基于极限学习机和优化算法的润滑油添加剂种类识别与含量预测

为了快速识别润滑油中添加剂种类和含量，将添加剂硫化异丁烯(T321)、烷基二苯胺(T534)、硫代磷酸胺盐(T307)以不同配比混合在基础油中，使用极限学习机(ELM)对油样的红外光谱数据构建模型进行训练测试，并采用贪心算法、遗传算法(GA)对输入波段优化，筛选出最优波段区间组合以剔除相关性过高的波段从而提高运算效率. 测试结果表明:ELM模型可对润滑油添加剂进行有效的种类识别和含量预测，相比于传统理化检测方法是一种经济快速的新型润滑油添加剂检测手段；且经GA波段筛选优化后模型输出结果更具优势，对三种添加剂的种类识别准确率均达到100%、含量预测决定系数(R2)分别提升了43.8%、39.0%和24.4%.      

Theoretical and Experimental Study of the Diffraction Intensity of Gradient Refractive-Index Lenses

We develop the wave-optics approach for calculating the diffraction distribution of gradient refractiveindex lenses and observing the diffraction pattern of gradient refractive-index lenses in the experiments. The results of our calculation are in good agreement with the experimental results obtained. We show that the diffraction can be regarded as a method to check the quality of the refractive-index distributions of gradient refractive-index lenses.     

Charged surface states scattering in AlGaN/GaN heterostructures

The electron mobility limited by charged surface states scattering was calculated accurately based on a self-consistent Schrödinger and Poisson equations solver. For a AlGaN thickness less than 5 nm, this new scattering mechanism is the dominated scattering for electron mobility. Moreover, there is a “high electron mobility window” for a certain AlGaN thickness range from 4 to 7 nm, where the electron mobility can be as high as 3000 cm²/Vs at a low temperature, which is consistent with reported experimental data [1] (Cao and Jena, 2007).     

Stability of anti-reflection coatings via the self-assembly encapsulation of silica nanoparticles by diazo-resins

A modified silica nanoparticle (MSNP) solution was formed by the encapsulation of negatively charged silica nanoparticles by the UV-crosslinkable polycation oligomer diazo-resin (DAR). Appropriate DAR encapsulation concentrations were determined by use of zeta-potential and dynamic light scattering measurements. The MSNPs were used in conjunction with poly(styrene sulfonate) (PSS) to grow homogenous ionic self-assembled multilayer anti-reflection coatings. Stability was induced within the films by the exposure of UV-irradiation that allowed for crosslinking of the DAR and PSS. The films were characterized by UV/vis/IR spectroscopy and field emission scanning electron microscopy. The transmission and reflection levels were >98.5% and <0.05%, respectively. The refractive indices resided in the 1.25–1.26 range. The solvent stability was tested by sonication of the films in a ternary solvent (H₂O/DMF/ZnCl₂ 3:5:2 w/w/w).     

Theoretical investigation on submicron particle penetration through circular tubes inside a porous medium

The analytical infinite series solution of submicron particle transport in a circular tube bounded by a porous wall, such as a pinhole, is determined under the slip velocity boundary condition, and the solution is verified by using the experimental data in the previous studies for the specific cases. The results show that particle penetration rate increases with the increase of the porous parameter, the axial pressure drop, and the pinhole radius, whereas it decreases with increasing the pinhole length. The penetration rate of nano-particles are more sensitive to the variation of these parameters. However, the differences between the penetrations of particles ranging from 0.3 μm to 1 μm are not evident because the diffusion becomes weak gradually in this size range. In addition, a further comparison is performed between the analytical solution and the existing studies, and approximate expressions are presented for accurate calculation of particle penetration rate through pinholes appearing in porous materials including filter devices and masks.     

Ab initio studies of the structural,elastic, electronic and thermal properties of NiTi2 intermetallic

First principles calculations were performed in the framework of the density functional theory (DFT) using the Full Potential–Linear Augment Plane Wave method (FP–LAPW) within the generalized gradient approximation (GGA) to predict the structural, electronic, elastic and thermal properties of NiTi₂ intermetallic compound. By using the Wien2k all-electron code, calculations of the ground state and electronic properties such as lattice constants, bulk modulus, presure derivative of bulk modulus, total energies and density of states were also included. The elastic constants and mechanical properties such as Poisson’s ratio, Young’s modulus and shear modulus are estimated from the calculated elastic constants of the single crystal. Through the quasi-harmonic Debye model, the preasure and temperature dependences of the linear expansion coefficient, bulk modulus and heat capacity have been investigated. Finally, the Debye temperature has been estimated from the average sound velocity according to the predicted polycrystal bulk properties and from the single crystal elastic constants.     

