Mutual information maximizing GAN inversion for real face with identity preservation

Recent generative adversarial networks (GANs) have yielded remarkable performance in face image synthesis. GAN inversion embeds an image into the latent space of a pretrained generator, enabling it to be used for real face manipulation. However, current inversion approaches for real faces suffer the dilemma of initialization collapse and identity loss. In this paper, we propose a hierarchical GAN inversion for real faces with identity preservation based on mutual information maximization. We first use a facial domain guaranteed initialization to avoid the initialization collapse. Furthermore, we prove that maximizing the mutual information between inverted faces and their identities is equivalent to minimizing the distance between identity features from inverted and original faces. Optimization for real face inversion with identity preservation is implemented on this mutual information-maximizing constraint. Extensive experimental results show that our approach outperforms state-of-the-art solutions for inverting and editing real faces, particularly in terms of face identity preservation.     

The formed surface characteristics of SiCf/SiC composite in the nanosecond pulsed laser ablation

For the advantages of high-temperature resistance, corrosion resistance and ultra-high hardness, SiC_f/SiC composite is becoming a preferred material for manufacturing aero-engine parts. However, the anisotropy and heterogeneity bring great challenges to the processing technology. In this study, a nanosecond pulsed laser is applied to process SiC_f/SiC composite, where the influence of the scanning speed and laser scanning direction to the SiC fibers on the morphology of ablated grooves is investigated. The surface characteristics after ablation and the involved chemical reaction of SiC_f/SiC are explored. The results show that the increased laser scanning speed, accompanied by the decreasing spot overlap rate, leads to the less accumulation of energy on the material surface, so the ablation effect drops. In addition, for the anisotropy of the SiC_f/SiC material, the obtained surface characteristics are closely dependent on the laser scanning direction to the SiC fibers, resulting in different groove morphology. The element composition and phase analysis of the machined surface indicate that the main deposited product is SiO₂ and the carbon substance. The results can provide preliminary technical support for controlling the machining quality of ceramic matrix composites.     

农村综合能源系统多层协同优化运行方法

农村综合能源系统通过多种能源的协同互补,在满足农村用户多元化用能需求的同时,能有效提升能源利用效率和用能经济性。首先,在考虑适应农村典型场景的基础上,提出了一种三层协同自律的农村综合能源分层协同运行优化框架。然后,基于农村综合能源系统典型设备,建立了三层农村综合能源系统源-储-荷联合优化调度模型以及相应的优化调度流程。调度模型中冬季通过对沼气发电机组的余热回收与空气源热泵协同对用户供热;夏季对沼气发电机组余热进行回收,并通过溴化锂制冷机与空气源热泵联合对农村用户供冷。最后,对农村综合能源系统多层协同优化方法进行了算例分析,结果表明该优化方法提高了农村居民用能的经济性,验证了所提方法的有效性。     

高职模具类人才培养新模式的研究与实践

为了解决传统教学无法全面培养人才的问题,采用"课程融合+时空耦合+过程结合"的方式开展人才培养研究。以注射模设计与制造生产过程为导向,构建大项目牵引下的跨越课程、跨越学期的课程融合新模式,利用信息通信技术及互联网平台,将互联网与传统教学进行深度融合,创造新的教学生态,构建多元多维互动的时空耦合教学新关系,以全方位评价为引导,构建课程多元多维的评价新机制。以3个班级为对象开展不同模式下的教学,对试验班级成绩进行分析,研究结果表明:课程融合+时空耦合+过程结合新模式培养的学生更好地完成了综合项目,获得了较好的教学效果。     

Laser damage resistance of plasma-sprayed alumina and honeycomb skeleton/alumina composite ceramic coatings

Al₂O₃ and honeycomb skeleton-Al₂O₃ composite coatings on Titanium alloy (Ti–6Al–4V) were prepared by atmospheric plasma spraying. A laser ablation experiment on as-sprayed coatings was performed. In this paper, the laser damage resistance, microstructure, phase composition of Al₂O₃ coatings were examined. 3D Dimensional Confocal Microscopy, Scanning Electron Microscopy (SEM), X-ray Diffraction (XRD), and Energy Dispersive Spectrometry (EDS) characterized the laser damage morphology, microstructure, phase composition, and element analysis, respectively. The influence of the honeycomb skeleton on the laser ablation damage on as-sprayed coatings was investigated by a comparative analysis of the laser damage morphology with different laser ablation times and gas flow. The results show that the honeycomb skeleton raises thermal conductivity and thermal diffusivity. Moreover, a “tower”-like dendrite was generated during the laser irradiation of the composite coating. The honeycomb skeleton refined the structure, suppressed crack propagation, and reduced the influence of gas flow on cracks. Under the same experimental laser ablation parameters, the laser damage area of the honeycomb skeleton-Al₂O₃ composite coating was smaller than that of the Al₂O₃ coating. It was demonstrated that the laser damage resistance of the honeycomb skeleton-Al₂O₃ composite coating was superior to that of the Al₂O₃ coating.     

