Search

In the pursuit of rapid atomic migration in lightweight Fe–Al diffusion couples, rationally designing short-circuit diffusion paths has become paramount. Herein, a strain-mediated defect engineering strategy was proposed for reducing the vacancy activation energy and enhancing diffusion behaviors along dislocations (DLs) and grain boundaries (GBs). Combining the modified Arrhenius-type relationship, an interfacial apparent activation energy of 139 kJ mol–1was acquired utilizing defect engineering, which was decreased by about 49%. This was closely related to high-density vacancies, DLs, and GBs formed in strained Fe and Al materials, which provided more low activation energy paths for atomic migration. First-principles calculations indicated that the lattice diffusion barrier mediated by monovacancy was reduced with strain incorporation, attributed to the weakened atom–vacancy bond as a consequence of less electron transport. The synergistic effect of abnormal electron-charge distribution in the bulk and strong attraction force at the Al/Fe interface radically resulted in rapid atomic migration, collectively regulating the “breaking–forming bond” process.

Existing approaches for vehicle repositioning on large-scale ride-hailing platforms either ignore the spatial-temporal mismatch between supply and demand in real-time or overlook the long-term balance of the system. To account for both, we propose a lookahead repositioning policy in this paper, which is a novel approach to repositioning idle vehicles from both a dynamic system and a long-term performance perspective. Our method consists of two parts; the first part utilizes linear programming (LP) to formulate the nonstationary system as a time-varying, $T$ -step lookahead optimization problem and explicitly models the fraction of drivers who follow repositioning recommendations (called the repositioning rate). The second step is to incorporate a reinforcement learning (RL) method to maximize long-term return based on learned value functions after the $T$ time slots. Extensive studies utilizing a real-world dataset on both small-scale and large-scale simulators show that our method outperforms previous baseline methods and is robust to prediction errors.

AbstractWe used reinforcement learning to improve how Lyft matches drivers and riders. The change was implemented globally and led to more than $30 million per year in incremental driver revenue.

In reinforcement learning, a promising direction to avoid online trial-and-error costs is learning from an offline dataset. Current offline reinforcement learning methods commonly learn in the policy space constrained to in-support regions by the offline dataset, in order to ensure the robustness of the outcome policies. Such constraints, however, also limit the potential of the outcome policies. In this paper, to release the potential of offline policy learning, we investigate the decision-making problems in out-of-support regions directly and propose offline Model-based Adaptable Policy LEarning (MAPLE). By this approach, instead of learning in in-support regions, we learn an adaptable policy that can adapt its behavior in out-of-support regions when deployed. We give a practical implementation of MAPLE via meta-learning techniques and ensemble model learning techniques. We conduct experiments on MuJoCo locomotion tasks with offline datasets. The results show that the proposed method can make robust decisions in out-of-support regions and achieve better performance than SOTA algorithms.

Creating a man‐made life in the laboratory is one of life science’s most intriguing yet challenging problems. Advances in synthetic biology and related theories, particularly those related to the origin of life, have laid the groundwork for further exploration and understanding in this field of artificial life or man‐made life. But there remains a wealth of quantitative mathematical models and tools that have yet to be applied to this area. In this paper, we review the two main approaches often employed in the field of man‐made life: the top‐down approach that reduces the complexity of extant and existing living systems and the bottom‐up approach that integrates well‐defined components, by introducing the theoretical basis, recent advances, and their limitations. We then argue for another possible approach, namely “bottom‐up from the origin of life”: Starting with the establishment of autocatalytic chemical reaction networks that employ physical boundaries as the initial compartments, then designing directed evolutionary systems, with the expectation that independent compartments will eventually emerge so that the system becomes free‐living. This approach is actually analogous to the process of how life originated. With this paper, we aim to stimulate the interest of synthetic biologists and experimentalists to consider a more theoretical perspective, and to promote the communication between the origin of life community and the synthetic man‐made life community.

