This paper mainly investigated the effect of Mn/Ag ratio on microstructure and room temperature high-temperature (350 °C) tensile mechanical properties as-cast heat-treated Al-6Cu-xMn-yAg (x + y = 0.8, wt.%) alloys. The alloy has α-Al, Al2Cu, a small amount Al7Cu2 (Fe, Mn) Al20Cu2 (Mn, Fe)3 phases. After T6 heat treatment, massive dispersive fine θ′-Al2Cu phase (100~400 nm) is precipitated from matrix. influences quantity size precipitates; when 1:1, precipitation reaches highest smallest....

The quality of welding is directly related to the performance and life structure. nondestructive testing technology mainly used for defect defection, while X-ray can reliably reflect shape, location size defects. For images, manual inspection at present, it easily caused by subjective factors, such as professional level, which may lead low accuracy. This paper focuses on establishing an end-to-end automatic detection model defects improve accuracy efficiency based a deep learning algorithm....

For a long time, the extremely high chemical activity of magnesium melt with ceramics at temperatures makes it challenging to synthesize Mg-matrix syntactic foams (MgSFs) by casting. In this work, for first porosity AZ31B alloy matrix containing Al 2 O 3 hollow spheres (Al 3-HS ) were successfully synthesized gravity infiltration The coated MgO using MgCl solution after high-temperature calcination prevent interfacial reaction. results show that concentration strongly influences because...

Since the rotorcraft can easily be recognized by using micro-Doppler (m-D) signature of rotor blades, m-D effect induced micro-motion dynamics plays an important role in target recognition and classification. However, existing researches on blades pay little attention to mechanism time-domain time-frequency-domain flash phenomena. To comprehensively explain phenomena from physics, modeling are studied this paper. Firstly, for cannot represented as a rigid, homogeneous line nor several...

Augmented reality (AR) by analyzing the characteristics of scene, computer-generated geometric information which can be added to real environment in way visual fusion, reinforces perception world. Three-dimensional (3D) registration is one core issues AR. The key issue estimate sensor’s posture 3D and figure out objects scene. Recently, computer vision has made significant progress, but based on natural feature points space for AR system still a severe problem. There difficulty working...

Cylinder liners are important to automobile engines. The appearance quality will directly affect the life and safety of At present, inspection cylinder mainly relies on manual visual judgment, which is easily affected by subjective factors inspectors. This paper studies improved machine vision realize surface defect detection. It proposes improvement attention mechanism a feature fusion method locate classify defect. Experiments show that proposed in this has both accuracy speed, it can...

Cylinder liner plays an important role in the internal combustion engine. The surface defects of cylinder will directly affect safety and service life At present, defect detection mainly relies on manual visual inspection, which is easily affected by subjective factors inspectors. Aiming at bottleneck traditional inspection technology appearance this paper proposes a algorithm based deep learning to realize location classification. Based characteristics research object paper, improved YOLOv4...

The natural laminar flow airfoil shape design at transonic regime is solved using multi-objective evolutionary algorithms in this paper. A shock wave control bump used to reduce drag of flow, and the e N transition prediction method based on linear stability theory predict location. multi-island parallel implemented optimize equipped with for obtaining a larger region weaker simultaneously. Optimization experiment shows that it easy capture Pareto front minimization maximization. Results...

According to the characteristics of spinning targets, narrow-band radar echoes can be directly used for imaging targets. However, spurious peaks appear due azimuth down sampling with a low pulse repetition frequency (PRF). By exploiting sparsity compressed sensing (CS) theory adopted obtain super resolution image under sub-sampling condition. This paper mainly focuses on analyzing physical mechanism CS-based method. Firstly, radar's under-sampling echoes' model from rapidly targets is...

Drag reduction by laminarization, an innovative technology in aerodynamics design, provides the possibility to improve significantly aerodynamic performance of aircraft and remains be one most promising effective technologies. In this paper, design method for transonic natural laminar flow (NLF) wing at high Reynolds number is investigated details a new optimization model established solve conflicts between increase region shock wave strength trailing edge. order numerically handle...

Image matching plays an important role in Augmented Reality, Simultaneous Localization and Map ping (SLAM), unmanned. The key issue of image is the accuracy feature between adjacent images. Due to complicated environment, images through geometry runs into bottlenecks. In this paper, we focus on improving by incorporating instance-aware semantic segmentation Oriented FAST Rotated BRIEF (ORB) matching, which broadly utilized registration Visual SLAM. We segment objects pixel level qualify...

Larger workpiece's size visual detection needs to combine motion control and image mosaic ensure accuracy. Usually the complex flat parts have many test items therefore require tedious path planning which affect efficiency This paper researched developed a intelligent generation method using interaction. The is based on DXF drawing file analysis, combined with smart interaction, restructure sample point automatic positioning achieve plane processing planning. Application results shown that...

Micro-motion modulation caused by multi-rotor can be used as an important feature to identify different types of aircrafts. In this paper, modeling and systematical research on micro-motion characteristics aircraft's rotors is studied. Firstly, the mathematic model proposed. Secondly, time-domain micro-Doppler (m-D) signature are analyzed based model. Finally, various rotors, which compared with that single-rotor, simulated Simulation results show there a clear distinction m-D signatures...

Limited echoes and large Doppler bandwidth in a low PRF radar system is challenging problem for rapidly spinning targets imaging. To simultaneously improve the efficiency accuracy when imaging targets, this paper proposes novel fast sequential orthogonal matching pursuit algorithm with threshold (FSOMP-T). Firstly, complexity reduced by decreasing total iteration numbers computational burden each iteration. Secondly, designed to eliminate false reconstructed scatters, which can help image...

Abstract The AZ31B Magnesium alloy matrix syntactic foams containing Al 2 O 3 hollow spheres (Al − HS ) were successfully synthesized by the process of gravity infiltration casting. Extremely high chemical activity magnesium melt with ceramics at temperature makes it challenging to synthesize Mg-matrix (MgSFs) In this work, coated MgO using MgCl solution after high-temperature calcination prevent interfacial reaction. porosity is strongly influenced concentration because determines coverage...

An algorithm is presented for adaptation and self-learning of the hidden Markov model (HMM). It makes HMM-based speech recognition robust, so that well-trained models can be adapted to new speaking conditions or a speaker. The consists fact that, during recognition, all test tokens used augment current model. Both procedures increase size training set. was tested on speaker-dependent system whole Chinese vocabulary speaker-independent 0-9 digits. Experiments show very successful, both...