Cold sintered,temperature-stable CaSnSiO5-K2MoO4 composite microwave ceramics and its prototype microstrip patch antenna

Dense (1-x)wt%CaSnSiO₅-xwt%K₂MoO₄ (CSSO-KMO) composite ceramics were fabricated by the cold sintering process at 180 °C under 400 MPa for 60 min. X-ray diffraction, Energy dispersive X-ray and Raman spectroscopy confirmed that CSSO and KMO coexisted without intermediate phases. As KMO weight fraction increased, relative permittivity (ε_r) and temperature coefficient of resonant frequency (τ_f) decreased and the microwave quality factor (Q×f, where f is resonant frequency) increased. Near-zero τ_f (-0.5 ppm/°C) was obtained for 65 wt%CSSO-35 wt%KMO with ε_r ～ 9.2 and Q×f ～ 6240 GHz. No chemical reaction between ceramic composites and silver was observed, demonstrating potential for cofiring with Ag-paste. A prototype antenna was fabricated from 65 wt%CSSO-35 wt%KMO composite ceramic with a bandwidth of 144 MHz @ -10 dB, a gain of 5.7 dBi and a total efficiency of 88.4 % at 5.2 GHz, suitable for 5 G mobile communication systems.     

Molding,patterning and driving liquids with light

When a laser beam induces surface tension gradient at the free surface of a liquid, a weak surface depression is expected and has been observed. Here we report giant depression and rupture in “optothermocapillary fluids” under the illumination of laser and sunlight. Computational fluid dynamics models were developed to understand the surface deformation and provided desirable physical parameters of the fluid for maximum deformation. New optothermocapillary fluids were created by mixing transparent lamp oil with different candle dyes. They can be cut open by sunlight and be patterned to different shapes and sizes using an ordinary laser show projector or a common laser pointer. Laser driving and elevation of optothermocapillary fluids, in addition to the manipulation of different droplets on their surface, were demonstrated as an efficient controlling method and platform for optofluidic operations. The fundamental understanding of light-induced giant depression and creation of new optothermocapillary fluids encourage the fundamental research and applications of optofluidics.     

An approach to predict lower-order dynamic behaviors of a 5-DOF hybrid robot using a minimum set of generalized coordinates

This paper presents an effective semi-analytical approach for predicting lower-order dynamics of a five degrees-of-freedom (DOF) hybrid robot named TriMule, which is composed of a 3-DOF parallel mechanism plus a 2-DOF A/C wrist. In this method, the governing equations of motion of limbs within the parallel mechanism are first formulated by finite element analysis (FEA) and then reduced to super-element models. This is followed by exploiting a general stiffness model of multiple DOF joints connecting the super-elements. These two threads lead to the reduced dynamic model of the parallel mechanism while keeping the full set of lower-order modes retained. Finally, the dynamic model of entire system is established by merging the models of parallel mechanism and wrist. The computational results show that the lower-order natural frequencies, mode shapes of the entire system, and the frequency response functions (FRFs) of the robot tool center point (TCP) estimated by the proposed approach have very good agreement with those obtained by a full order FE model and experimental modal tests. The merits of this approach lie in that the established model allows the full set of lower-order dynamics of the entire system to be predicted effectively and accurately by only using fourteen generalized coordinates.     

Resolving Deep Sub-Wavelength Scattering of Nanoscale Sidewalls Using Parametric Microscopy

The quantitative optical measurement of deep sub-wavelength features with sub-nanometer sensitivity addresses the measurement challenge in the semiconductor fabrication process. Optical scatterings from the sidewalls of patterned devices reveal abundant structural and material information. We demonstrated a parametric indirect microscopic imaging (PIMI) technique that enables recovery of the profile of wavelength-scale objects with deep sub-wavelength resolution, based on measuring and filtering the variations of far-field scattering intensities when the illumination was modulated. The finite-difference time-domain (FDTD) numerical simulation was performed, and the experimental results were compared with atomic force microscopic (AFM) images to verify the resolution improvement achieved with PIMI. This work may provide a new approach to exploring the detailed structure and material properties of sidewalls and edges in semiconductor-patterned devices with enhanced contrast and resolution, compared with using the conventional optical microscopy, while retaining its advantage of a wide field of view and relatively low cost.