Ride hailing has become prevailing. Central in ride hailing platforms is taxi order dispatching which involves recommending a suitable driver for each order. Previous works use pure combinatorial optimization solutions for taxi dispatching, which suffer in practice due to complex dynamics of demand and supply and temporal dependency among dispatching decisions. Recent studies try to adopt data-driven method into combinatorial optimization hoping knowledge from history data would help overcome these challenges. Among these attempts, adoption of reinforcement learning shows great promise but current adoptions are a unidirectional integration which restricts the potential performance gains. In this work, we propose Learning To Dispatch(LTD), a systematic solution that allows synergic integration of reinforcement learning and combinatorial optimization for large-scale taxi order dispatching. We demonstrate the necessity of online learning and taxi scheduling for reinforcement learning to work in synergy with combinatorial optimization, and devise corresponding algorithms. We also devise many tricks for more efficient calculation of the bipartite matching. Experiments show our methods can improve $36.4\%$36.4% and $42.0\%$42.0% on utility and efficiency at most, respectively. Especially, it achieves state-of-the-art performance in terms of utility.

Solid-state lithium batteries (SSLBs) based on Ta-doped Li6.5La3Zr1.5Ta0.5O12(LLZTO) suffer from lithium dendrite growth, which hinders their practical application. Herein, first principles simulations indicate that the Ta element prefers to segregate along grain boundaries in the form of Ta2O5precipitates due to a high energy difference induced by Ta doping. Grain boundary engineering is employed to regulate the distribution of the Ta element and enhance the density of LLZTO by introducing the La2O3additive. The sufficient La2O3additive reacts with the Ta2O5precipitates, while the residual La2O3nanoparticles fill up void defects, promoting the homogeneous distribution of the Ta element and improving the relative density to ∼98%. Critical current density of the symmetric Li battery reaches 2.12 mA·cm–2at room temperature with the solid-state electrolyte (LLZTO + 5 wt % La2O3), which increases by 41% compared to pure LLZTO. SSLBs with the LiFePO4cathode achieve a stable cycling performance with a discharge capacity of 138.6 mA·h·g–1after 400 cycles at 0.2 C. This work provides theoretical insights into the distribution of Ta-doped LLZTO and inhibits lithium dendrite growth through grain boundary engineering.

Strength-ductility trade-off dilemma remains a significant obstacle to high-strength composites due to the undesirable dislocation storability. Herein, an ingenious nano-micro SiC-reinforced Al matrix composite (AMC) with heterogeneous grain structures of coarse and fine grains was designed viaa novel deformation-driven metallurgy method. The accumulated geometrically necessary dislocations and the intragranularly dispersed SiC particles were tailored based on the principle of back stress amelioration, which was a key point to maintain the dynamic balance with the applied stress toward strength-ductility synergy. The ultimate tensile strength and uniform elongation of the designed SiC5np-5µp/Al composite reached 324 MPa and 12.9%, respectively, and the strength was 181% as high as that of the SiC10µp/Al with only 3% ductility loss. As such, a new strategy was provided herein to promote strength-ductility synergy viathe further modified back stress.

Currently, the exploration of wearable strain sensors that can work under subzero temperatures while simultaneously possessing anti-interference capability toward temperature is still a grand challenge. Herein, we present a low-temperature wearable strain sensor that is constructed via the incorporation of a Ag nanowires/graphene (Ag NWs/G) composite into the polydimethylsiloxane (PDMS) polymer. The Ag NWs/G/PDMS strain sensor exhibits promising flexibility at a very low temperature (−40 °C), outstanding fatigue resistance with low hysteresis energy, and near-zero temperature coefficient of resistance (TCR). The Ag NWs/G/PDMS strain sensor shows excellent sensing performance under subzero temperatures with a very high gauge factor of 9156 under a strain of >36%, accompanied by a noninterference characteristic to temperature (−40 to 20 °C). The Ag NWs/G/PDMS strain sensor also demonstrates the feasibility of monitoring various human movements such as finger bending, arm waving, wrist rotation, and knee bending under both room temperature and low-temperature conditions. This work initiates a new promising strategy to construct next-generation wearable strain sensors that can work stably and effectively under very low temperatures.

