Search

Postconsumer plastics are generally perceived as valueless with only a small portion of plastic waste being closed-loop recycled into similar products while most of them are discarded in landfills. Depositing plastic waste in landfills not only harms the environment but also signifies a substantial economic loss. Alternatively, constructing value-added chemical feedstocks via mining the waste-derived intermediate species as a carbon (C) source under mild electrochemical conditions is a sustainable strategy to realize the circular economy. This proof-of-concept work provides an attractive turning trash to treasure strategy by integrating electrocatalytic polyethylene terephthalate (PET) plastic upcycling with a chemical C-S coupling reaction to synthesize organosulfur compounds, hydroxymethanesulfonate (HMS). HMS can be produced efficiently (Faradaic efficiency, FE of ∼70%) via deliberately capturing electrophilic intermediates generated in the PET monomer (ethylene glycol, EG) upcycling process, followed by coupling them with nucleophilic sulfur (S) species (i.e., SO32- and HSO3-). Unlike many previous studies conducted under alkaline conditions, PET upcycling was performed over an amorphous MnO2 catalyst under near-neutral conditions, allowing for the stabilization of electrophilic intermediates. The compatibility of this strategy was further investigated by employing biomass-derived compounds as substrates. Moreover, comparable HMS yields can be achieved with real-world PET plastics, showing its enormous potential in practical application. Lastly, Density function theory (DFT) calculation reveals that the C-C cleavage step of EG is the rate-determining step (RDS), and amorphous MnO2 significantly decreases the energy barriers for both RDS and C-S coupling when compared to the crystalline counterpart.

Laser-powder DED (LP-DED) has been investigated extensively for the rapid prototype of complex and customized metallic parts and in-situ repairing tasks. However, unlike the material extrusion type which allows a relatively large overhang angle, the conventional 2.5-axis LP-DED faces a severe overhang issue that deteriorates the printing stability and quality. While five-axis LP-DED is able to effectively resolve the overhang problem, it however introduces new kinematic issues and limited nozzle orientation range due to physical restrictions of DED. The existing computer numerical control systems are mainly designed for machining, where the above issues can be mostly solved by feedrate scheduling, which, however, is not applicable for LP-DED processes, as the feedrate must be preserved as programmed lest deposition height would be altered since in most cases it is not practical to vary the powder flow rate in real-time. In this paper, a feedrate-preserved motion planning method is proposed to optimize the rotational angles of the machine tool in the machine coordinate system (MCS) under the constraints of MCS and workpiece coordinate system (WCS). Given a five-axis DED path as well as its accompanying initial nozzle orientation in WCS, we first map them via inverse kinematics to MCS, which is represented by a cylindrical coordinate system. Then, the motion path in the cylindrical coordinate system is optimized to avoid the singularity. Finally, the physical constraint of nozzle orientation in WCS is introduced into the time-weighted constrained Laplacian smoothing in MCS with the axial speed limits upheld. The experimental results confirm that the actual feedrate is successfully preserved as the theoretical one under the constraints of the axial kinematic limits, while the singularity is avoided.

Freeform surface inspection is vital in manufacturing-related processes, and the newly emerged five-axis continuous sweep scanning technology offers to be a powerful means for accurate inspection of freeform surfaces. However, at present, for an arbitrary freeform surface, it still heavily depends on human intervention in planning an interference-free sweep scanning path, wherein typically the surface is simply partitioned into several subregions and sweep scanning paths are then individually planned for each subregion. In this paper, we present a novel and practical algorithm for automatically planning an efficient five-axis inspection path for an arbitrary freeform surface. The strategy is based on the idea of converting the inspection path planning problem into a set-covering problem, whose solution then gives out a near-minimum set of inspection paths for the freeform surface while subject to necessary constraints such as sampling resolution, collision-free, and others. Both computer simulation and physical inspection experiments of the proposed method have been conducted, and the results give a preliminary confirmation on the effectiveness and advantages of the proposed method.

