Digestive protection of probiotic Lacticaseibacillus rhamnosus in Ricotta cheese by monoglyceride structured emulsions

This research aimed at studying the potential use of monoglyceride (MG) structured emulsions (MSEs) as delivery and protective systems for probiotic bacteria in Ricotta cheese. To this purpose, a low-fat commercial Ricotta cheese was added with MSEs formulated with milk, as water phase, and sunflower oil (MSE-SO) or anhydrous milk fat (MSE-AMF), as lipid phase. A commercial whole milk Ricotta cheese (W-RC) was considered as reference. A probiotic Lacticaseibacillus rhamnosus strain was inoculated as free cells in W-RC or embedded into the MSEs and added to the low-fat Ricotta at the same reference fat content. After physico-chemical characterisation, L. rhamnosus viability and sample destructuring behaviour upon in vitro digestion were evaluated. At the end of in vitro digestion, both W-RC and sample containing MSE-SO were unable to protect cells. By contrast, sample with AMF ensured a sufficient probiotic viability, even after 14 days of storage at 4 °C. This result was attributed to system composition and structure. During the gastric phase, the presence of caseins and MG-AMF mixed structures induced the formation of clots, entrapping and protecting cells against the acidic pH of the stomach, as confirmed by confocal micrographs and particle size. During the intestinal phase, cell viability was guaranteed by the formation of mixed micelles promoted by MG. It was demonstrated that microbial cells located near MG structures where they found protection.     

LPNN-based approach for LASSO problem via a sequence of regularized minimizations

In compressive sampling theory, the least absolute shrinkage and selection operator (LASSO) is a representative problem. Nevertheless, the non-differentiable constraint impedes the use of Lagrange programming neural networks (LPNNs). We present in this article the -LPNN model, a novel algorithm that tackles the LASSO minimization together with the underlying theory support. First, we design a sequence of smooth constrained optimization problems, by introducing a convenient differentiable approximation to the non-differentiable -norm constraint. Next, we prove that the optimal solutions of the regularized intermediate problems converge to the optimal sparse signal for the LASSO. Then, for every regularized problem from the sequence, the -LPNN dynamic model is derived, and the asymptotic stability of its equilibrium state is established as well. Finally, numerical simulations are carried out to compare the performance of the proposed -LPNN algorithm with both the LASSO-LPNN model and a standard digital method.     

Impact of high-intensity ultrasound on physical properties and degree of oxidation of lipase modified menhaden oil with caprylic acid and/or stearic acid

The purpose of this research was to determine the effect of high-intensity ultrasound (HIU) on physical properties, degree of oxidation, and oxidative stability of structured lipids (SLs). Caprylic acid (C) and stearic acid (S) were incorporated into menhaden oil using Lipozyme® 435 lipase to obtain five samples: (1) LC 20 (menhaden oil with 20% of C), (2) LC 30 (menhaden oil with 30% C), (3) LS 20 (menhaden oil with 20% S), (4) LS 30 (menhaden oil with 30% S), and (5) Blend C (menhaden oil with 16.24% C and 13.04% S). Samples were crystallized for 90 min at the following temperatures: (1) LC 20 at 15.5°C, (2) LC 30 at 17.5°C, (3) LS 20 at 24°C, (4) LS 30 at 30°C, and (5) Blend C at 18.0°C, and HIU was applied at the onset of crystallization. Physical properties, degree of oxidation, and oxidative stability were evaluated in sonicated and nonsonicated samples. All SLs had statistically higher G′ after sonication. Sonicated LS 30, LC 30, and Blend C had a higher melting enthalpy than the nonsonicated ones, while enthalpy values in sonicated LS 20 and LC 20 samples were not statistically different than the nonsonicated ones. No significant difference between sonicated and nonsonicated samples was observed in peroxide values (1.2 ± 0.1 meq/kg, p > 0.05) and in the oxidative stability index (6.3 ± 0.2 h, p > 0.05). These results showed that HIU was effective at changing physical properties without affecting the oxidation of the samples.     

