30 November 2015 in Continuum Mechanics and Thermodynamics
29 November 2019 in The Journal of Engineering
A shock tube‐based calibration installation for dynamic pressure transducers was designed and manufactured. The hardware of the calibration installation contained a shock tube, two pressure transmitters, three transducers for velocity measurement, a temperature transmitter, two data sampling cards and a industrial personal computer, etc. The software of the calibration installation was a data sampling and data processing system. The functions of the software included test conditions monitoring, performance parameters computation, and calibration reports generation, etc. The performance of the calibration installation was tested. The test result showed that the range of the step pressure was 9.60 kPa–1.101 MPa, the maximum rise time of the step pressure was no more than 1μs, the minimum duration of the step pressure plateau was longer than 4 ms, the fluctuation of the step pressure plateau was better than ± 2%, and the maximum 2nd order expand uncertainty of the step pressure was less than 4%. Thus the calibration installation was qualified to be a shock tube‐based dynamic pressure standard because that the performance of the calibration installation was verified to meet the requirements of ‘Verification Regulation of Dynamic Pressure transducers’ and ‘Verification Regulation of Dynamic Pressure standards’.
24 March 2025 in Image Analysis and Stereology
Deep learning methods have demonstrated significant advancements in single image super-resolution (SISR), with Transformer-based models frequently outperforming CNN-based counterparts in performance. However, due to the self-attention mechanism in Transformers, achieving lightweight models remains challenging compared to CNN-based approaches. In this paper, we propose a lightweight Transformer model termed Multi-Branch Mixer Transformer (MBMT) for SR. The design of MBMT is motivated by two main considerations: while self-attention excels at capturing long-range dependencies in features, it struggles with extracting local features. Secondly, the quadratic complexity of self-attention forms a significant challenge in building lightweight models. To address these problems, we propose a Multi-Branch Token Mixer (MBTM) to extract richer global and local information. Compared to other Transformer-based SR networks, MBTM achieves a balance between capturing global information and reducing the computational complexity of self-attention through its compact multi-branch structure. Specifically, MBTM consists of three parts: shifted window attention, depthwise convolution, and active token mixer. This multi-branch structure handles both long-range dependencies and local features simultaneously, enabling us to achieve excellent SR performance with just a few stacked modules. Experimental results demonstrate that MBMT achieves competitive performance while maintaining model efficiency compared to SOTA methods.