Search

In this study, we propose a lightweight surrogate random forest (L-SRF) algorithm that can be interpreted through a new rule distillation method. The common surrogate models replace the existing heavy and deep but high-performance black box model using a teacher–student learning framework. However, the student model obtained in this way must maintain the performance of the teacher model, and thus the degree of model simplification and transparency is extremely limited. Therefore, to increase model transparency while maintaining the performance of the surrogate model, we propose two methods. First, we propose a cross-entropy Shapley value to evaluate the contribution of each rule in the student surrogate model. Second, a random mini-grouping method is devised to effectively distilless important rules while minimizing the overfitting problem caused by a model simplification. The proposed L-SRF based on a rule contribution has the advantage of improving the degree of simplification and transparency of the model by realizing the large distillation ratio against the initial SRF model. In addition, because the proposed L-SRF removes unnecessary rules, it is possible to minimize the loss of the importance and relevance of each feature. To demonstrate the superior performance of the proposed L-SRF method, several comparative experiments were conducted on various data sets. We proved experimentally that the proposed method achieves a more effective performance than black box AI models in terms of model transparency and memory requirement, as well as the interpretation of the feature relevance.

Understanding uncertainty is essential in using life cycle assessment (LCA) to support decisions. Monte Carlo simulation (MCS) is widely used to characterize the variability in LCA results, be them life cycle inventory (LCI), category indicator results, normalized results, or weighted results. In this study, we present a new method to decompose MCS results into underlying contributors using the logarithmic mean Divisia index (LMDI) decomposition method with a case study on natural gas focusing on two impact categories: global warming and USETox human health impacts. First, after each run of MCS, the difference in simulated and deterministic results is decomposed using the LMDI decomposition method, which returns the contribution of each factor to the difference of the run. After repeating this for 1000 MCS runs, the statistical properties of the contributions by each factor are analyzed. The method quantifies the contribution of underlying variables, such as characterization factors and LCI items, to the overall variability of the result, such as characterized results. The method presented can decompose the variabilities in LCI, characterized, normalized, or weighted results into LCI items, characterization factors, normalization references, weighting factors, or any subset of them. As an illustrative example, a case study on natural gas LCA was conducted, and the variabilities in characterized results were decomposed into underlying LCI items and characterization factors. The results show that LCI and characterization phases contribute 65% and 35%, respectively, to the uncertainty of the characterized result for global warming. For the human health impact category, LCIs and characterization factors contribute 32% and 68%, respectively, to the overall uncertainty. In particular, methane emissions in LCI contributed the most to the overall uncertainties in global warming impact, while the characterization factor of chromium was identified as the main driver of the overall uncertainties in human health impact of natural gas. Using this approach, LCA practitioners can decompose the overall variability in the results to the underlying contributors under the MCS setting, which can help prioritize the parameters that need further refinement to reduce overall uncertainty in the results. The method reliably estimates the uncertainty contributions of the variables with large variabilities without the need for large computational resources, and it can be applied to any stage of an LCA calculation including normalization and weighting, or to other fields than LCA such as material flow analysis and risk assessment.

We investigated the relationship between grain boundary (GB) oxidation of Cu–Ag thin-film catalysts and selectivity of the (photo)electrochemical CO2 reduction reaction (CO2 RR). The change in the ...

All solid-state batteries (ASSBs) are now anticipated to be an ultimate solution to the persistent safety issues of conventional lithium-ion batteries (LIBs). Contemporary society’s expanding power demands and growing energy consumption require energy storage with greater reliability, safety and capacity, which cannot be easily achieved with current state-of-the-art liquid-electrolyte-based LIBs. In contrast, these conditions are expected to be met by implementing ASSBs with high-performance solid-state electrolytes (SSEs). In this work, we altered the microscopic structure and Li diffusional behaviors of argyrodites (Li6-xPS5-xCl1+x), which were precisely monitored with cooling protocols. It was shown that, at the cooling speed of -3 oC·h-1, as the cooling rate decreased, impurities in Li5.5PS4.5Cl1.5 such as LiCl and Li3PO4 gradually diminished and eventually disappeared. At the same time, differences in the lattice sizes of Li5.5PS4.5Cl1.5 crystallites gradually decreased, resulting in a single phase Li5.5PS4.5Cl1.5. It was also found that the Cl content of the 4d crystallographic sites increased, eventually contributing to the improvement in ionic conductivity. This work also revealed the effect of cooling rates on the crystallographic atomic arrangements, which became weaker as a decrease in x. The correlations between ionic conductivities and structural features were experimentally verified via XRD and solid-state NMR studies.

Adsorption cycles have been gaining significant interest in waste-heat recovery and renewable energy utilization. Adsorption isotherm data and the equilibrium cycle analysis are crucial steps in evaluating a typical adsorbent + adsorbate pair. In this paper, the performance of Maxsorb III + R245fa and spherical activated carbon, SAC-2 + R245fa were studied for adsorption cooling and adsorption heat transformer (AHT) cycles. Adsorption isotherms of these pairs were measured using the constant-volume-variable-pressure apparatus for temperatures ranging from 30 °C to 60 °C, and fitted with the Dubinin–Astakhov (D–A) and the Toth isotherm model. An improved equilibrium model was developed, accounting the effects of thermal masses. The specific cooling energy (SCE) and the coefficient of performance (COP) of the adsorption cooling cycle were evaluated for various thermal mass to adsorbent mass ratios. It is observed that SAC-2 + R245fa pair offers better SCEs (20 kJ kg−1and 160 kJ kg−1 at 60 °C and 90 °C, respectively) when compared to that of Maxsorb III + R245fa. The impact of thermal mass is found to be significant for all regeneration temperatures for Maxsorb III + R245fa while the deterioration of COP in SAC-2 + R245fa is notable for high regeneration temperatures (> 75 °C). When employed in the AHT cycle, Maxsorb III offers a slightly higher useful heat while SAC-2 provides a better Q u h / Q Q R albeit by a small margin. The Q u h / Q Q R values for both studied pairs are more than 0.6 for all regeneration temperatures for the heat extraction at 120 °C.

