Search

Therapeutic drug monitoring (TDM) has played an important role in clinical medicine for precise dosing. Currently, chromatographic technology and immunoassay detection are widely used in TDM and have met most of the needs of clinical drug therapy. However, some problems still exist in practical applications, such as complicated operation and the influence of endogenous substances. Surface plasmon resonance (SPR) has been applied to detect the concentrations of small molecules, including pesticide residues in crops and antibiotics in milk, which indicates its potential for in vivo drug detection. In this study, a new SPR-based biosensor for detecting chloramphenicol (CAP) in blood samples was developed and validated using methodological verification, including precision, accuracy, matrix effect, and extraction recovery rate, and compared with the classic ultra-performance liquid chromatography-ultraviolet (UPLC-UV) method. The detection range of SPR was 0.1-50 ng/mL and the limit of detection was 0.099 ± 0.023 ng/mL, which was lower than that of UPLC-UV. The intra-day and inter-day accuracies of SPR were 98%-114% and 110%-122%, which met the analysis requirement. The results show that the SPR biosensor is identical to UPLC-UV in the detection of CAP in rat blood samples; moreover, the SPR biosensor has better sensitivity. Therefore, the present study shows that SPR technology can be used for the detection of small molecules in the blood samples and has the potential to become a method for therapeutic drug monitoring.

The nuclear charge pickup cross sections of 40Ar on polyethylene(CH2), carbon, aluminum, copper and lead targets at the highest energy of 495 MeV/n were measured using CR-39 nuclear track detector. The cross sections for H were calculated from those measured on C and CH2 targets. The dependence of charge pickup cross section on target mass was studied, it was found that the nuclear charge pickup cross section exhibits a linear dependence on the target mass.

The pricing and reverse logistics network design problem in remanufacturing has attracted considerable attention in recent years due to increasingly serious environmental problems. In this study, we consider a pricing and reverse logistics network design problem with price-dependent return quality uncertainty. To handle the high uncertainty in return quality, we propose a distributionally robust risk-averse model to safeguard the profits of investors in extreme situations. We propose a Benders decomposition approach to solve the proposed model. It is enhanced through valid inequalities, local branching, in-out variant methods and scenario-based aggregated cuts. Computational experiments demonstrate that the distributionally robust model can effectively hedge against high uncertainty and that the enhanced Benders decomposition methods significantly outperform their classical counterparts and the off-the-shelf solver Gurobi. Lastly, managerial insights are explored, and future research directions are outlined.

Perovskite solar cells (PSCs) are attracting much attention and are on the way to commercialization. However, some modules are subject to reverse bias in actual fields, so it is meaningful to investigate the reverse-bias behavior of PSCs. Herein, an in-situ temperature and current measurement technique was developed. Intriguingly, some hot spots were observed and then quickly disappeared in the reverse biased PSCs, along with the simultaneous increase in the current and local temperature. Also, the potential mechanism has been revealed and analyzed. An abnormal bulge in the perovskite film was found at the hot spot. Accordingly, the appearance and disappearance of hot spots were perfectly explained by band bending and tunneling current caused by ion accumulation. Additionally, statistical analysis suggested that sparkling hot spots were related to reverse voltage and efficiencies of PSCs. The research provides a great significance for the study of PSCs under reverse bias.

Given an image generated by the convolution of point sources with a band-limited function, the deconvolution problem involves reconstructing the source number, positions, and amplitudes. This problem is related to many important applications in imaging and signal processing. It is well known that it is impossible to resolve the sources when they are sufficiently close in practice. Rayleigh investigated this problem and formulated a resolution limit known as the Rayleigh limit for the case of two sources with identical amplitudes. However, many numerical experiments have demonstrated that stable recovery of the sources is possible even if the sources are separated below the Rayleigh limit. In this study, a mathematical theory for the deconvolution problem in one dimension is developed. The theory addresses the problem when the source number can be recovered exactly from noisy data. The key component is a new concept called the “computational resolution limit,” which is defined as the minimum separation distance between the sources such that exact recovery of the source number is possible. This new resolution limit is determined by the signal-to-noise ratio and the sparsity of sources, as well as the cutoff frequency of the image. Quantitative bounds for this limit are derived, and they demonstrate the importance of the sparsity and signal-to-noise ratio for the recovery problem. The stability of recovering the source positions is also analyzed under a condition on their separation distances. Moreover, a singular value thresholding algorithm is proposed to recover the source number for a cluster of closely spaced point sources and to verify our theoretical results regarding the computational resolution limit. The results are based on a multipole expansion method and a nonlinear approximation theory in Vandermonde space.

This paper introduces a new charging service mode that is emerging in the market for electric vehicles called the valet charging service. Through this service mode, a company provides free valet charging services to users that are far away from charging stations, and users can request the fee-based service when they are unwilling to charge electric vehicles themselves. The valet charging service is a new service mode that promises to alleviate users’ worries regarding the lack of nearby charging stations. However, whether such a mode truly benefits a specified system remains open to careful study. We build the charging station locating and sizing model with the valet charging service under demand uncertainty. We first formulate this problem as a two-stage stochastic mixed-integer model (TSMIP). Then, considering that the demand probability distribution is difficult to estimate, we reformulate TSMIP into a risk-averse two-stage stochastic mixed-integer model (RTSMIP) with the ambiguity set. Next, we propose an SAA-based hybrid-cut L-shaped algorithm to solve this model. By comparing the total cost of the system with and without the valet charging service under different scenarios through hypothetical numerical experiments and a realistic case study in Shanghai, meaningful insights are provided that include identifying the conditions in which the valet charging service works well and how it influences the location and capacity of charging stations. These insights provide some understanding of this new service mode that can help companies make rational decisions.

