基于彩色图像处理的浮选尾煤灰分软测量研究

浮选尾煤灰分是浮选产品的一个重要指标。针对选煤厂浮选尾煤灰分多采用离线检测而无法实现在线准确测量，以及当前浮选软测量多采用单一的灰度图像从而导致软测量模型精度及适应性较差的问题，提出了一种基于彩色图像处理的浮选尾煤软测量方法，建立了基于最小二乘支持向量机(LSSVM)的浮选尾煤灰分软测量模型。模型以不同颜色空间的彩色特征、灰度均值以及浓度特征为输入变量，以尾煤灰分作为输出变量，采用粒子群优化算法对LSSVM模型参数进行优化。结果表明：所建立的尾煤灰分软测量模型可以较好地实现浮选尾矿灰分的在线预测，引入浮选尾矿图像的彩色特征可以提高尾煤图像分析的精度，预测精度达96.89%。研究成果在柳湾选煤厂现场应用，并取得了较好的尾矿灰分测量效果。     

Additional high-energy milling to enhance the performance of porcelain stoneware manufacturing

Porcelain stoneware tile is the best class of ceramic tiles regarding technical performance. Low porosity and high glass content are some of its highlighted characteristics. The manufacturing cost is highly dependent on the feldspar content and the processing flow rate. Certain technical bottlenecks in the manufacturing steps, such as milling, forming, and firing, are intrinsically associated with limitations in the processing properties, such as the dry strength, bulk density, and pyroplastic deformation. In this work, improvements in these properties were achieved using high-energy milling (HEM) after conventional milling (CM). This study was carried out on a pilot industrial scale in the milling stage. Six experimental runs were evaluated. Slurries were spray-dried. The powders were humidified with 6.5% moisture. Specimens were conformed under a specific pressure of 45 MPa. The firing was performed using temperature ranging from 1150 to 1230°C. The use of HEM, in comparison to traditional milling for the similar particle-size distributions, has increased the dry density, +0.2 g.cm⁻³, dry bending strength, +1.0 MPa, and decreased the pyroplastic deformation index, −1.10⁻⁵ cm⁻¹. These results allow an estimated thickness reduction of 10%.     

Direct processing of PbZr0.53Ti0.47O3 films on glass and polymeric substrates

This work reports on direct crystallization of PbZr_0.53Ti_0.47O₃ (PZT) thin films on glass and polymeric substrates, using pulsed thermal processing (PTP). Specifically, xenon flash lamps deliver pulses of high intensity, short duration, broadband light to the surface of a chemical solution deposited thin film, resulting in the crystallization of the film. Structural analysis by X-ray diffraction (XRD) and transmission electron microscopy show the existence of perovskite structure in nano-sized grains (≤5 nm). Local functional analysis by band excitation piezoelectric spectroscopy and electrostatic force microscopy confirm the presence of a ferroelectric phase and retention of voltage-written polarization for multiple days. Based on structural and functional analyses, strategies are discussed for optimization of pulse voltage and duration for the realization of crystalline ferroelectric thin films. For ∼200 nm-thick PZT films on glass substrates, 500 μs-long pulses were required for crystallization, starting with 100 pulses at 350 V, 10 or 25 pulses at 400 V and in general lower number of pulses at higher voltages (resulting in higher radiant energy). Overall power densities of >6.4 kW/cm² were needed for appearance of peaks corresponding to the perovskite phase in the XRD. Films on glass processed at 350–400 V had a higher degree of 111-oriented perovskite grains. Higher applied radiant energy (through increased pulse voltage or count) resulted in more random and/or partially 001-oriented films. For ∼1 μm-thick PZT films on polymeric substrates, 10 to 25 250 μs-long pulses at voltages ranging between 200 to 250 V, corresponding to power densities of ∼2.8 kW/cm², were optimal for maximized perovskite phase crystallization, while avoiding substrate damage.     

煤矿地面轨道运输矿车自动避障技术研究

介绍了煤矿地面轨道运输矿车自动避障技术。首先介绍了基于雷达的矿车前方障碍物检测技术及其相关算法,然后详细介绍了基于机器视觉的矿车前方障碍物检测系统和常用的图像处理方法以及相关的算法,最后介绍了雷达和机器视觉融合的障碍物检测技术在煤矿地面轨道运输矿车自动避障过程中的应用。     

Headspace fingerprinting approach to identify the major pathway influencing volatile patterns of vinasse-cured duck processed by high pressure,as well as its impact on physicochemical and sensory attributes

As a decisive attribute, flavour could be influenced by HP treatments through multiple physical and chemical pathways within the high pressure (HP)-assisted meat curing process. This investigation aimed to identify the major pathway influencing volatile flavour patterns of two representative vinasse-cured duck (VCD) products with HP treatments (150–300 MPa/15 min), including wet and dry types, by employing headspace fingerprinting as an untargeted approach. Results suggested that HP treatments greatly lowered moisture contents and increased Warner-Bratzler shear force and thiobarbituric acid reactive substances of the cured samples. According to multivariate models, the volatile flavour patterns of the HP-processed VCD could be clearly separated from the unprocessed samples, but the VCD pressurised at different intensities represented similar volatile fingerprinting, which was validated by e-nose analysis. The discriminant analysis (OPLS-DA) model outlined vinasse-derived ethanol, acetic acid, 3-methyl-1-butanol, 2-methyl-1-butanol, phenethyl alcohol and 2-methyl-3-octanone as the major discriminant aromas across the unpressurised and pressurised samples.     