In tactical environments, mobility has a significant impact on the performance such as handover delay. In this paper, we design and implement an integrated tactical mobility testbed that consists of (1) host identify protocol module for secure initial registration, (2) Proxy Mobile IPv6-based distributed mobility management module for infrastructure-based mobility cases, and (3) delay/disruption tolerant networking module for opportunistic mobility cases. The implementation and experimental results demonstrate that the integrated testbed can be used to consider diverse tactical mobility scenarios and to evaluate the handover performance.

We propose a preprocessing system that transforms the real world into 3D mesh virtual world to provide augmented reality (AR) services. The proposed system uses a monocular RGB camera to obtain sequential images to generate real space. First, we create 3D real-world space composed of point clouds using sequential color images. And then, we segment the objects in each color image and assign an identification number for each object. Also, we execute ‘3D point labeling’, which assigns identification numbers from 2D segmented object to 3D point cloud through a simple projection manner with several parameters. This process could collect 3D point cloud with same the label by as an object. Finally, only the 3D point clouds that have the same identification number by as an object are used to generate 3D mesh. This paper confirmed that 3D mesh could be created using the only general monocular camera without the help of special 3D sensors.

Si nanosheets (SiNS) as two dimensional Si nanomaterials are promising candidates for anodes of lithium ion batteries (LIBs) by their large surface area and excellent mechanical durability. In this study, we fabricated polymeric binder (carboxymethyl cellulose, CMC) coated SiNS electrodes using a spin-coating process and characterized the electrochemical properties. The CMS coated SiNS electrodes were homogeneously covered with a CMC polymer layer and showed the improved capacity retention with maintaining over 90% of initial capacity after several cycles. The result of electrochemical impedance spectroscopy indicates that the CMC layer provide the chemical stability of Si surface and suppress the continuous growth of the SEI layer. Furthermore, CMC coated layer on SiNS provide improved mechanical stability that maintains the adhesion of SiNS to the current collector during the discharge–charge cycles.

An autoencoder that automatically generates an initial guess for the minimum energy pathway (MEP) calculations has been designed. Specifically, our autoencoder takes in the trajectories of molecular dynamics simulations as its input and facilitates the generation of feasible molecular coordinates. Two molecules (acetonitrile and alanine dipeptide) were tested using the nudged elastic band calculations and the results provided improvements over linear interpolation and image dependent pair potential methods in terms of the number of SCF iterations, demonstrating the utility of using an autoencoder type of an approach for MEP calculations.

Domestic heat pumps constitute a significant part of the global heat pump industry. R401A offers excellent performance with no influence on the ozone depletion and has been a dominant refrigerant in most domestic heat pumps. However, R410A has a significant impact on the climate due to its high global warming potential (GWP). Thus, the heat pump industry has been focusing on the development of R410A substitutes. Pure refrigerant alternates, such as HFC32, and the blends mixed with low-GWP refrigerants, such as HFO1234yf and HFO1234ze(E), are widely reported. Besides the direct impact of refrigerants (which is estimated as GWP), domestic heat pumps indirectly affect the environment from the energy consumption and manufacturing processes. Hence the life cycle climate performance (LCCP) analysis accounts for all emissions through the lifetime of a heat pump. This article reports the LCCP evaluation of various low-GWP refrigerants for R410A replacement on domestic heat pumps. Six refrigerants, i.e., HFC32, binary blends of HFC32 and HFO1234yf (with GWP values of 300, 200 and 150), and HFC32 and HFO1234ze(E) with GWP values of 300 and 200, were compared against R410A. The performance data of these refrigerants from the experimental heat pump facility were utilized to evaluate the LCCP. Among the selected refrigerants, the binary blend of HFC32/HFO1234ze(E) with GWP 300 shows the lowest LCCP. Low-GWP refrigerants would become more competitive than R410A when CO2 emission from energy generations can be reduced by the optimization of the system or the usage of renewable energy.

This paper proposes a keypoint regressor (KeyReg), which consists of multi-layer random forest (MRF) regressor and single random forest (SRF) classifier modules. To increase the keypoints’ repeatability, the MRF regressor is applied to multi-scale images in a shared rules manner, and keypoints predicted at each scale are given a confidence score through the SRF for reliability measurement. Each candidate point is detected as the final keypoint through a non-maxima suppression process based on a confidence score. The MRF structure of KeyReg is designed to maintain a coarse-to-fine structure by varying the number of nodes per layer. In addition, the accuracy of the matching can be improved by removing less confidential keypoints through the continuous SRF classifier. KeyReg is the first approach to apply an MRF to the keypoint regression and is designed to run on a CPU rather than a GPU compared to DNN-based approaches. KeyReg training was conducted using COCO, and positive and negative examples were automatically obtained under a self-supervised learning method between the original image and a warped image. The proposed KeyReg showed superior performance in terms of repeatability, the accuracy of the homography, mean matching accuracy (MMA), and localization errors on HPatches dataset compared to state-of-the-art methods.

It has been believed that stereo-irregular polymers hardly form crystalline structure because of the deficient ordering along the backbone direction. Here we show the crystallization of atactic polymers by rapid thermal quenching of polymers co-melted with a chain-stretching agent. To this end, crystalline PMMAs (hc-PMMAs) are prepared from stereo-irregular atactic-PMMAs (Mn = 35–1500 kg/mol) as well as stereo-regular isotactic- and syndiotactic-PMMAs by rapid quenching of molten PMMA/benzoic acid (BA) solutions. BA acts as a chain-stretching agent. PMMA chains are elongated up to 60% of their contour length when they are co-melted with BA. The preferential hydrogen bonding between PMMA and BA compensates the conformational entropy loss of PMMA chains in molten PMMA/BA solutions. Upon quenching, PMMA chains are further stretched because of the large tensile stress induced by the sudden temperature drop and because of mechanical squeezing by concomitant directional crystallization of anisotropic lath-like BA crystals. After sublimation of BA, the PMMA shows a clear melting transition: Tm = 276–283 °C for hc-s-PMMA and 293–306 °C for hc-a-PMMA. Unlike other conventional crystals having 3D crystallinity, hc-PMMAs exhibits 1D crystallinity due to ordered lateral packing between chains (d = ∼4.3 A) which is independent on the tacticity while the packing is amorphous-like for the other two directions. For the reason of the reduced dimensionality in crystallinity, we named the crystalline PMMA as 1D hypo-crystalline PMMA (hc-PMMA). Because of their high 1D crystallinity (55–65%), hc-PMMAs exhibit the storage modulus (E′) enhanced by 20 times comparing with those of pristine PMMA.