Since Chinese government work report in 2010 put forward the content of strengthening the construction of smart grids, the construction of smart grids has become one of the important contents of the national development strategy. In the in-depth study of smart grids, the issue of security defense systems and disaster prevention frameworks has always been Research hotspots by scholars at home and abroad, but most of my country’s existing security defenses are aimed at large-scale interconnected systems such as transmission grids, and there is a lack of research on disaster prevention of distribution networks. Therefore, in this research context, this article starts from the perspective of multi-source information fusion, takes into account the uncertainty of distribution network information and other factors, constructs a distribution network disaster response resilience index system and a distribution network resilience assessment model, and uses accidents at the same time. The scenario optimization method improves the disaster warning capability of the distribution network, and provides a scientific basis for the distribution network to take active defense measures. Then, a coordinated defense evolution strategy for the distribution network considering the optimization of accident scenarios was established. This strategy takes limited resources as the constraint, takes the minimum load loss as the objective function, and considers the economics of strategy conversion. Power supply equipment output, load transfer, and switching linkage mechanism between the three load is established, forming islands of security, and network transfer, emergency coordination of the three islands of the distribution network evolution strategy of active defense. This strategy looks for a directional evolution path from real-time accident scenarios to defense strategies from both horizontal and vertical perspectives, and realizes the coordinated control of the distribution network’s active defense strategy locally and globally. Finally, this paper takes the ice disaster climate as an example for simulation verification to prove the comprehensiveness of the risk assessment model and the rationality of the security defense method, and prove that the defense strategy can meet the online adaptive evolution requirements of the active defense of the distribution network, while taking into account The local autonomy and global optimal coordinated control of the active defense of the distribution network make up for the shortcomings of the current active defense strategy and are of great significance to further improve the security defense level of the active distribution network.

Area under the ROC curve (AUC) is a widely used performance measure for imbalanced classification. Oftentimes, the ubiquitous imbalanced data such as financial records from fraud detection or genomic data from cancer diagnosis contains sensitive information, and therefore it is of practical and theoretical importance to develop privacy-preserving AUC maximization algorithms. In this paper, we propose differentially private empirical risk minimization (ERM) for AUC maximization, and systematically study their privacy and utility guarantees. In particular, we establish guarantees on the generalization (utility) performance of the proposed algorithms with fast rates. The technical novelty contains fast rates for the regularized ERM in AUC maximization, which is established using the peeling techniques for Rademacher averages [1] and properties of U-Statistics [2,3] to handle statistically non-independent pairs of examples in the objective function, and a new error decomposition to handle strongly smooth losses (e.g. least square loss). In addition, we revisit the private ERM with pointwise loss [4,5] and show optimal rates can be obtained using the uniform convergence approach.

Graphene oxide (GO) has been considered as a promising material to develop advanced nanofiltration membranes to resolve the current worldwide water scarcity issue, benefiting from its extraordinary physicochemical properties. However, loose-GO-nanosheet-stacking rendered membrane compaction during operation significantly harms the mass transfer of GO membranes. Here, we proposed an ice-crystal templating approach to simultaneously tailor the two mass transfer channels of GO membranes, i.e., nanochannels originated from the interlayer spacing and microporous defects arising from poorly stacking of nanosheets. The tunability of the ice-crystal templating strategy was verified by low-field nuclear magnetic resonance (LF-NMR) coupled with X-ray diffraction patterns technique. The result demonstrated that the interlayer spacing can be precisely tuned from 7.5 to 9.3 A while the volume of microporous defects can be adjusted from 2.9% to 24.6%. Thus, the optimized GO membrane (M2) was utilized for desalination of dye/NaCl mixtures and accomplished a high separation performance, for example, high water permeability of ∼22.6 LMH/bar (10-fold enhancement compared with traditional GO membranes), 100% rejection to EB dyes, 19% rejection to NaCl, and long-term running stability (180 h). Therefore, the innovative ice-crystal templating fabrication techniques opens the door for the design of high-efficiency 2D material-based membranes.

Silica-based ceramic cores are widely used in the manufacturing of hollow, nickel-based, superalloy turbine blades. However, elemental Hf, Ti, Al, and other active metals in the superalloy can react with silica-based ceramic cores during casting, resulting in a reduction in the quality of the turbine blades. In this study, both plasma spraying and sol-gel dipping methods were used to prepare alumina coatings on silica-based ceramic substrates to prevent the interfacial reaction. The performance of the alumina coatings prepared by both methods was evaluated by comparative analysis of the surface roughness, bonding interface morphologies, and the adhesive characteristics of the coating. The plasma-sprayed alumina coating has a roughness greater than 5 μm and peeled away from the substrate due to the difference in thermal expansion between SiO2 and Al2O3 at temperatures above 1500 °C, rendering the silica-based substrate with the plasma-sprayed alumina coating unfit for the application requirements of the casting process. The alumina coating prepared by the sol-gel dipping method improved the roughness of the substrate from Ra 2.39 μm to Ra 1.83 μm, and no peeling was observed when heated to 1550 °C for 30 min due to the pinning characteristics of the coating on the substrate. Furthermore, the interfacial reaction between the DZ125 superalloy melt and the silica-based substrate coated with alumina by sol-gel dipping method were investigated. The alumina coating effectively inhibited the interfacial reaction and no reaction products were detected during the directional solidification with pouring temperature of 1550 °C and withdraw rate of 5 mm/min. While a uniform, 4–5 μm thick HfO2 reaction layer formed between the uncoated substrate and the DZ125 alloy melt. Two dipping-drying cycles were required to ensure the alumina sol completely covered the surface of the substrate.