Solid-state lithium batteries (SSLBs) based on garnet-type solid-state electrolytes (SSEs) have attracted much attention due to their high energy density and chemical stability. However, poor room-temperature ionic conductivity and low density of SSEs induced by conventional preparation routes limit their potential future applications. In this work, an oriented attachment strategy is employed to enhance the Li-ion conductivity and density of garnet-type SSE Li6.5La3Zr1.5Ta0.5O12by introducing La2O3nanoparticles. The oriented attachment of the ZrO2(Ta2O5) matrix mediates the epitaxial growth of the La–Zr(Ta)–O intermediate phase due to the addition of La2O3nanoparticles. Continuous Li-ion transport pathways along grain boundaries are produced by the combination of residual La2O3and gas Li2O. A densification interface is obtained when 10 wt % La2O3is doped. The maximum value of Li-ion conductivity reaches 8.20 × 10–4S·cm–1, with a relative density of 97.3%. SSLBs with a LiFePO4cathode showing a stable cycling performance with a discharge capacity of 123.1 mA·h·g–1and a Coulombic efficiency of 99.2% after 300 cycles (0.5C) at room temperature. This work is comparable to the state-of-the-art methodology, which provides a feasible approach to creating SSEs with high performances for SSLBs.

Deformation-driven metallurgy was implemented to prepare graphene nanoplatelet (GNP)-reinforced aluminum matrix composites with a time-dependent self-enhancement in corrosion resistance. Severe plastic deformation contributed to the sufficient brokenness, thinning, enfolding, and redispersion of GNPs, as well as grain refinement. The homogeneously dispersed GNPs showed a great corrosion inhibition mechanism in a chloride-containing environment, ascribed to the formation of a carbon-doped protective film via diffusion and chemical bonding between GNPs and the surface oxide film. Electrochemical and intergranular corrosion tests were conducted to show the enhancement of long-term corrosion resistance. First-principles calculations were performed to explore the high corrosion resistance of the carbon-doped protective film. The energy barriers of vacancy formation, Cl ingress, and charge transfer were synchronously enhanced with the addition of GNPs into aluminum matrix composites as long-term corrosion inhibitors.

The service areas of the BeiDou Navigation Satellite System (BDS) are China, the Asia–Pacific, and the world via three stages of development. The BDS constellation is designed to maintain a geostationary Earth orbit (GEO), inclined geosynchronous orbit (IGSO), and medium Earth orbit (MEO). These orbital maneuvers yield certain difficulties for data processing, especially for the BeiDou satellites, such as a decrease in the real-time service performance, which causes a missing precise orbit product from the maneuvered satellites. The frequencies of the orbital maneuvers for the GEO and IGSO satellites are higher than those for the MEO satellites. The maneuvering time and strategy cannot be obtained by common users due to secrecy, which can yield a decline in the service performance of the BDS. Thus, in this study, we analyzed the variation in the orbital semimajor axis for the satellites. The long-term variation in the orbital semimajor axis changes linearly. The short-term variation in the orbital semimajor axis also has periodicity, which can be described by the sum of sine functions. According to the long-term variation in the orbital semimajor axis for the satellites, we propose a detection method for in-plane orbital maneuvering of the GEO, IGSO, and MEO satellites. We then propose a detection method for out-of-plane maneuvering of the GEO satellites. BDS and Global Positioning System (GPS) data from the broadcast ephemeris were analyzed to verify the proposed methods. The experimental results from 2013 to 2019 show that the frequency of orbital maneuvering is approximately once a month for the GEO satellites, once every six months for the IGSO satellites, and once every 1.4 years for the MEO satellites.

Order dispatching is instrumental to the marketplace engine of a large-scale ride-hailing platform, such as the DiDi platform, which continuously matches passenger trip requests to drivers at a scale of tens of millions per day. Because of the dynamic and stochastic nature of supply and demand in this context, the ride-hailing order-dispatching problem is challenging to solve for an optimal solution. Added to the complexity are considerations of system response time, reliability, and multiple objectives. In this paper, we describe how our approach to this optimization problem has evolved from a combinatorial optimization approach to one that encompasses a semi-Markov decision-process model and deep reinforcement learning. We discuss the various practical considerations of our solution development and real-world impact to the business.