Role of the bolus degree of structure on the protein digestibility during in vitro digestion of a pea protein-fortified sponge cake chewed by elderly

This study investigated the digestibility of proteins in a pea protein-fortified sponge cake, as well as the impact of the degree of structure of the bolus produced by elderly subjects on the digestibility of proteins by combining ex vivo and in vitro approaches via the standardized protocol INFOGEST. The sponge cakes were consumed by a group of 20 elderly subjects with contrasting physiology, their boli were recovered just before swallowing, and their apparent viscosity was measured to delineate the bolus degree of structure. According to this criterion, two pools were formed with boli from subjects selected at the extremes: low viscosity and high viscosity, with apparent viscosity values (at 120 s⁻¹) of 124 ± 18 and 208 ± 19 Pa s, respectively. The sponge cakes and the two pools underwent in vitro digestion. Protein hydrolysis kinetics was followed by measuring the released primary amino groups (NH₂) and by sodium-dodecyl-sulfate polyacrylamide gel electrophoresis at different time points. For all samples, the representative bands of pea proteins disappear gradually during digestion, accompanied by the appearance of bands indicating the presence of proteins with M_W < 15 kDa. In addition, the NH₂ concentrations increase over time and do not differ between sponge cake and pea protein isolate. Moreover, the degree of structure of the food bolus has no significant effect on the concentration of NH₂ released. These results showed that pea proteins in a fortified sponge cake are bioaccessible under standardized conditions and that the degree of structure of the bolus did not influence protein digestibility for these foods.     

The effects of ozone,ultrasound and coating with shellac and lysozyme–chitosan on fresh egg during storage at ambient temperature – part 1: interior quality changes

The objective of this research was to investigate and compare the effect of different treatments including gaseous ozone (6-ppm/4-min), ultrasound (450 W/4 min) and coating of shellac, lysozyme–chitosan (L-C) on fresh eggs internal quality during storage for six weeks at ambient temperature. The internal quality attributes such as weight loss (WL), albumen viscosity, Haugh unit (HU), yolk index, total soluble solids, albumen and yolk's pH were monitored. Control groups had the worst HU (50.04), WL (8.35%), albumen pH (9.27) and viscosity (7.72 mPa·s) values after storage. The best interior qualities were obtained by shellac coating [HU (72.37), WL (1.37) and viscosity (30.10 mPa·s)] and sonication (albumen pH: 8.22). Thus, the shellac-coated eggs maintained all the internal quality parameters at the highest values, while the ozone and ultrasound only helped to keep the internal quality and enhance the shelf life of eggs.     

CLOI-NET: Class segmentation of industrial facilities’ point cloud datasets

Shape segmentation from point cloud data is a core step of the digital twinning process for industrial facilities. However, it is also a very labor intensive step, which counteracts the perceived value of the resulting model. The state-of-the-art method for automating cylinder detection can detect cylinders with 62% precision and 70% recall, while other shapes must then be segmented manually and shape segmentation is not achieved. This performance is promising, but it is far from drastically eliminating the manual labor cost. We argue that the use of class segmentation deep learning algorithms has the theoretical potential to perform better in terms of per point accuracy and less manual segmentation time needed. However, such algorithms could not be used so far due to the lack of a pre-trained dataset of laser scanned industrial shapes as well as the lack of appropriate geometric features in order to learn these shapes. In this paper, we tackle both problems in three steps. First, we parse the industrial point cloud through a novel class segmentation solution (CLOI-NET) that consists of an optimized PointNET++ based deep learning network and post-processing algorithms that enforce stronger contextual relationships per point. We then allow the user to choose the optimal manual annotation of a test facility by means of active learning to further improve the results. We achieve the first step by clustering points in meaningful spatial 3D windows based on their location. Then, we apply a class segmentation deep network, and output a probability distribution of all label categories per point and improve the predicted labels by enforcing post-processing rules. We finally optimize the results by finding the optimal amount of data to be used for training experiments. We validate our method on the largest richly annotated dataset of the most important to model industrial shapes (CLOI) and yield 82% average accuracy per point, 95.6% average AUC among all classes and estimated 70% labor hour savings in class segmentation. This proves that it is the first to automatically segment industrial point cloud shapes with no prior knowledge at commercially viable performance and is the foundation for efficient industrial shape modeling in cluttered point clouds.