Application of robust machine learning methods to modeling hydrogen solubility in hydrocarbon fuels

Having accurate information about the hydrogen solubility in hydrocarbon fuels and feedstocks is very important in petroleum refineries and coal processing plants. In the present work, extreme gradient boosting (XGBoost), multi-layer perceptron (MLP) trained with Levenberg–Marquardt (LM) algorithm, adaptive boosting support vector regression (AdaBoost?SVR), and a memory-efficient gradient boosting tree system on adaptive compact distributions (LiteMORT) as four novel machine learning methods were used for estimating the hydrogen solubility in hydrocarbon fuels. To achieve this goal, a database containing 445 experimental data of hydrogen solubilities in 17 various hydrocarbon fuels/feedstocks was collected in wide-spread ranges of operating pressures and temperatures. These hydrocarbon fuels include petroleum fractions, refinery products, coal liquids, bitumen, and shale oil. Input parameters of the models are temperature and pressure along with density at 20 °C, molecular weight, and weight percentage of carbon (C) and hydrogen (H) of hydrocarbon fuels. XGBoost showed the highest accuracy compared to the other models with an overall mean absolute percent relative error of 1.41% and coefficient of determination (R²) of 0.9998. Also, seven equations of state (EOSs) were used to predict hydrogen solubilities in hydrocarbon fuels. The 2- and 3-parameter Soave-Redlich-Kwong EOS rendered the best estimates for hydrogen solubilities among the EOSs. Moreover, sensitivity analysis indicated that pressure owns the highest influence on hydrogen solubilities in hydrocarbon fuels and then temperature and hydrogen weight percent of the hydrocarbon fuels are ranked, respectively. Finally, Leverage approach results exhibited that the XGBoost model could be well trusted to estimate the hydrogen solubility in hydrocarbon fuels.     

Enhancement of super-hydrophilic/underwater super-oleophobic performance of ceramic membrane with TiO2 nanowire array prepared via low temperature oxidation

A gradient porous ceramic membrane with surface super-hydrophilic and underwater super-oleophobic performance was prepared by combining hydrogel directional freezing method and low temperature oxidation process. The effects of solid contents and sintering temperature on the ceramic membrane matrix were examined. The reaction time and synthesis temperature on the TiO₂ nanowire array were also evaluated. In addition, the related effects on pore size distribution, permeation flux, contact angle, and oil-in-water emulsion separation were systematically investigated. The ceramic membrane matrix pore size changed from 0.5 μm to 25 μm gradually, indicating the gradient structure controlled by the growth of ice. The super-hydrophilic and underwater super-oleophobic performance of ceramic membrane surface was obtained with surface modification by TiO₂ nanowire array, and the surface water contact angle and underwater oil contact angle were less than 5° and over 158°, respectively. The bonding strength between TiO₂ nanowire and ceramic membrane matrix was high enough to withstand ultrasonic waves. The ceramic membrane modified with TiO₂ nanowire array was used for 1000 ppm diesel oil-in-water emulsion separation, and the stable separation efficiency and flux were about 97% and 100–200 L/(m² h bar) even after 10 filtration cycles.     

Effects of alkali metal ion on imprinting GRIN microstructure in GeS2-Ga2S3-MCl (M=Na,K, Cs) glasses for visible to mid-infrared microgratings

Gradient refractive index (GRIN) micro-optics present unique opportunities for control of the chromatic properties, new degrees of freedom for optical design as well as the potential for use in new optical system applications. GRIN microgratings were imprinted in GeS₂-Ga₂S₃-MCl (M = Na, K, Cs) chalcohalide glasses by microthermal poling, and the effects of the type and concentration of alkali cations on their performance were investigated. Two effective imprinting formation regions of the GRIN microstructure based on the poling saturation voltage (U_s) and glass composition are observed at fixed poling time and temperature. The U_s increases from 140 to 750 and 2600 V in accordance with the activation energy (E_a) of alkali ions (Na⁺ to K⁺ and Cs⁺) increasing from 45.15 to 58.62 and 92.58 kJ/mol for studied samples. The saturated numbers of diffraction order (N_s) of the gratings in these samples are 7, 9 and 6, respectively, the highest number being provided by the K⁺-containing sample. This is in accordance with imprinting-induced phase differences (0.14λ, 0.19λ and 0.09λ) measured in the fabricated samples containing Na⁺, K⁺ and Cs⁺ ions. Furthermore, the U_s of samples decreases from 1500 to 300 V with four concentrations of K⁺ from 10 to 30%, associated with their E_a of K⁺ decreasing from 69.62 to 53.46 kJ/mol, while N_s increases from 8 to 14, which is attributed to the increase of the phase difference in the GRIN structures. The controllable GRIN microstructures are realized by adjusting the type and concentration of alkali cations in chalcohalide glasses, which is expected to drive the design of broadband GRIN microgratings.     

