An efficient model for underwater noise prediction during pile driving.

Underwater noise pollution from pile driving is now attracting increasing attention. However, most of the current numerical and semi-analytical models for predicting the noise are still expensive and time-consuming, and the near-field noise and far-field noise have to be obtained from different models. This paper proposes an efficient semi-analytical solution for predicting underwater noise in both near field and far field with only one model, whose computational efficiency is orders of magnitude higher than that of the finite element model. It is the first time that the Baranov-Novak thin-layer model for soil-pile interaction has been extended to the subject of underwater noise prediction during pile installation, taking into account pile-fluid-soil interaction. The solutions are obtained using the Laplace transform and the variable separation method. By comparing the prediction results with the five reported research cases, it is shown that the error of the proposed model is within reasonable limits for both near-field and far-field noise predictions.

Quantitative characterization of the carbon/carbon composites components based on video of polarized light microscope.

The components of carbon/carbon (C/C) composites have significant influence on the thermal and mechanical properties, so a quantitative characterization of component is necessary to study the microstructure of C/C composites, and further to improve the macroscopic properties of C/C composites. Considering the extinction crosses of the pyrocarbon matrix have significant moving features, the polarized light microscope (PLM) video is used to characterize C/C composites quantitatively because it contains sufficiently dynamic and structure information. Then the optical flow method is introduced to compute the optical flow field between the adjacent frames, and segment the components of C/C composites from PLM image by image processing. Meanwhile the matrix with different textures is re-segmented by the length difference of motion vectors, and then the component fraction of each component and extinction angle of pyrocarbon matrix are calculated directly. Finally, the C/C composites are successfully characterized from three aspects of carbon fiber, pyrocarbon, and pores by a series of image processing operators based on PLM video, and the errors of component fractions are less than 15%.

Quantitative characterization of carbon/carbon composites matrix texture based on image analysis using polarized light microscope.

A quantitative characteristic method was proposed for characterizing the matrix texture of carbon/carbon(C/C) composites, which determined the mechanical and physical properties of C/C composites. Based on the cloud theory that was commonly used for uncertain reasoning and the transformation between quantitative and qualitative characterization, so the relationship between the extinction angle and texture types was built by the cloud models for describing the texture of microstructure, moreover, linguistic controllers were established to analyze the matrix texture in accordance with the features of the polarized light microscope (PLM) image. On this basis, the extinction angle could be calculated from the PLM image of the C/C composites. In contrast to the results of measurement, the errors between calculative values and measured values were maintained 1-2° in basically. Meanwhile, the PLM image of C/C composites was segmented by the component, in particular, the matrix with mixed textures was further segmented by the difference of texture. It means that the quantitative characterization of C/C composites matrix based on single PLM image has been realized.