Titanium dioxide (TiO2) is a semiconductor photocatalyst widely applied in numerous fields due to possessing prominent photocatalytic properties. However, its practical applications in the form of nanoparticles or powders still have remained several limitations. Recently, novel photocatalytic porous composites have been discovered to be potential alternative approaches. In the present study, nanostructured magnesium-aminoclay-based TiO2 (MgAC–TiO2) was successfully deposited on an activated carbon fiber (ACF) matrix using the sol–gel approach followed by calcination at 350°C in an air atmosphere. The structure and photocatalytic activity of this as-prepared photocatalyst composite were characterized by scanning electron microscopy (SEM), X-ray photoelectron spectroscopy (XPS), the Brunauer-Emmett-Teller (BET), and UV-vis diffuse reflectance spectral analysis. The photocatalytic activity of MgAC–TiO2/ACF was investigated under batch conditions for the removal of methylene blue (MB) in solution under UV irradiation and dark conditions. The results revealed that MB is absorbed by MgAC–TiO2/ACF and that its photodecomposition occurs under UV irradiation. The addition of MgAC can prevent the sintering of TiO2 act as a dispersing agent to create a high specific surface area, and thus enhance photocatalytic efficiency. In addition, ACF in the MgAC–TiO2/ACF composite can additionally improve the photocatalytic activity by hindering electron-hole recombination, which is known as a synergetic effect, and thereby enhancing the photodegradation and removal efficiency of MB.

Location-aware technologies and big data are transforming the ways we capture and analyze human activities. This has particularly affected tourism geography, which aims to study tourist activities ...

The emission properties of the rare earth (RE) elements have been found to be sensitive to the local lattice environment around them, and could be employed for probing the local lattice environment. Because each constituent ion of a material should have its own particular structural environment, the emission profiles in RE elements depend strongly on the doping sites of RE elements inside the host crystals. We investigated the emission properties of the Eu3+ ion doped BaZrO3 (BaZrO3:Eu) with doping site dependence (A-site vs. B-site), as well as ambient dependence in the post-annealing process. The site-selective doping was identified from the Rietveld refinement analysis on the XRD patterns of our BaZrO3:Eu samples. Photoluminescence and photoluminescence excitation measurements showed that the emission properties of the samples with the Eu3+ ions substituted at Zr sites revealed much greater emission properties than those at the Ba sites. This behavior was found to accord with the change in intensity ratio of 5D0 → 7F0,2 to 5D0 → 7F1 in Eu ions, which should be a measure of the local lattice asymmetry around the Eu ions. We also found that with post-annealing in H2 atmosphere, the emission intensities of the Eu ions increased significantly, while with post-annealing in O2 atmosphere, they were suppressed. Our findings indicated that the cation/oxygen vacancies could change the local lattice environment around the Eu ions, as well as the valence states of Eu ions, depending on the doping site in the cubic perovskite.

Understanding ferroelectric domain switching dynamics at the nanoscale is a great of importance in the viewpoints of fundamental physics and technological applications. Here, we investigated the intriguing polarity-dependent switching dynamics of ferroelectric domains in epitaxial BiFeO3 (001) capacitors using transient switching current measurement and piezoresponse force microscopy. We observed the distinct behavior of nucleation and domain wall motion depending on the polarity of external electric bias. When applying the negative bias to the top electrode, the sideways domain wall motion initiated by only few nuclei was dominant to polarization switching. However, when applying the positive bias, most of domains started to grow from the pre-existed pinned domains and their growth velocity was much smaller. We suggest that the observed two distinct domain switching behavior is ascribed to the interfacial defect layer.

Summary As the share of variable renewable energy (VRE) grows in the electric grid, so does the risk of curtailment. While energy storage and hydrogen production have been proposed as solutions to the curtailment problem, they often pose technological and economic challenges. Here, we analyze the potential of data center load migration for mitigating curtailment and greenhouse gas (GHG) emissions. Using historical hourly electricity generation, curtailment, and typical data center server utilization data, we simulate the effect of migrating data center workloads from the fossil-fuel-heavy PJM to the renewable-heavy CAISO. The results show that load migration within the existing data center capacity during the curtailment hours in CAISO has the potential to reduce 113–239 KtCO2e per year of GHG emissions and absorb up to 62% of the total curtailment with negative abatement costs in 2019. Our study demonstrates the overlooked role that data centers can play for VRE integration and GHG emissions mitigation.

Halide vacancy defect is one of the major origins of non-radiative recombination in the lead halide perovskite light emitting devices (LEDs). Hence the defect passivation is highly demanded for the high-performance perovskite LEDs. Here, we demonstrated that FA doping led to the enrichment of Br in Cs1−xFAxPbBr3 QDs. Due to the defect passivation by the enriched Br, the trap density in Cs1−xFAxPbBr3 significantly decreased after FA doping, and which improved the optical properties of Cs1−xFAxPbBr3 QDs and their QD-LEDs. PLQY of Cs1–xFAxPbBr3 QDs increased from 76.8% (x = 0) to 85.1% (x = 0.04), and Lmax and CEmax of Cs1–xFAxPbBr3 QD-LEDs were improved from Lmax = 2880 cd m−2 and CEmax = 1.98 cd A−1 (x = 0) to Lmax = 5200 cd m−2 and CEmax = 3.87 cd A−1 (x = 0.04). Cs1–xFAxPbBr3 QD-LED device structure was optimized by using PVK as a HTL and ZnO modified with b-PEI as an ETL. The energy band diagram of Cs1–xFAxPbBr3 QD-LEDs deduced by UPS analyses.

A facile method for preparing nitrogen-doped graphite felt electrodes with high electrocatalytic activity for vanadium redox flow batteries (VRFBs) is developed. These nitrogen-doped graphite felts are fabricated by coating 1-ethyl-3-methylimidazolium dicyanamide (EMIM dca) on the surface of graphite felts followed by thermal treatment under a N2 atmosphere. EMIM dca, an ionic liquid containing a high content of nitrogen, is used as an effective precursor for nitrogen doping on graphite felt surfaces. The physical and electrochemical properties of the nitrogen doped graphite felts are characterized by scanning electron microscopy (SEM), X-ray photoelectron spectroscopy (XPS) and cyclic voltammetry (CV). When employed in charge-discharge tests, the single cells with the nitrogen doped graphite felts show outstanding performance. The graphite felt coated with 20 wt% EMM dca solution (GF-Ed20) exhibits three times increased discharge capacity and 10% higher energy efficiency at a current density of 150 mA cm−2 compared to the graphite felt with the oxygen functional groups (GF-Ref) prepared by the conventional heat treatment method. The improved performance is attributed to the high nitrogen content on the graphite felt, which increases the electrocatalytic activity of vanadium redox reactions.