基于k-shell的复杂网络最短路径近似算法

复杂网络最短路径经典算法的处理效率较低，不适用于大规模复杂网络，而现有近似算法通用性有限，且计算准确率不理想，不能满足规模日益扩大的复杂网络中的最短路径计算需求。针对于此，提出基于[k]-shell的复杂网络最短路径近似算法。算法利用节点的[k]-shell值进行网络划分并引导搜索路径，利用超点聚合处理[k]-shell子网来降低路径搜索中节点和连边的规模，通过在路径搜索过程使用双向搜索树方法提高算法的计算效率和准确率。实验结果表明，算法通用性较好，在现实与仿真大规模复杂网络中均具有较高的计算效率和准确率。     

A gradient boosting decision tree approach for insider trading identification: An empirical model evaluation of China stock market

Insider trading is a kind of criminal behavior in stock market by using nonpublic information. In recent years, it has become the major illegal activity in China’s stock market. In this study, a combination approach of GBDT (Gradient Boosting Decision Tree) and DE (Differential Evolution) is proposed to identify insider trading activities by using data of relevant indicators. First, insider trading samples occurred from year 2007 to 2017 and corresponding non-insider trading samples are collected. Next, the proposed method is trained by the GBDT, and initial parameters of the GBDT are optimized by the DE. Finally, out-of-samples are classified by the trained GBDT–DE model and its performances are evaluated. The experiment results show that our proposed method performed the best for insider trading identification under time window length of ninety days, indicating the relevant indicators under 90-days time window length are relatively more useful. Additionally, under all three time window lengths, relative importance result shows that several indicators are consistently crucial for insider trading identification. Furthermore, the proposed approach significantly outperforms other benchmark methods, demonstrating that it could be applied as an intelligent system to improve identification accuracy and efficiency for insider trading regulation in China stock market.     

2.5 GHz城市环境电磁态势的计算与补偿

复杂城市空间的电磁态势分布可以通过电磁场数值计算的方法进行预测。文章使用矩量法(MOM)结合一致性几何绕射理论(UTD)的方法对2.5 GHz频率下某小区环境中的电磁态势进行了仿真计算以及实际测量。场景中的树木结构在此频率下显示出比较强烈的波动性,对电磁仿真计算结果产生了不可忽略的影响,文中将其对仿真计算结果的影响进行了单独建模仿真,给出其产生影响的经验值,对仿真结果进行修正。修正结果与测量结果对比证明该经验值在该频率下可以有效补偿计算结果的缺失。     

Design structure matrix (DSM) method application to issue of modeling and analyzing the fault tree of a wind energy asset

The perpetual energy production of a wind farm could be accomplished (under proper weather conditions) if no failures occurred. But even the best possible design, manufacturing, and maintenance of a system cannot eliminate the failure possibility. In order to understand and minimize the system failures, the most crucial components of the wind turbines, which are prone to failures, should be identified. Moreover, it is essential to determine and classify the criticality of the system failures according to the impact of these failure events on wind turbine safety. The present study is processing the failure data from a wind farm and uses the Fault Tree Analysis as a baseline for applying the Design Structure Matrix technique to reveal the failure and risk interactions between wind turbine subsystems. Based on the analysis performed and by introducing new importance measures, the “readiness to fail” of a subsystem in conjunction with the “failure riskiness” can determine the “failure criticality.” The value of the failure criticality can define the frame within which interventions could be done. The arising interventions could be applied either to the whole system or could be focused in specified pairs of wind turbine subsystems. In conclusion, the method analyzed in the present research can be effectively applied by the wind turbine manufacturers and the wind farm operators as an operation framework, which can lead to a limited (as possible) design‐out maintenance cost, failures' minimization, and safety maximization for the whole wind turbine system.     

太阳能及空气源热泵复合供热系统故障诊断研究与应用

为确保复杂太阳能热水系统的稳定运行,有必要采集实时运行数据从中获取故障特征信号,通过专家系统进行自动故障诊断。采用故障树分析方法,以太阳能热利用系统典型故障为例构建故障树,求解最小割集,并以此为基础建立诊断知识库;运用正向推理结合人工辅助决策机制进行故障诊断和定位,建立故障诊断专家系统。开发故障诊断软件,并在工程中进行测试,验证故障辨识的准确性。