Experimental Research on the Similarity of Rime Icing on a Cylinder Rotating around Its Horizontal Axis

Ice accumulation on the blade of a wind turbine surface seriously threatens the operational safety of the turbine; therefore, the research on this problem is very important. In this paper, a new similarity criterion of icing shape for a rotational model was proposed based on the similarity criterion for translational motion models in the aviation field, and experimental studies on the similarity of the rotational model icing were carried out. To validate the similarity criterion, icing wind tunnel tests were carried out with aluminum cylinders with diameters of 40 mm and 20 mm. Key parameters for the experiment, such as wind speed, temperature, liquid water content, medium volume diameter, and test time, were selected based on the criterion. All the icing tests were carried out in a new self-designed icing wind tunnel test system based on natural low-temperature conditions. The icing shapes observed in the tests were confirmed after many repetitions. To quantitatively analyze the similarity between different sizes of ice shapes, a dimensionless method for evaluating the similarity of ice shapes of different sizes was defined based on the typical characteristics of ice shapes. The research results show that the similarity score between two sizes of ice shapes under different test conditions is 81%~90%. The accuracy and applicability of the icing shape similarity criterion were thus validated. The research results in this paper lay a theoretical and experimental foundation for exploring the icing shape similarity of a rotating model.

RESONANCE RISK AND MODE SHAPE MANAGEMENT IN THE FREQUENCY DOMAIN TO PREVENT SQUEAK AND RATTLE

Avoiding quality problems in passenger cars, such as squeak and rattle (S&R), has been a remarkable cost-saving consideration. The introduction of electric engines and less engaged drivers due to autonomous driving is expected to further stress the need for quieter cabins. However, the complexity of the virtual evaluation of S&R events has obstructed the practical treatment of these quality issues in the pre-design-freeze phases of product development. In this study, new quantified frequency-domain metrics are proposed to measure the risk for the generation of S&R in subsystem assemblies. The proposed metrics measure the resonance risk and the mode shape similarity in the critical interfaces for S&R. The calculations are done based on the system response in the frequency domain. Compared to the time-domain evaluation methods, the knowledge about the system excitation levels is not essential and the calculations are more time-efficient. The proposed metrics can be used in closed-loop design optimisation processes to involve S&R attributes in the pre-design-freeze attribute trade-off activities besides other attributes. In this work, these metrics were used in a two-stage optimisation problem to optimise the connection configuration in two industrial cases. As compared to the baseline design, the risk for S&R was reduced by improving the system behaviour in terms of resonance risk and mode shape similarity. This was achieved by applying some adjustments to the location of the fasteners while maintaining the same general connection configuration concept.

An image-computable model of human visual shape similarity

Shape is a defining feature of objects, and human observers can effortlessly compare shapes to determine how similar they are. Yet, to date, no image-computable model can predict how visually similar or different shapes appear. Such a model would be an invaluable tool for neuroscientists and could provide insights into computations underlying human shape perception. To address this need, we developed a model (‘ShapeComp’), based on over 100 shape features (e.g., area, compactness, Fourier descriptors). When trained to capture the variance in a database of >25,000 animal silhouettes, ShapeComp accurately predicts human shape similarity judgments between pairs of shapes without fitting any parameters to human data. To test the model, we created carefully selected arrays of complex novel shapes using a Generative Adversarial Network trained on the animal silhouettes, which we presented to observers in a wide range of tasks. Our findings show that incorporating multiple ShapeComp dimensions facilitates the prediction of human shape similarity across a small number of shapes, and also captures much of the variance in the multiple arrangements of many shapes. ShapeComp outperforms both conventional pixel-based metrics and state-of-the-art convolutional neural networks, and can also be used to generate perceptually uniform stimulus sets, making it a powerful tool for investigating shape and object representations in the human brain.

Forecasting of Hazard Zone due to Storm Surge Using SIND Model

We have developed the SIND (scientific interpolation for natural disasters) model to forecast natural hazard zone for storm surge. Most previous studies have been conducted to predict hazard zone with numerical simulations based on various scenarios. It is hard to predict hazard zone for all scenarios and to respond immediately because most numerical models are requested a long simulation time and complicated postprocess, especially in coastal engineering. Thus, in this study, the SIND model was developed to overcome these limitations. The principal developing methods are the scientific interpolation for risk grades and trial and error for parameters embedded in the governing equation. Even designed with hatch files, applying disaster characteristics such as the risk propagation, the governing equation for storm surge in coastal lines was induced from the mathematical solver, COMSOL Multiphysics software that solves partial differential equations for multiple physics using FEM method. The verification process was performed through comparison with the official reference, and the accuracy was calculated with a shape similarity indicating the geometric similarity of the hazard zone. It was composed of position, shape, and area criteria. The accuracy of about 80% in terms of shape similarity was archived. The strength of the model is high accuracy and fast calculation time. It took only less than few seconds to create a hazard map for each scenario. As future works, if the characteristics of other disasters would be understood well, it would be able to present risk propagation induced from each natural disaster in a short term, which should help the decision making for EAP.

Towards Measuring Shape Similarity of Polygons Based on Multiscale Features and Grid Context Descriptors

In spatial analysis applications, measuring the shape similarity of polygons is crucial for polygonal object retrieval and shape clustering. As a complex cognition process, measuring shape similarity should involve finding the difference between polygons, as objects in observation, in terms of visual perception and the differences of the regions, boundaries, and structures formed by the polygons from a mathematical point of view. In existing approaches, the shape similarity of polygons is calculated by only comparing their mathematical characteristics while not taking human perception into consideration. Aiming to solve this problem, we use the features of context and texture of polygons, since they are basic visual perception elements, to fit the cognition purpose. In this paper, we propose a contour diffusion method for the similarity measurement of polygons. By converting a polygon into a grid representation, the contour feature is represented as a multiscale statistic feature, and the region feature is transformed into condensed grid of context features. Instead of treating shape similarity as a distance between two representations of polygons, the proposed method observes similarity as a correlation between textures extracted by shape features. The experiments show that the accuracy of the proposed method is superior to that of the turning function and Fourier descriptor.

Sequential shape similarity for active contour based left ventricle segmentation in cardiac cine MR image

<abstract> <p>Delineation of the boundaries of the Left Ventricle (LV) in cardiac Magnetic Resonance Images (MRI) is a hot topic due to its important diagnostic power. In this paper, an approach is proposed to extract the LV in a sequence of MR images. In the proposed paper, all images in the sequence are segmented simultaneously and the shape of the LV in each image is supposed to be similar to that of the LV in nearby images in the sequence. We coined the novel shape similarity constraint, and it is called sequential shape similarity (SSS in short). The proposed segmentation method takes the Active Contour Model as the base model and our previously proposed Gradient Vector Convolution (GVC) external force is also adopted. With the SSS constraint, the snake contour can accurately delineate the LV boundaries. We evaluate our method on two cardiac MRI datasets and the Mean Absolute Distance (MAD) metric and the Hausdorff Distance (HD) metric demonstrate that the proposed approach has good performance on segmenting the boundaries of the LV.</p> </abstract>