Serial femtosecond crystallography (SFX) enables the determination of the room-temperature crystal structure of macromolecules, as well as providing time-resolved molecular dynamics data in pump–probe experiments. Fixed-target SFX (FT-SFX) can minimize sample consumption and physical effects on crystals during sample delivery. In FT-SFX studies, having a sample holder that can stably fix crystal samples is one of the key elements required for efficient data collection. Hence, development of sample holders from new materials capable of supporting various crystal sizes and shapes may expand the applications of FT-SFX. Here, a viscous-media-based crystal support in a sample holder for FT-SFX is introduced. Crystal samples were embedded in viscous media, namely gelatin and agarose, which were enclosed in a polyimide film. In the vertically placed sample holder, 10–15%(w/v) viscous gelatin and 1–4%(w/v) agarose gel stably supported crystals between two polyimide films, thereby preventing the crystals from descending owing to gravity. Using this method, FT-SFX experiments were performed with glucose isomerase and lysozyme embedded in gelatin and agarose, respectively. The room-temperature crystal structures of glucose isomerase and lysozyme were successfully determined at 1.75 and 1.80 Å resolutions, respectively. The glucose isomerase and lysozyme diffraction analyses were not impeded by excessive background scattering from the viscous media. This method is useful for delivering crystal samples of various sizes and shapes in FT-SFX experiments.

Ammonia (NH3) is pivotal to the fertilizer industry and one of the most commonly produced chemicals1. The direct use of atmospheric nitrogen (N2) had been challenging, owing to its large bond energy (945 kilojoules per mole)2,3, until the development of the Haber–Bosch process. Subsequently, many strategies have been explored to reduce the activation barrier of the N≡N bond and make the process more efficient. These include using alkali and alkaline earth metal oxides as promoters to boost the performance of traditional iron- and ruthenium-based catalysts4–6 via electron transfer from the promoters to the antibonding bonds of N2 through transition metals7,8. An electride support further lowers the activation barrier because its low work function and high electron density enhance electron transfer to transition metals9,10. This strategy has facilitated ammonia synthesis from N2 dissociation11 and enabled catalytic operation under mild conditions; however, it requires the use of ruthenium, which is expensive. Alternatively, it has been shown that nitrides containing surface nitrogen vacancies can activate N2 (refs. 12–15). Here we report that nickel-loaded lanthanum nitride (LaN) enables stable and highly efficient ammonia synthesis, owing to a dual-site mechanism that avoids commonly encountered scaling relations. Kinetic and isotope-labelling experiments, as well as density functional theory calculations, confirm that nitrogen vacancies are generated on LaN with low formation energy, and efficiently bind and activate N2. In addition, the nickel metal loaded onto the nitride dissociates H2. The use of distinct sites for activating the two reactants, and the synergy between them, results in the nickel-loaded LaN catalyst exhibiting an activity that far exceeds that of more conventional cobalt- and nickel-based catalysts, and that is comparable to that of ruthenium-based catalysts. Our results illustrate the potential of using vacancy sites in reaction cycles, and introduce a design concept for catalysts for ammonia synthesis, using naturally abundant elements. Ammonia is synthesized using a dual-site approach, whereby nitrogen vacancies on LaN activate N2, which then reacts with hydrogen atoms produced over the Ni metal to give ammonia.

High-mobility low-hysteresis electrolyte-gated thin-film transistors were successfully fabricated using a diketopyrrolopyrrole (DPP) and benzotriazole (BTz) copolymer semiconductor and high-capacitance ion gels based on 1-ethyl-3-methylimidazolium bis[(trifluoromethyl)sulfonyl]imide ([EMI][TFSI]) and tetraoctyl phosphonium ([P8888]) [TFSI]. The DPP-BTz transistors gated with both ion gels operated effectively at low voltages below 1V with high on-to-off current ratios exceeding 105 and very low device hysteresis. Specifically, when the [EMI][TFSI] ion gel was employed as a gate dielectric layer, the DPP-BTz transistors exhibited a very high carrier-mobility value of 8.50 ± 1.09 cm2/Vs. In addition, device hysteresis of the DPP-BTz transistor was almost invariant to the voltage sweeping rate and was much lower than that of the poly(3-hexylthiophene) (P3HT) transistor under the same conditions. Overall, these results indicate that the performance of electrolyte-gated transistors can be improved using the DPP-BTz semiconductor.

Global low‐carbon transition demands the development of large‐scale infrastructure, which is sand intensive. Natural sand is widely considered abundant, whereas recent research has pointed out the increasing risk of supply shortage in a number of world regions. In the current research, we examine the implication of future sand demand in the context of low‐carbon infrastructure development. We mapped the projected investments on buildings and low‐carbon infrastructure up to 2030 and estimated the sand intensity of the two types of construction. We translated these investments and sand intensity to country‐specific sand‐demand projections under three economic development scenarios. Our results indicate that China is expected to face the highest sand demand, followed by India, and that should the current sand extraction rates and construction practices be maintained, developing countries will be exposed to a significant supply risk of construction sand. Under the scenario of sustainable growth, developed economies are expected to have a relatively stable sand demand whereas South‐East Asia and Africa will see a rapid increase in their sand demand over the next 15 years. Our results call regional sand supply security into attention in low‐carbon transition planning.

Wireless Body Area Networks (WBAN) is becoming highly distinctive within health application. Security is important for individual nodes because such applications deal with sensitive data. Within WBAN, approaches with higher security with less computation are preferred along with smaller key size because of limitations of WBAN. However reducing key size can result in reduction in security level. In this paper we propose the use of message authentication for WBAN with pre-processed symmetric RSA (PSRSA). PSRSA is based on RSA with smaller key with multiple layers of pre-processing to increase the total combinations. Based on our analysis, PSRSA provides high security level with smaller key size compare to RSA. We implemented PSRSA to check feasibility and tested couple of scenarios.