Cost-effective and convenient additive manufacturing technologies are recognized as suitable alternatives for achieving lightweight aluminum alloy components with dense and defect-free parts toward high mechanical properties. Herein, a high-efficient hybrid arc and friction stir additive manufacturing (HAFS-AM) technique was employed to fabricate a high-performance Al-Cu-Mg-Ag-Zr alloy part. The AC pulsed tungsten inert gas (TIG) was utilized for the rapid pre-deposition of Al-Cu-Mg-Ag-Zr alloy with high efficiency. Friction stir processing contributed to eliminating pore defects and refining coarse microstructure in the pre-deposited arc layers, accounting for enhancing mechanical properties. During the HAFS-AM, the coarse θ phase with continuous network morphology was completely broken and gradually dissolved into the α-Al matrix. The addition of Zr element promoted the precipitation of coherent L12-Al3Zr inducing dislocation shearing behavior. The Mg and Ag additions accelerated the precipitation of semi-coherent plate-like Ω and σ phases, where a critical size (thickness and diameter) of precipitates existed for deciding the behaviors of dislocation shearing and dislocation looping. The ultimate tensile strength and elongation of the HAFS-AMed steady-state layer reached 262 MPa and 18.8 %, which were increased by 21 % and 96 % compared to that of the arc layer, respectively. The combined effect of arc and friction stir with high efficiency achieved high-performance manufactured components, expediting the future industry application of advanced additive manufacturing technology.

The electrochemical nitrate reduction reaction (NO3RR) is considered as a promising strategy to degrade nitrate-containing wastewater and synthesize recyclable ammonia at atmospheric pressure and room temperature. In this work, the copper oxides-derived nano-polycrystalline Cu (NPC Cu) was integrated with Ti3+-self-doped TiO2nanotube arrays (NTA) to fabricate the NPC Cu/H–TiO2NTA. Ti3+-self-doped TiO2NTAs and the NPC Cu facilitate electron transfer and mass transportation and create abundant active sites. The unique nanostructure in which Cu nano-polycrystals interlace with the TiO2nanotube accelerates the electron transfer from the substrate to surface NPC Cu. The density functional theory calculations confirm that the built-in electric field between Cu and TiO2improves the adsorption characteristic of the NPC Cu/H–TiO2NTA, thereby converting the endothermic NO3–adsorption step into an exothermic process. Therefore, the high NO3–conversion of 98.97%, the Faradic efficiency of 95.59%, and the ammonia production yield of 0.81 mg cm–2h–1are achieved at −0.45 V vsreversible hydrogen electrode in 10 mM NaNO3(140 mg L–1)–0.1 M Na2SO4. This well-designed NPC Cu/H–TiO2NTA as an effective electrocatalyst for the 8e–NO3RR possesses promising potential in the applications of ammonia production.

A green solid-state recycling strategy for recycling short-circuited SSEs is proposed in this work, contributing to SSLBs large-scale industrial application and green recycling study.

During past decades, the capacity of NAND flash memory has been increasing dramatically, leading to the use of nonvolatile flash in the system’s memory hierarchy. The increasing capacity of NAND flash memory introduces a large RAM footprint to store the logical to physical address mapping. The demand-based approach can effectively reduce and well control the RAM footprint. However, extra address translation overhead is also introduced which may degrade the system performance.In this article, we present CDFTL, an adaptive Caching mechanism for Demand-based Flash Translation Layer, for NAND flash memory storage systems. CDFTL adopts both the fine-grained entry-based caching mechanism to exploit temporal locality and the coarse-grained translation-page-based caching mechanism to exploit spatial locality of workloads. By selectively caching the on-demand address mappings and adaptively changing the space configurations of two granularities, CDFTL can effectively utilize the RAM space and improve the cache hit ratio. We evaluate CDFTL under a real hardware embedded platform using a variety of I/O traces. Experimental results show that our technique can achieve an 11.13% reduction in average system response time and a 35.21% reduction in translation block erase counts compared with the previous work.