基于结构光的星箭对接环相对位姿测量方法

空间非合作目标位姿测量是空间在轨维护的前提。面向空间机械手对空间载体自动抓捕的应用需求，提出了基于结构光的空间非合作目标视觉测量方法。该方法以双套双线结构光测量装置作为测量传感器，以空间载体上普遍存在的星箭对接环作为抓捕目标。根据对接环的共同特点，选择直线特征和点特征相结合作为相对位姿的求解特征；基于直线特征求解圆环平面法向量，基于点特征求解圆环圆心坐标，进而得到机械手工具坐标系与对接环坐标系间的相对位姿关系；基于多重几何约束实现了不同光照条件下图像上目标直线的鲁棒识别。建立了演示验证实验系统，在大量演示实验的基础上，进行了实验结果分析。     

基于分段加权的反向稀疏跟踪算法研究

为提高稀疏表示跟踪模型性能，提出一种分段加权的反向稀疏跟踪算法，将跟踪问题转化为在贝叶斯框架下寻找概率最高的候选对象问题，构造不同的分段权重函数来分别度量候选目标与正负模板的判别特征系数。通过池化来降低跟踪结果的不确定性干扰，选择正负模板加权系数差值最大的候选表示作为跟踪结果。实验表明，在光照变化、遮挡、快速运动、运动模糊情况下，所提出的算法可以确保跟踪结果的准确性和鲁棒性。     

多共振分量融合CNN的行星齿轮箱故障诊断

针对行星齿轮箱中各部件所激起的振动成分混叠、早期故障特征经常被较强的各级齿轮谐波成分以及环境噪声所湮没的问题，提出一种多共振分量融合卷积神经网络（multi-resonance component fusion based convolutional neural network，简称MRCF-CNN）的行星齿轮箱故障诊断方法。首先，对振动信号进行共振稀疏分解，得到包含齿轮谐波成分的高共振分量和可能包含轴承故障冲击成分的低共振分量；其次，构建多共振分量融合卷积神经网络，将得到的高、低共振分量和原始振动信号进行自适应的特征级融合，通过有监督的方式训练模型并进行行星齿轮箱故障诊断。对行星齿轮箱实验数据的分析结果表明，该方法能够有效分类行星齿轮箱中滚动轴承和齿轮的故障，成功对行星齿轮箱故障进行诊断，同时能够进一步增强卷积神经网络对振动信号所蕴含的故障信息的辨识能力。     

Durable,Sensitive, and Wide‐Range Wearable Pressure Sensors Based on Wavy‐Structured Flexible Conductive Composite Film

Flexible pressure sensors have potential applications in human motion monitoring and electronic skins. To satisfy the practical applications, pressure sensors with a high sensitivity, a low detection limit, a broad response range, and an excellent stability are highly needed. Here, a piezoresistive pressure sensor based on wavy‐structured single‐walled carbon nanotube/graphite flake/thermoplastic polyurethane (SWCNT/GF/TPU) composite film is fabricated by a prestretching process. Due to the random wavy structure, high conductivity, and good flexibility, the prepared sensor displays a low detection limit of 2 Pa, a wide sensing range of 0–60 kPa, and a high sensitivity of 5.49 kPa^?1 for 0–50 Pa. Furthermore, the sensor shows a remarkable repeatability of over 1.1 × 10⁴, 9.0 × 10³, and 2.0 × 10³ pressure loading/unloading cycles at 50 Pa, 500 Pa, and 30 kPa, respectively, and a fast responsibility of 100–150 ms of loading response time and 400–600 ms of relaxation time. Therefore, the pressure sensor is successfully adopted to monitor both the large‐scale human activities (e.g., walk and jump) and the small‐scale signals (e.g., wrist pulse). Furthermore, a sensor array is assembled to map the weight and shape of an object, indicating its various potential applications including human–machine interactions, human health monitoring, and other wearable electronics.     

Advanced visible-light photocatalytic property of biologically structured carbon/ceria hybrid multilayer membranes prepared by bamboo leaves

Biologically structured carbon/cerium dioxide materials are synthesized by biological templates. The microscopic morphology, structure and the effects of different oxidation temperatures on materials are characterized by scanning electron microscopy (SEM), transmission electron microscopy (TEM), X-ray diffraction (XRD) ultraviolet-visible light spectrum (UV–Vis) and X-ray Photoelectron Spectroscopy (XPS). Moreover, by splitting water under visible light irradiation, the hydrogen production is measured to test the photocatalytic property of these materials. The results show that materials made with bamboo biological templates which are immersed in 0.1 mol L^?1 of cerium nitrate solution, then carbonizated in nitrogen (700 °C) and oxidized in air (500–600 °C), can obtain the biological structure of bamboo leaves. The product is in the composition of hybrid multilayer membrane, which one is carbon membrane form plant cell carbonation and another is ceria membrane by nanoparticle self assembly. The best oxidation temperature is 550 °C and the band gap of carbon/cerium dioxide materials synthesized at this optimum oxidation temperature could be reduced to 2.75 eV. After exposure to visible light for 6 h, the optimal hydrogen production is about 302 μmol g^?1, which is much higher than that of pure CeO₂.