In this Letter, two comparator-based voltage-mode hysteretic controller designs are proposed to enhance dynamic response for single-inductor multiple-output (SIMO) switching converters to minimise over-/under-shoot voltages due to self- and cross-regulations in fast load transients, especially with high output power and large current steps. The proposed converters are designed and simulated in a 65 nm CMOS process with standard I/O devices. With the proposed techniques, the output voltage fluctuations, including steady-state voltage ripples and over-/under-shoot voltages due to self- and cross-regulations during a 1.8 A load transient in 1 ns, are kept within 60 mV. The proposed converter delivers more than 3 A maximum load current per output and a maximum total output power above 6 W, with a peak efficiency above 90%. Compared to state-of-the-art SIMO converters, this work achieves a significantly higher output power capacity and faster dynamic response thus smaller self- and cross-regulations within 0.033 mV/mA.

As the number of network devices is increasing and they are highly connected, network attacks have become more complex and varied. To mitigate these attacks, multiple types of network security equipment are used in combination, requiring considerable security knowledge of each type of network security equipment. Also, the deployment of network security equipment requires considerable cost and time. To solve these problems, this article proposes an IBCS based on automation and virtualization. IBCS provides not only an automated, virtualized security system for security service providers, but also easy, quick, and flexible cloud-based security services for security service consumers based on their intents. Finally, this article demonstrates the feasibility of IBCS with a prototype in a real network environments.

We report a facile method for the nitrogen-doping process using the same precursor material for graphite felt electrodes, polyacrylonitrile (PAN). To utilize the nitrogen content in PAN, two steps ...

A new space vector pulse-width modulation (SVPWM) technique to eliminate common-mode voltage (CMV) is proposed for a dual three-phase inverter fed-dual-winding fault-tolerant permanent magnet motor (FTPMM) drive. In general, since both the PWM module-1 of an inverter-1 drive and the PWM module-2 of an inverter-2 drive adopt the conventional SVPWM (CSVPWM) technique, two identical CMVs are generated in the drives. Each of the CMCs generated by the CMVs flows through the stray capacitances between the stator windings and the motor frame. These currents flow into the ground of the drive, where they are combined. Hence, in the proposed technique, each of the PWM modules adopts the remote-state PWM (RSPWM)2A and RSPWM2B. As a result, two CMVs with same absolute values and opposite polarities are generated at every switching period, and the corresponding generalized total CMV of the dual-winding FTPMM drive always becomes zero. Analytical and experimental results are provided to verify the validity of the proposed technique.

A new one-dimensional fixed-target sample delivery method combined with a real-time visual servo scan system has been developed for serial femtosecond crystallography. The developed high-throughput microcrystal delivery system shows significant improvements for microcrystal delivery efficiency at an optimal crystal stability and significantly reduced sample consumption to reveal protein structures utilizing X-ray free-electron laser pulses., Microcrystal delivery methods are pivotal in the use of serial femtosecond crystallography (SFX) to resolve the macromolecular structures of proteins. Here, the development of a novel technique and instruments for efficiently delivering microcrystals for SFX are presented. The new method, which relies on a one-dimensional fixed-target system that includes a microcrystal container, consumes an extremely low amount of sample compared with conventional two-dimensional fixed-target techniques at ambient temperature. This novel system can deliver soluble microcrystals without highly viscous carrier media and, moreover, can be used as a microcrystal growth device for SFX. Diffraction data collection utilizing this advanced technique along with a real-time visual servo scan system has been successfully demonstrated for the structure determination of proteinase K microcrystals at 1.85 Å resolution.

Understanding the environmental consequences of actions is becoming increasingly important in the field of industrial ecology in general, and in life cycle assessment (LCA) more specifically. However, a consensus on how to operationalize this idea has not been reached. A variety of methods have been proposed and applied to case studies that cover various aspects of consequential life cycle assessment (CLCA). Previous reviews of the topic have focused on the broad agenda of CLCA and how different modeling frameworks fit into its goals. However, explicit examination of the spectrum of methods and their application to the different facets of CLCA are lacking. Here, we provide a detailed review of methods that have been used to construct models of the environmental consequences of actions in CLCA. First, we cover the following structural modeling approaches: (a) economic equilibrium models, (b) system dynamics models, (c) technology choice models, and (d) agent‐based models. We provide a detailed review of particular applications of each model in the CLCA domain. The advantages and disadvantages of each are discussed, and their relationships with CLCA are clarified. From this, we are able to map these models onto the established aspects of CLCA. We learn that structural models alone are not sufficient to quantify the uncertainty distributions of underlying parameters in CLCA, which are essential components of a robust analysis of consequences. To address this, we provide a brief introduction to a counterfactual‐based causal inference approach to parameter identification and uncertainty analysis that is emerging in the CLCA literature. We recommend that one potential research path forward is the establishment of feedback loops between empirical estimates and structural models.

Randomized sampling-based ensemble learning is emerging as a new alternative to deep neural networks (DNNs) because it supports diversity and locality and does not require backpropagation in the learning process. By connecting randomized weak classifiers in a layer-by-layer manner, it is possible to create models having a performance similar to that of DNNs, but with less overfitting and better generalizability than before. Therefore, in this paper, we propose a deep random ferns (d-RFs) model, in which extremely randomized ferns are connected to multilayers, allowing a high classification performance and a lightweight and fast structure. The input vector is first encoded as a transformed feature vector in the feature encoder layer and then is input to the cascade layers. The feature encoding process is similar to the DNN convolution and helps improve the performance of the final output layer. Unlike in the backpropagation paradigm, the cascade layer adjusts the number of ferns and layers required for the d-RFs adaptively, using only a small scale of data. RFs ensemble approaches have considerably fewer hyper-parameters than a DNN or deep forest model, and the complexity can be determined automatically in a data-dependent manner. In addition, experimental results show that an ensemble of multiple weak classifiers reduces the bias between models through an averaging of the weakly correlated classifiers, resulting in a better overall model. The proposed lightweight d-RFs was successfully applied to benchmark datasets and yielded a similar or better accuracy level and a smaller number of parameters and operations as compared with state-of-the-art methods.