Investigation of the performance of bipolar transverse plate ESP in the sintering flue control

In order to improve the particle collection efficiency of the electrostatic precipitator (ESP), a transverse plate ESP with bipolar discharge electrodes is proposed. The simulations of the velocity distribution have shown that when the inlet velocity is 1 m/s, within the range of 40 mm from electrode plate, the average velocities of windward side and leeward side are less than 0.7 m/s and 0.3 m/s respectively. It is clear that the velocity near the collection electrode plate of this bipolar ESP is much lower than that of the ordinary ESP at the same inlet velocity. This low velocity can lead to higher efficiency for fine dust collection due to the less dust re-entrainment in ESP. It is also found that the average velocities are getting lower when the distance between plates electrodes are greater than 150 mm in accordance with the simulations. The voltage current characteristics of the bipolar ESP are superior to the ordinary ESP. The pressure drop of the bipolar ESP is about 30% higher than that of the ordinary one. The dust penetration of the bipolar ESP is about 54% less than that of the ordinary ESP when the sintering dust with 25.405 μm mass median diameter is used as the test particulate under the condition of the electric field from 2.1 kV/cm to 3.2 kV/cm and the velocity from 1.0 m/s to 1.5 m/s.     

Dynamic and reversible electrowetting with low voltage on the dimethicone infused carbon nanotube array in air

Unremitting efforts have been intensively making for pursuing the goal of the reversible transition of electrowetting owing to its vital importance to many practical applications, but which remains a major challenge for carbon nanotubes due to the irreversible electrochemical damage. Herein, we proposed a subtly method to prevent the CNT array from electrochemical damage by using liquid medium instead of air medium to form a liquid/liquid/solid triphase system. The dimethicone dynamically refills in CNT arrays after removing of voltage that makes the surface back to hydrophobic, which is an elegant way to not only decrease energy dissipation in electrowetting process but also obtain extra energy in reversible dewetting process. Repeated cycles of in situ experiments showed that more than four reversible electrowetting cycles could be achieved in air. It worth mention that the in situ reversible electrowetting voltage of the dimethicone infused CNT array has been lowered to 2 V from 7 V which is the electrowetting voltage for the pure CNT array. The surface of the dimethicone infused CNT array can maintain hydrophobicity with a contact angle of 145.6° after four cycles, compared with 148.1° of the initial state. Moreover, a novel perspective of theoretical simulations through the binding energy has been provided which proved that the charged CNTs preferred binding with water molecules thereby replacing the dimethicone molecules adsorbed on the CNTs, whereas reconnected with dimethicone after removing the charges. Our study provides distinct insight into dynamic reversible electrowetting on the nanostructured surface in air and supplies a way for precise control of wettability in surface chemistry, smart phase-change heat transfer enhancement, liquid lenses, microfluidics, and other chemical engineering applications.     

Remarkable improvement of cyclic stability in Li–O2 batteries using ruthenocene as a redox mediator

Nonaqueous Li–O₂ batteries attract attention for their theoretical specific energy density. However, due to the difficulty of decomposition of Li₂O₂, Li–O₂ batteries have high charge overpotential and poor cycling life. So all kinds of catalysts have been studied on the cathode. Compared to heterogeneous solid catalysts, soluble catalysts achieve faster and more effective transport of electrons by reversible redox pairs. Here, we first report ruthenocene (Ruc) as a mobile redox mediator in a Li–O₂ battery. 0.01 mol/L Ruc in the electrolyte effectively reduces the charging voltage by 610 mV. Additionally, Ruc greatly increases the cycling life by four-fold (up to 83 cycles) with a simple ketjen black (KB) cathode. The results of SEM, XPS and XRD confirm that less discharge product residue accumulated after recharge. To verify the reaction mechanisms of the mediator, free energy profiles of the possible reaction pathways based on DFT are provided.     

Impulsive synchronization seeking in general complex delayed dynamical networks

The present paper investigates the issues of impulsive synchronization seeking in general complex delayed dynamical networks with nonsymmetrical coupling. By establishing the extended Halanay differential inequality on impulsive delayed dynamical systems, some simple yet generic sufficient conditions for global exponential synchronization of the impulsive controlled delayed dynamical networks are derived analytically. Compared with some existing works, the distinctive features of these sufficient conditions indicate two aspects: on the one hand, these sufficient conditions can provide an effective impulsive control scheme to synchronize an arbitrary given delayed dynamical network to a desired synchronization state even if the original given network may be asynchronous itself. On the other hand, the controlled synchronization state can be selected as a weighted average of all the states in the network for the purpose of practical control strategy, which reveals the contributions and influences of various nodes in synchronization seeking processes of the dynamical networks. It is shown that impulses play an important role in making the delayed dynamical networks globally exponentially synchronized. Furthermore, the results are applied to a typical nearest-neighbor unidirectional time-delay coupled networks composed of chaotic FHN neuron oscillators, and numerical simulations are given to demonstrate the effectiveness of the proposed control methodology.