation, which increase the complexity and difficulty of the actual deployment. In this paper, a novel spatial attention-based convolutional transformer (SAConvFormer) is proposed to establish an accurate bearing RUL prediction model based on raw vibration data without prior knowledge or feature engineering. In this algorithm, firstly, a convolutional neural network enhanced by a spatial attention mechanism is proposed to squeeze the feature maps and extract the local and global features from raw bearing vibration data effectively. Then, the extracted senior features are fed into a transformer network to further explore the sequential patterns relevant to the bearing RUL. An experimental study using the XJTU-SY rolling bearings dataset revealed the merits of the proposed deep learning algorithm in terms of root-mean-square-error (RMSE) and mean-absolute-error (MAE) in comparison with other state-of-the-art algorithms.