In a multiple object tracking (MOT) system, an association check between the tracker and detected objects is an important factor in determining the tracking performance. Siamese convolution neural network (CNN) is the most popular data association method in MOT owing to its good matching performance and network sharing support. However, it is unsuitable for real-time online tracking in low-end systems because numerous parameters and operations are still required. In this article, instead of a CNN, we propose using a SiameseRF algorithm which combines Siamese structure and random forest (RF), enabling high-speed learning and classification. SiameseRF has a shared-rule based Siamese structure rather than shared weight, which improves the matching performance and solves existing slow CNN-based tracking issues. During the learning process, the shared RFs consisting of tree rules are learned in the directions of increasing similarity to the positive pair {anchor, positive} and increasing difference between the negative pair {anchor, negative}. However, because many rules that make up SiameseRF remain a burden for online processing, this study proposes an additional rule distillation algorithm to effectively remove redundant and unimportant rules causing an overfitting with SiameseRF. This reduction in the number of rules reduces the processing time and number of parameters in the rule distilled SiameseRF. In experiments conducted on MOT benchmark datasets, our proposed 30% rule distilled SiameseRF achieved up to a 1.12-times faster speed and a 1.13-times higher compression rate than basic SiameseRF while maintaining a similar or somewhat better tracking performance than other state-of-the-art CNN-based MOT algorithms.

In this paper, we propose a new approach to detect Deepfakes generated through the generative adversarial network (GANs) model via an algorithm called DeepVision to analyze a significant change in the pattern of blinking, which is a voluntary and spontaneous action that does not require conscious effort. Human eye blinking pattern has been known to significantly change according to the person's overall physical conditions, cognitive activities, biological factors, and information processing level. For example, an individual's gender or age, the time of day, or the person's emotional state or degree of alertness can all influence the pattern. As a result, Deepfakes can be determined through integrity verification by tracking significant changes in the eye blinking patterns in deepfakes by means of a heuristic method based on the results of medicine, biology, and brain engineering research, as well as machine learning and various algorithms based on engineering and statistical knowledge. This means we can perform integrity verification through tracking significant changes in the eye blinking pattern of a subject in a video. The proposed method called DeepVision is implemented as a measure to verify an anomaly based on the period, repeated number, and elapsed eye blink time when eye blinks were continuously repeated within a very short period of time. DeepVision accurately detected Deepfakes in seven out of eight types of videos (87.5% accuracy rate), suggesting we can overcome the limitations of integrity verification algorithms performed only on the basis of pixels.

In general, transition metal-doped La0.6Sr0.4FeO3 (LSF) has been used as a cathode material for intermediate temperature solid oxide fuel cells (IT-SOFCs) because of its high mixed electronic−ionic conductivity and catalytic properties. Recently, some research groups have been investigating the doped LSF as an anode material. In this study, we evaluated the influence of dopant in LSF on anodic properties of LSF in SOFCs. Whereas Mn-doped LSF showed typical perovskite oxide structure even after reduction in hydrogen at high temperature, the LSF and Co-doped LSF exhibited phase transition partially to LaSrFeO4 and exsolution of metal particles after reduction. The phase transition and metal exsolution occurred at temperature higher than 1008 K in a reducing atmosphere. Despite the partial phase transition, the cell using Co-doped LSF anode exhibited fairly high power density of 1.33 W/cm2 at 1173 K with the lowest polarization resistance. These results may originate from the high oxygen-ion conductivity of LaSrFeO4–La(Sr)Fe(Co)O3 and the high hydrogen oxidation property of the Co–Fe particles on ceramic anode surface.

Stabilization performance of an air-borne EO/IR camera is severely affected mainly by two factors: One is the vibration transmitted from the platform, and the other is the torque disturbance due to mass imbalance of the inner gimbal installed on the camera. Elastomeric isolators are commonly employed in aeronautical equipment due to their ease of installation, light-weight, and excellent dissipation characteristics. The dynamic characteristics of an elastomeric isolator are a function of frequency and also of temperature. Therefore, it is very important to obtain accurate parameters for its dynamic characteristics for the control of an EO/IR camera system. The parameters are identified inversely by testing and by using finite element calculation with proper modeling of the isolators. In this paper, the process of identifying the dynamic parameters of the isolators using a simple pilot test and a finite element model of the camera system is presented. A pilot test equipment is built to put real environmental condition (e.g., temperature condition, heat flow condition, etc.) on the elastomeric isolator. Therefore, the characteristics of an elastomer identified on tests are the dynamic values reflecting the temperature environment characteristic. In the process, 3 dB method is introduced for fast and effective parameter identification. On matching the isolator model on FEA to the disturbance transfer model on stabilization control simulator, it is able to narrow the difference between the test and the real condition by using more reliable parameters.

This study investigated the influence of the initial grain size on the plastic deformation and tunnel defects that occurred from friction stir welding of CoCrFeMnNi high entropy alloys (HEAs). The rolled and cast HEAs had a grain size of 2.8 and 308 μm, respectively. After friction stir welding, the cast HEA weld had a grain size of 1.8 μm, which was coarser than that of the rolled HEA weld (1.4 μm). Therefore, the dynamic recrystallization ratios were 60.7 and 99.6% for the rolled and cast HEAs, respectively. The cast HEA weld with a large grain size contained a higher density of high-angle boundaries and twins than the rolled HEA weld with the small grain size. The cast HEA had a larger resistance to plastic deformation owing to the larger fraction of ∑3 twin boundaries than the rolled HEA during friction stirring. This was associated with the high strain hardening rate during tensile testing and to the significant amount of W dissolved from the stirring tool in the cast HEA weld. Thus, the cast HEA weld had a higher tunnel defects ratio than the rolled HEA weld. The total unbonded ratios of the rolled and cast HEA welds were 0.2 and 7.2%, respectively, indicating that the rolled HEA had better weldability than the cast HEA.

