AST3DRNet: Attention-Based Spatio-Temporal 3D Residual Neural Networks for Traffic Congestion Prediction.

Traffic congestion prediction has become an indispensable component of an intelligent transport system. However, one limitation of the existing methods is that they treat the effects of spatio-temporal correlations on traffic prediction as invariable during modeling spatio-temporal features, which results in inadequate modeling. In this paper, we propose an attention-based spatio-temporal 3D residual neural network, named AST3DRNet, to directly forecast the congestion levels of road networks in a city. AST3DRNet combines a 3D residual network and a self-attention mechanism together to efficiently model the spatial and temporal information of traffic congestion data. Specifically, by stacking 3D residual units and 3D convolution, we proposed a 3D convolution module that can simultaneously capture various spatio-temporal correlations. Furthermore, a novel spatio-temporal attention module is proposed to explicitly model the different contributions of spatio-temporal correlations in both spatial and temporal dimensions through the self-attention mechanism. Extensive experiments are conducted on a real-world traffic congestion dataset in Kunming, and the results demonstrate that AST3DRNet outperforms the baselines in short-term (5/10/15 min) traffic congestion predictions with an average accuracy improvement of 59.05%, 64.69%, and 48.22%, respectively.

A Complete Method for Evaluating the Performance of Photocatalysts for the Degradation of Antibiotics in Environmental Remediation.

Various antibiotics such as tetracycline, aureomycin, amoxicillin, and levofloxacin are found in large quantities in groundwater and soil systems, potentially leading to the development of resistant and multi-drug resistant bacteria, posing a threat to humans, animals, and environmental systems. Photocatalytic technology has attracted keen interest due to its rapid and stable treatment and direct use of solar energy. However, most studies evaluating the performance of semiconductor catalysts for the photocatalytic degradation of organic pollutants in water are currently incomplete. In this paper, a complete experimental protocol is designed to comprehensively evaluate the photocatalytic performance of semiconductor catalysts. Herein, rhombic dodecahedral silver phosphate was prepared by a simple solvent phase synthesis method at room temperature and atmospheric pressure. BrSubphthalocyanine/Ag3PO4 heterojunction materials were prepared by the solvothermal method. The catalytic performance of as-prepared materials for the degradation of tetracycline was evaluated by studying different influencing factors such as catalyst dosage, temperature, pH, and anions at atmospheric pressure using a 300 W xenon lamp as a simulated solar light source and a light intensity of 350 mW/cm2. Compared with the first cycle, the constructed BrSubphthalocyanine/Ag3PO4 maintained 82.0% of the original photocatalytic activity after five photocatalytic cycles, while the pristine Ag3PO4 maintained only 28.6%. The stability of silver phosphate samples was further tested by a five-cycle experiment. This paper provides a complete process for evaluating the catalytic performance of semiconductor catalysts in the laboratory for the development of semiconductor catalysts with potential for practical applications.

Multiwavelength generation based on a mode-locked fiber laser using carbon nanotube and fiber Fabry-Perot filter.

In this paper, a novel and compact configuration of stable multiwavelength generation with a uniform wavelength interval is proposed for the first time to our knowledge. It employs a mode-locked fiber laser using a carbon nanotube and spectrum-slicing technique. A flat rectangular optical output spectrum is demonstrated by adjusting the dispersion value of the fiber-loop cavity and the pump power. With a fiber Fabry-Perot filter, 33 wavelengths with 0.2 nm spacing are obtained among the power uniformity of 2.3 dB. Moreover, the variations of output power at each wavelength are all less than 0.1 dB, which implies excellent stability of the whole structure.

All-fiber Mach-Zehnder interferometers for sensing applications.

We propose and demonstrate a thinned fiber based Mach-Zehnder interferometer for multi-purpose sensing applications. The sensor head is formed by all-fiber in-line singlemode-multimode-thinned-singlemode (SMTS) fiber structure, only using the splicing method. The principle of operation relies on the effect that the thinned fiber cladding modes interference with the core mode by employing a multimode fiber as a mode coupler. Experimental results showed that the liquid refractive index information can be simultaneously provided from measuring the sensitivity of the liquid level. A 9.00 mm long thinned fiber sensor at a wavelength of 1538.7228 nm exhibits a water level sensitivity of -175.8 pm/mm, and refractive index sensitivity as high as -1868.42 (pm/mm)/RIU, respectively. The measuring method is novel, for the first time to our knowledge. In addition, it also demonstrates that by monitoring the wavelength shift, the sensor at a wavelength of 1566.4785 nm exhibits a refractive index sensitivity of -25.2935 nm/RIU, temperature sensitivity of 0.0615 nm/°C, and axial strain sensitivity of -2.99 pm/µÎµ, respectively. Moreover, the sensor fabrication process is very simple and cost effective.