Deterministic and probabilistic ship pitch prediction using a multi-predictor integration model based on hybrid data preprocessing,reinforcement learning and improved QRNN

The deterministic and probabilistic prediction of ship motion is important for safe navigation and stable real-time operational control of ships at sea. However, the volatility and randomness of ship motion, the non-adaptive nature of single predictors and the poor coverage of quantile regression pose serious challenges to uncertainty prediction, making research in this field limited. In this paper, a multi-predictor integration model based on hybrid data preprocessing, reinforcement learning and improved quantile regression neural network (QRNN) is proposed to explore the deterministic and probabilistic prediction of ship pitch motion. To validate the performance of the proposed multi-predictor integrated prediction model, an experimental study is conducted with three sets of actual ship longitudinal motions during sea trials in the South China Sea. The experimental results indicate that the root mean square errors (RMSEs) of the proposed model of deterministic prediction are 0.0254°, 0.0359°, and 0.0188°, respectively. Taking series #2 as an example, the prediction interval coverage probabilities (PICPs) of the proposed model of probability predictions at 90%, 95%, and 99% confidence levels (CLs) are 0.9400, 0.9800, and 1.0000, respectively. This study signifies that the proposed model can provide trusted deterministic predictions and can effectively quantify the uncertainty of ship pitch motion, which has the potential to provide practical support for ship early warning systems.     

Machine fault diagnosis with small sample based on variational information constrained generative adversarial network

In actual engineering scenarios, limited fault data leads to insufficient model training and over-fitting, which negatively affects the diagnostic performance of intelligent diagnostic models. To solve the problem, this paper proposes a variational information constrained generative adversarial network (VICGAN) for effective machine fault diagnosis. Firstly, by incorporating the encoder into the discriminator to map the deep features, an improved generative adversarial network with stronger data synthesis capability is established. Secondly, to promote the stable training of the model and guarantee better convergence, a variational information constraint technique is utilized, which constrains the input signals and deep features of the discriminator using the information bottleneck method. In addition, a representation matching module is added to impose restrictions on the generator, avoiding the mode collapse problem and boosting the sample diversity. Two rolling bearing datasets are utilized to verify the effectiveness and stability of the presented network, which demonstrates that the presented network has an admirable ability in processing fault diagnosis with few samples, and performs better than state-of-the-art approaches.     

The effects of washing a collagen sample prior to TEM examination

Transmission electron microscopy (TEM) is an important analysis technique to visualize (bio)macromolecules and their assemblies, including collagen fibers. Many protocols for TEM sample preparation of collagen involve one or more washing steps to remove excess salts from the dispersion that could hamper analysis when dried on a TEM grid. Such protocols are not standardized and washing times as well as washing solvents vary from procedure to procedure, with each research group typically having their own protocol. Here, we investigate the influence of washing with water, ethanol, but also methanol and 2-propanol, for both mineralized and unmineralized collagen samples via a protocol based on centrifugation. Washing with water maintains the hydrated collagen structure and the characteristic banding pattern can be clearly observed. Conversely, washing with ethanol results in dehydration of the fibrils, often leading to aggregation of the fibers and a less obvious banding pattern, already within 1 min of ethanol exposure. As we show, this process is fully reversible. Similar observations were made for methanol and propanol. Based on these results, a standardized washing protocol for collagenous samples is proposed.     

Colostrum source and passive immunity transfer in dairy bull calves

Colostrum is essential for good neonate health; however, it is not known whether different calves absorb the nutrients from colostrum equally well. In this study, the absorption of protein, IgG, and γ-glutamyl transferase was compared in newborn dairy bull calves for 1 wk after feeding colostrum from different sources. Thirty-five Holstein-Friesian bull calves were randomly allocated into 3 groups and fed colostrum within 4 h after birth. Group A calves (n = 12) were bottle fed colostrum from their own dam for 3 d. Colostrum from these group A cows was also used as foster cow colostrum for the group B calves (n = 12), such that each group A and B calf pair received identical colostrum from each milking of the respective group A dam (10% of birth weight per day). The group C calves (n = 11) were fed 1 bottle (2 L) of pooled colostrum and transition milk (referred to as pooled colostrum), as was the standard practice on the dairy farm. The pooled colostrum was collected from the other dairy cows on the farm 0 to 4 d postpartum and stored at 4°C for less than 12 h. Blood was sampled from calves before the first feeding and at 1, 2, 3, and 7 d after birth. Levels of total solids, total protein, and IgG were higher in the dam colostrum than in the pooled colostrum. At birth, there were no differences between the calf groups for any measurements, and all calves had very low IgG levels. After receiving colostrum, the glucose, plasma γ-glutamyl transferase, serum total protein, and IgG concentrations increased significantly in all calves. There were no differences in any blood measurements at any time point between the pairs of group A and group B calves that received colostrum from the same cow except for the IgG concentration 2 d after birth. However, the group A calves had a higher total serum protein level and IgG concentration than the group C calves for all the time points after the first feeding. The group B calves had a higher IgG concentration than the group C calves on d 1, 2, and 7 after birth. Compared with groups A and B, there was no difference in the proportion of calves in group C that failed to have passive immunity transferred adequately based on the IgG threshold (<10 g/L). Thus, the calves receiving identical colostrum from the same cow had the same levels of IgG, and even the pooled colostrum provided sufficient transfer of IgG as the calves were fed within 4 h after birth.     