Heavy rainfall events account for most socioeconomic damages caused by natural disasters in South Korea. However, the microphysical understanding of heavy rain is still lacking, leading to uncertainties in quantitative rainfall prediction. This study is aimed at evaluating rainfall forecasts in the Local Data Assimilation and Prediction System (LDAPS), a high-resolution configuration of the Unified Model over the Korean Peninsula. The rainfall of LDAPS forecasts was evaluated with observations based on two types of heavy rain events classified from K-means clustering for the relationship between surface rainfall intensity and cloud-top height. LDAPS forecasts were characterized by more heavy rain cases with high cloud-top heights (cold-type heavy rain) in contrast to observations showing frequent moderate-intensity rain systems with relatively lower cloud-top heights (warm-type heavy rain) over South Korea. The observed cold-type and warm-type events accounted for 32.7% and 67.3% of total rainfall, whereas LDAPS forecasts accounted for 65.3% and 34.7%, respectively. This indicates severe overestimation and underestimation of total rainfall for the cold-type and warm-type forecast events, respectively. The overestimation of cold-type heavy rainfall was mainly due to its frequent occurrence, whereas the underestimation of warm-type heavy rainfall was affected by both its low occurrence and weak intensity. The rainfall forecast skill for the warm-type events was much lower than for the cold-type events, due to the lower rainfall intensity and smaller rain area of the warm-type. Therefore, cloud parameterizations for warm-type heavy rain should be improved to enhance rainfall forecasts over the Korean Peninsula.

This study investigates the effects of welding velocity on weldability of dissimilar cast/rolled high-entropy alloys and the applicability of cryogenic temperatures. Cast HEA side of dissimilar wel...

Recent increases in Distributed Generation (DG) into distribution system have led to changes in the conventional protection scheme because of the reverse fault current coming from the DG side. However, it is not yet clear whether the reverse fault current is caused by the type of interconnection transformer or the type of DG. Therefore, for reliable coordination between protective devices, it is necessary to analyze the reverse fault current contribution from the DG side to fault location according to the system configuration. In this paper, the single-line to ground fault current contribution is analyzed when different combinations of the transformer and DG are interconnected in the system. For this, a methodology for calculating the fault current according to the type of transformer and DG is introduced. Using this methodology, the fault current contribution from each transformer and DG is analyzed separately to eliminate the mixed contribution from both the transformer and DG. The mixed fault current contribution from the different combinations of these devices is then analyzed. For the interconnection transformer, delta-grounded wye (D-Yg), Yg-Yg, and Yg-D transformer configurations are considered; for DG, synchronous machine-based and inverter-based DG are considered. Based on this analysis, the present paper provides effective guidelines for determining the reverse fault current according to the type of transformer and DG in the distribution system, enabling reliable protection coordination under the integration of DG.

We investigated the emission property of Ce3+ and Tb3+ co-doped CaHfO3 phosphor. Using photoluminescence and photoluminescence excitation spectroscopy, we found that as a result of the energy transfer from Ce3+ to Tb3+, the visible emission intensity of Tb3+ increased dramatically with near ultraviolet (UV) light excitation, and consequently, the colors of our phosphors were tuned from blue to green/yellow and through to white. The temperature dependent emission measurement revealed that the energy transfer from Ce3+ to Tb3+ in the near UV region could be more efficient at low temperatures.

1-(6-bromohexyl)-3-methyl-2-(2,4,6-trimethylphenyl)-2H-benzimidazolium bromide was synthesised and grafted onto polybenzimidazole (PBI). The highest obtained degrees of functionalisation were 0.7 and 1.1 ionic groups per repeat unit, corresponding to ion exchange capacities of 1.25 and 1.55 mmol chloride/g. While the latter material had much reduced tensile strength and Young's modulus, the material with a lower degree of functionalisation showed properties similar to PBI and was therefore further investigated. X-ray scattering showed that the functional group reduced the crystallinity. Conductivity was in the range of 2–3 mS cm−1, but increased to 23 mS cm−1 after doping in 2 M sulfuric acid, 2–4 times higher than for doped PBI. Permeability for VO2+ ions was 0.8 × 10−12 m2 s−1. In the vanadium redox flow battery, the grafted PBI showed higher coulomb and voltage efficiency than PBI, and therefore also higher energy efficiency.

PBI membranes are recently discussed as stable, well performing membranes for vanadium redox flow batteries (VRFB). Blending meta-PBI with an anion exchange polymer (FAA3i) slightly reduces the coulomb efficiency from 99.7 to 97.8%, but strongly increases the voltage efficiency from 82.5 to 88.2%, leading to an increased energy efficiency (86.2% at 80 mA cm−2), exceeding that of meta-PBI (82.2%) and N212 (83%). Apparently, since the conductivity of sulfuric acid has a maximum around a concentration of 3.8 M, the concentration of the absorbed acid has a dominant influence on the conductivity. Addition of FAA3i decreases the concentration of the acid absorbed by PBI membranes. Furthermore, an ex-situ stability test in 1.5 M V5+ solutions in 2 M sulfuric acid for 87 days showed a very high stability for meta-PBI and Nafion 212, while the commercial FAA3 membrane disintegrated into pieces. Blending of meta-PBI and FAA3 decreased the stability, as proven by formation of V4+, but all tested blend membranes retained their membrane shape and could still be handled. Blending with FAA3 reduces the tensile strength and Young's modulus of meta-PBI, and doping with sulfuric acid leads to a further decrease in the mechanical strength. However, an acid doped PF-21 still showed a tensile strength of 37 MPa and a Young's modulus of 0.7 GPa.

In this paper, a model for a prediction of conducted emission from vehicular PMSM-drive braking system is presented. The proposed model is based on the high-frequency circuit model combined with EM simulation. The parameter for high-frequency circuit modeling is derived from the measurement as well as 3D EM simulation based on FEM. To establish meaningful high-frequency parameters, the lumped parameters concerning the three-phase motor coil, motor busbar, heatsink, LISN, harness wire, meshed strap wire, and the passive-EMI filters including parasitics are either electromagnetically simulated or directly measured through an impedance analyzer. The distributed parameters of the PCB’s, connector, and motor housing are calculated via the 3D EM simulation. Eventually, the presented high-frequency parameters were integrated with PWM inverter circuit for the system-level analysis, and the model was verified through the correlation between the measurement and simulation with the EMI filters installed in the system. Moreover, the EMI characteristics with/without particular noise propagation path were analyzed in the model. The calculated conducted emission of the proposed model showed a very good agreement with the experimental data, which proves the validness of the proposed model.

