机器自监督学习的建筑物面要素几何形状度量

针对传统方法在度量建筑物面要素几何形状时,未能考虑形状认知的视觉特征因素且形状特征需要人为定义等问题,该文提出一种建筑物几何形状度量方法。首先,利用深度卷积神经网络的图像特征学习特性,结合自动编码机的自监督学习能力,构建基于机器自监督学习的建筑物面要素几何形状度量神经网络;其次,利用建筑物图像形状大数据对网络进行训练;最后,利用训练完成的神经网络识别并提取建筑物形状特征集并作为形状度量的结果。实验表明,该方法形状度量结果区分度高,一定程度上克服了人为定义形状特征的缺点,且与视觉感知结果基本一致。

Pre-compensation of Warpage for Additive Manufacturing

Additive manufacturing technologies enable the production of parts by successively adding layers. In powder-based technologies, each powder layer is selectively solidified following the respective cross-section of the parts either by the application of high-energy radiation or by the selective deposition of binder. By repeating the steps of layer deposition and selective solidification, parts are fabricated. The layer-wise build-up and the ambient conditions lead to warpage of the parts due to the temporarily and locally uneven distribution of shrinkage throughout the part. This leads to deviations in shape and dimension. The development of these technologies fosters a change fi＇om prototyping to manufacturing applications, As a consequence, higher standards regarding the shape and dimensional accuracy are required. Therefore, new strategies to minimize the resulting deformations are necessary to reduce rejects and widen the range of applications of the described technologies. In this paper, an empirical, a knowledge-based and a simulative approach for warpage compensation are introduced. They are all based on the pre-deformation of the digital 3D part geometry inverse to the expected deformation during manufacturing. The aim of the research is the development of a comprehensive method that enables users to improve their part-quality by supporting the pre-deformation process. Contrary to existing work, this method should not be process-specific but cover a wide range of additive manufacturing techniques. Typical forms of deformation of the processes laser sintering, laser beam melting and 3D printing （powder-binder） are presented and compensation strategies are disenssed. Finally, an outlook on the ongoing research is given.

Experimental Study of Effects of Tip Geometry on the Flow Field in a Turbine Cascade Passage

This study investigates the effects of blade tip geometry on the flow field of a turbine cascade at the incidence angle of 0 degree experimentally. The tests were performed in a low-speed turbine cascade wind tunnel. The Reynolds number based on the blade chord was about 172300 at the exit. Traverses of the exit flow field were made in order to measure the overall performance. The effects of using fiat tip and grooved tip with a chord-wise channel were studied. The case with the flat tip is referenced as the baseline. The tip clearances are all 1 mm measuring 0.84 percent of the blade span. The depth of channel is 2mm.The flow field at 10% chord downstream from the cascade trailing edge was measured at 38 span-wise positions and 26 pitch-wise positions using a mini five-hole pressure probe. The static pressure distribution on the tip end wall is measured at 16 pitch-wise stations and 17 chord-wise stations. Results show that there exists great pressure gradient in the pressure side for the fiat tip and the pressure side squealer tip, which means strong leakage flow. The pressure gradient from the pressure side to the suction side is greatly decreased for the grooved tip, and the resulting leakage flow is weaker. The core of the leakage vortex moves closer to the suction side for the pressure side squealer tip and farther away from the suction side for the suction side squealer tip. The pressure side squealer has little advantages over the fiat tip in improving the flow capacity and reducing the overall losses. The suction side squealer tip and grooved tip can effectively decrease the intensity of the tip leakage vortex, improve the flow capacity and reduce loss of the turbine cascade passage and the grooved tip performs the best.

Dual mean Minkowski measures of symmetry for convex bodies

We introduce and study a sequence of geometric invariants for convex bodies in finite-dimensional spaces, which is in a sense dual to the sequence of mean Minkowski measures of symmetry proposed by the second author. It turns out that the sequence introduced in this paper shares many nice properties with the sequence of mean Minkowski measures, such as the sub-arithmeticity and the upper-additivity. More meaningfully, it is shown that this new sequence of geometric invariants, in contrast to the sequence of mean Minkowski measures which provides information on the shapes of lower dimensional sections of a convex body, provides information on the shapes of orthogonal projections of a convex body. The relations of these new invariants to the well-known Minkowski measure of asymmetry and their further applications are discussed as well.