A high-throughput chaotic advection microreactor for preparation of uniform and aggregated barium sulfate nanoparticles

A high-throughput (105.5 g/h) passive four-stage asymmetric oscillating feedback microreactor using chaotic mixing mechanism was developed to prepare aggregated Barium sulfate (BaSO₄) particles of high primary nanoparticle size uniformity. Three-dimensional unsteady simulations showed that chaotic mixing could be induced by three unique secondary flows (i.e., vortex, recirculation, and oscillation), and the fluid oscillation mechanism was examined in detail. Simulations and Villermaux–Dushman experiments indicate that almost complete mixing down to molecular level can be achieved and the prepared BaSO₄ nanoparticles were with narrow primary particle size distribution (PSD) having geometric standard deviation, σ_g, less than 1.43 when the total volumetric flow rate Q_total was larger than 10 ml/min. By selecting Q_total and reactant concentrations, average primary particle size can be controlled from 23 to 109 nm as determined by microscopy. An average size of 26 nm with narrow primary PSD (σ_g = 1.22) could be achieved at Q_total of 160 ml/min.     

DCNN深度特征与交替方向乘子的相关滤波跟踪

针对采用相关滤波的判别式目标跟踪遇到的瓶颈问题：由于目标快速移动引起边界效应，使得相关滤波器在学习与更新过程中可能会引入错误，最终错误的累积将导致跟踪失败。在采集深度学习特征与样本相似性度量的基础上，提出一种引入交替方向乘子方法的改进相关滤波目标跟踪算法，选择DCNN深度特征有效地表征待跟踪目标的初始状态，通过在线分类过程中样本相似性比对与半监督学习，辅助解决相关滤波器在学习过程中存在的自学习问题。所提目标跟踪算法特别适合训练样本为持续获得的、同时存储空间较小的机器学习过程，提高目标在快速运动与部分遮挡等复杂情况下的跟踪成功率，针对VOT2016标准测试视频的实验表明：当目标面临快速运动时，对比CN、SAMF、STC算法，所提DA-CFT跟踪算法将跟踪成功率分别由60.4%～73.4%、67.2%～82.9%、80.9%～88.1%提升至85.6%～91.0%。     

Optimal data-based binning for histograms and histogram-based probability density models

Histograms are convenient non-parametric density estimators, which continue to be used ubiquitously. Summary quantities estimated from histogram-based probability density models depend on the choice of the number of bins. We introduce a straightforward data-based method of determining the optimal number of bins in a uniform bin-width histogram. By assigning a multinomial likelihood and a non-informative prior, we derive the posterior probability for the number of bins in a piecewise-constant density model given the data. In addition, we estimate the mean and standard deviations of the resulting bin heights, examine the effects of small sample sizes and digitized data, and demonstrate the application to multi-dimensional histograms.     

二次侧非能动余热排出系统流动不稳定性研究

二次侧非能动余热排出（ASP）系统是压水堆核电厂应对全厂断电事故的重要措施之一。为研究ASP系统的运行特性，设计建造了ASP系统试验装置（ASPTF）。在ASPTF上开展了ASP系统运行稳定性影响试验，并对试验结果进行了理论分析。试验研究与理论分析结果表明：低压低功率下ASP系统中出现流动不稳定性；增加蒸汽管线或回水管线阻力系数可抑制ASP系统出现流动不稳定性；降低加热功率时ASP系统更易出现流动不稳定性；主泵运行状态影响ASP系统的输入功率进而影响系统的流动不稳定性。     

Understanding the determinants on household electricity consumption in Korea: OLS regression and quantile regression

The paper aims to investigate the determinants of household electricity consumption in Korea by using both the OLS regression and quantile regression. The results show that the effects of socio-demographic, dwelling, and electricity consumption characteristics on household electricity consumption may differ between two regressions and may differ across quantiles. We found that age group of household head, number of household, housing area, the number of household appliances, and refrigerator usage time were significant in all quantiles.