In this study, a new dithiolopyrrolone biosynthetic pathway was identified in Saccharothrix algeriensis NRRL B-24137, which was reported to produce a variety of dithiolopyrrolone natural products including thiolutin, a potential drug candidate for tumor angiogenesis inhibition. Bioinformatics analysis of the cluster revealed that it contains all the essential genes for holothin core biosynthesis and several other auxiliary genes. Interestingly, heterologous expression of the gene cluster in Streptomyces albus only induced the production of holomycin, implying that the gene responsible for the N4-methylation and the gene(s) involved in the formation of various acylated chains on N4 position of the holothin may locate outside the gene cluster. Incubation of holomycin with S-adenosyl-L-methionine (SAM) in the cell-free extract of Sa. algeriensis resulted in the production of thiolutin, suggesting that the N4-methyl group of thiolutin is originated from SAM, and the N4-methylation could be in the late stage of biosynthesis of thiolutin type dithiolopyrrolones. An evolution-based model for biosynthesis of thiolutin and its analogs was further proposed based on these results.

Bent varieties and straight varieties were made as parents for the genetic analysis to investigate cucumber bending genetic mechanism. The results showed that the bent characters of the cucumber fruit (BCCF) were quantitative inheritance controlled by multiple genes and major genes. The additive effect played the main role and the dominance effect played the lesser role. Compared with the additive environmental variance, the dominant-environmental variance was more important and the cucumber fruit was more easily affected by the additive effect. The broad heritability and the narrow heritability of BCCF were both higher. The varieties of D0455 and D07299 could be used as parents which were benefit for improving the straight characters of the cucumber fruit.

In this article, we focus on solving the energy optimization problem for real-time streaming applications on multiprocessor System-on-Chip by combining task-level coarse-grained software pipelining with DVS (Dynamic Voltage Scaling) and DPM (Dynamic Power Management) considering transition overhead, inter-core communication and discrete voltage levels. We propose a two-phase approach to solve the problem. In the first phase, we propose a coarse-grained task parallelization algorithm called RDAG to transform a periodic dependent task graph into a set of independent tasks by exploiting the periodic feature of streaming applications. In the second phase, we propose a scheduling algorithm, GeneS, to optimize energy consumption. GeneS is a genetic algorithm that can search and find the best schedule within the solution space generated by gene evolution. We conduct experiments with a set of benchmarks from E3S and TGFF. The experimental results show that our approach can achieve a 24.4% reduction in energy consumption on average compared with the previous work.

Abstract: Out of 851 soybean accessions from Vietnam, China and Japan analyzed for 7S β-subunit variants, a new β-reduced subunit line with normal growth was collected from the Mekong Delta, Vietnam. Protein of the `β-reduced' line is composed of β-reduced and extremely low-β types. (α-null + β-reduced) type and β-null (or extremely low-β) type were screened in the progeny seeds of the Japanese mutant `(α + β) null' line. Therefore, recombination between α- and β-subunits was identified. By comparing the nucleotide sequence of the partial β-subunit gene of Enrei (standard), Mo-shi-dou Gong 503 (α + β low), β-533 seed (β-reduced), β-57 seed (extremely low-β), and (α + β)-null16((α + β) null) seed, we found that the base `T' at 166 bp, in Enrei changed to `G' in Mo-shi-dou Gong 503, β-533 and β-57. Using the (α + β)-null16 individual as template, a distinct 305 bp β-subunit gene fragment was identified, instead of a 285 bp fragment.

An important part of deepening our understanding of the reaction mechanism of natural gas oxidation is validating a detailed model against experimental flame and ignition data. Ignition delays in shock‐heated CH4–O2mixtures, with and without ignition‐promoting additives, account for most of the available validation data. Most, but not all, ignition delays in such mixtures reported in the literature have been found to agree with calculations done using the GRI‐Mech 1.2 model. Additional reflected shock wave ignition experiments were done to explore possible reasons for the mismatches and to study the promotion of CH4ignition by small amounts of C2H6and C3H8additives. Empirical correlations were derived that describe ignition delays in CH4–O2–C2H6(C3H8).