Experimental and numerical studies have been carried out to improve the flow distribution by optimizing the local porosity of the electrodes of the all-vanadium redox flow batteries (VFBs) to increase the energy efficiency at high current density. We control the local porosity by inserting extra layer of electrode at inlet and outlet, and the flow field of electrolyte is analyzed numerically. First, the flow field of electrolyte is analyzed numerically to understand distribution of electrolyte in the electrode. Then, the charge and discharge curve is analyzed to understand the effect of local porosity of electrode on the energy efficiency. At 50 mA/cm2, the energy efficiency is the highest when using electrode with uniform porosity. At 150 mA/cm2, however, the energy efficiency of the cell using the electrode with low porosity at inlet is similar to that using the uniform electrode which is 66.6%. Lastly, we suggest an empirical equation for optimal local porosity distribution of the electrode according to current density. Using the empirical equation, we can increase the energy efficiency of the cell to 67.7% at 150 mA/cm2. This study shows the possibility of increase of the energy efficiency of VFBs by controlling local porosity of the electrode.

The semiconductor etching process, which is the most important part of the semiconductor manufacturing process, requires higher sophistication as 10 nm semiconductors are mass produced. Etching methods utilizing plasma are getting increasingly popular with the miniaturization of the etching process. As the process performance depends on the state of the plasma variables, such as electron density, it is essential to measure and control these variables in real time. Moreover, to control the plasma-based system, the sensitive and time-varying characteristics of plasma should be considered. This paper proposes a self-tuning adaptive model predictive controller that can effectively perform electron density control. As a first step, an integral squared error-based Bayesian optimization is used to tune the model predictive controller, and its performance is verified on a drift-free Ar plasma system. The self-tuning adaptive model predictive controller is constructed by combining a recursive model parameter estimator with a model predictive controller. The recursive model parameter estimator is designed using a recursive least squares algorithm with Kalman filter interpretation. The effectiveness of the proposed controller is verified through control simulations and a set-point tracking experiment on the electron density with artificial drift in real time. The experimental results show that the performance of the proposed controller is 21% better than that of the conventional model predictive controller. We expect that this result will make a significant contribution to control processes utilizing plasma.

Instant messaging applications store users' personal data (e.g., user profile, chat messages, photos and video clips). Because those data typically include privacy sensitive information, most instant messaging applications are trying to protect the stored data in an encrypted form so that the authorized messaging application itself can only access the data. In this paper, we analyzed the locations and file formats of personal data files in three instant messaging applications (KakaoTalk, NateOn, and QQ) which are the most popularly used in China and South Korea. We particularly examined the encryption and decryption procedures for internal databases in those messaging applications through reverse-engineering. Our analysis results demonstrate how the database files of those instant messaging applications are stored and encrypted. Moreover, in the cases of KakaoTalk and NateOn applications, we found that their encrypted database files can successfully be recovered without requiring user password. We also found that QQ messenger stores the encryption key for the database files into an external server. This implementation may raise another privacy concern because users’ personal data can be freely accessed by the service provider without user consent.

This letter considers simultaneous wireless information and power transfer (SWIPT) in multiple-input single-output broadcasting systems based on the integer forcing (IF) technique where multiple receivers are equipped with a power splitting circuit and able to harvest energy and decode information at the same time. The optimization problem that incorporates the beamforming matrix and IF coefficients matrix designs is formulated to minimize the transmit power at transmitter, while guaranteeing the achievable rate and harvested energy threshold of each receiver. To tackle the NP-hardness and non-convexity of the optimization problem, we propose an iterative algorithm consisting of two sub-problems. Simulation results provide the comparison of the performance of SWIPT systems with and without IF technique.

We propose a new simple and low-cost alignment method utilizing electric contact for die stacking on an interposer or a printed circuit board. Without any expensive aligning/bonding equipment assisted by microscopy, we successfully stacked and bonded two silicon dies on a printed circuit board achieving alignment accuracy of $20~\mu \text{m}$ by utilizing only cheap equipment: manual micropositioners, a stage, a stand, a heat gun, a hot plate, and a power supply. In the proposed method, conductive alignment marks are fabricated on the bonding surfaces, and the electric contact between them are used to characterize misalignment. Therefore, the proposed method does not require any expensive complex microscopy equipment for alignment. In addition, alignment can be quickly and simply examined in situ by detecting electrical conduction during bonding. In experiment, alignment accuracy better than $20~\mu \text{m}$ and 100% bonding yield was measured in testing of 1800 bonds connecting two silicon interposer dies which were stacked on a printed circuit board by the proposed method.

Ionic liquid-based polymer electrolyte composites (PECs) with high mechanical strength and ionic conductivity were fabricated by using a semicrystalline homopolymer network. Morphological, mechanical, and electrical properties of semicrystalline homopolymer-based PECs were investigated over the composition range of 10–50 polymer wt%. The PECs consisted of poly(vinylidene fluoride) (PVDF) and a room-temperature ionic liquid, 1-ethyl-3-methylimidazolium bis(trifluoromethylsulfonyl)imide ([EMI][TFSI]). The amorphous PVDF chains were solvated by the liquid electrolyte and interconnected the phase-separated PVDF crystalline-domains, acting as physical crosslinks. The ionic conductivity and specific capacitance of the homopolymer PECs were found to be very high, ranging from 0.4 to 4.9 mS/cm and from 2.6 to 9.1 μF/cm2, respectively. The elastic moduli of the PECs that range from 4 to 38 MPa were also greater than those of PECs prepared by using other polymeric networks. These PECs were successfully applied as electrolyte membranes in electrical double layer supercapacitors and as high-capacitance gate insulators in electrochemical thin-film transistors. These results provide detailed systematic data for fabricating high-performance solid polymer electrolytes for electrochemical thin-film devices.

As network attacks are constantly and dramatically evolving, demonstrating new patterns, intelligent Network Intrusion Detection Systems (NIDS), using deep-learning techniques, have been actively studied to tackle these problems. Recently, various autoencoders have been used for NIDS in order to accurately and promptly detect unknown types of attacks (i.e., zero-day attacks) and also alleviate the burden of the laborious labeling task. Although the autoencoders are effective in detecting unknown types of attacks, it takes tremendous time and effort to find the optimal model architecture and hyperparameter settings of the autoencoders that result in the best detection performance. This can be an obstacle that hinders practical applications of autoencoder-based NIDS. To address this challenge, we rigorously study autoencoders using the benchmark datasets, NSL-KDD, IoTID20, and N-BaIoT. We evaluate multiple combinations of different model structures and latent sizes, using a simple autoencoder model. The results indicate that the latent size of an autoencoder model can have a significant impact on the IDS performance.