Janzen-Connell effects state that the accumulation of host-specific natural enemies near parent plants can negatively affect their offspring. Negative plant-soil feedbacks can produce patterns of seedling performance predicted by Janzen-Connell effects and influence plant populations, but their relevance in field conditions remains unclear. Here, using spatial point-pattern analysis, we examine the spatial distribution of Jacobaea vulgaris to assess whether distance- and density-dependent predictions of Janzen-Connell effects are evident in the field. We established 27 replicated 64 m2 plots at two grassland sites and mapped positions of rosette-bearing and flowering J. vulgaris plants within each plot. To investigate temporal distribution patterns, we tracked plant positions repeatedly in three plots during a single season. Additionally, we tested whether these patterns are soil-mediated. Soil samples were collected underneath flowering plants and at a distance of 0.5 meter, and used to compare seed germination, seedling survival, and growth under controlled conditions. Furthermore, we measured J. vulgaris growth in soil from patches with high J. vulgaris densities and in soil from areas outside these patches. The density of rosette-bearing plants was lower at close distances from flowering plants than expected from null models, suggesting negative distance-dependent plant recruitment. The degree of clustering decreased over time from rosette-bearing to flowering plants, indicating density-dependent self-thinning. Seed germination was higher in soil further away from flowering J. vulgaris plants than in soil underneath plants at one site, but soil distance was not an overall significant factor in explaining seed germination. However, seedling mortality and biomass did not differ between soils collected at the two distances, and plants produced similar biomass in soil collected from inside and outside J. vulgaris patches. Our study demonstrates conspec

For rough machining of a complex narrow cavity, e.g., a complex blisk channel on an aero-engine, the typically used cutting tools are the slender cylindrical cutter and conical cutter. Nevertheless, as neither of the two is particularly suited for rough machining, wherein the main purpose is to remove a large volume as quickly as possible, the machining efficiency is low, especially when the part materials are of hard-to-cut types (e.g., Titanium-alloy) for which it often takes days to rough machine a blisk. Fortunately, disc machining provides a new and efficient roughing solution, since a disc cutter with a large radius enables a much larger cutting speed and thus a larger material removal rate. However, due to the large radius of the disc cutter, its potential collision with narrow and twisted channels becomes a serious concern. In this paper, we propose a novel two-phase approach for efficiently machining a complex narrow cavity workpiece using a disc-shaped cutter, i.e., 3 + 2-axis disc-slotting of the channel by multiple layers (rough machining) + five-axis disc-milling of the freeform channel side surfaces (semi-finish machining). Both simulation and physical cutting experiments are conducted to assess the effectiveness and advantages of the proposed method. The experimental results show that, with respect to a same cusp-height threshold on the channel side surfaces, the total machining time of the tested part by the proposed method is about only 36% of that by the conventional approach of plunging-milling (for roughing) plus milling by a slender cylindrical cutter (for semi-finishing). © 2023 Chinese Society of Aeronautics and Astronautics

The dark photon is a promising candidate for the dark matter which comprises most of the matter in our visible Universe. Via kinetic mixing with the Standard Model it can also be resonantly converted to photons in an electromagnetic cavity, offering novel experimental possibilities for the discovery and study of dark matter. We report the results of a pathfinder dark photon dark matter cavity search experiment performed at Hunan Normal University and the Institute of Physics, Chinese Academy of Sciences, representing the first stage of the APEX (Axion and dark Photon EXperiment) program. Finding no statistically significant excess, we place an upper limit on the kinetic mixing parameter $|\chi|<3.7\times 10^{-13}$ around $m_A\simeq 29.5$ $\mu$eV at 90% confidence level. This result exceeds other constraints on dark photon dark matter in this frequency range by roughly an order of magnitude., Comment: 5 pages, 4 figures

Possible light bosonic dark matter interactions with the Standard Model photon have been searched by microwave resonant cavities. In this paper, we demonstrate the cryogenic readout system calibration of a 7.138 GHz copper cavity with a loaded quality factor $Q_l=10^4$, operated at 22 mK temperature based on a dilution refrigerator. Our readout system consists of High Electron Mobility Transistors as cryogenic amplifiers at 4 K, plus room-temperature amplifiers and a spectrum analyzer for signal power detection. We test the system with a superconducting two-level system as a single-photon source in the microwave frequency regime and report an overall 95.6 dB system gain and -71.4 dB attenuation in the cavity's input channel. The effective noise temperature of the measurement system is 7.5 K., Comment: 7 pages, 5 figures, version to appear in CPC

Embedding visual representations within original hierarchical tables can mitigate additional cognitive load stemming from the division of users' attention. The created hierarchical table visualizations can help users understand and explore complex data with multi-level attributes. However, because of many options available for transforming hierarchical tables and selecting subsets for embedding, the design space of hierarchical table visualizations becomes vast, and the construction process turns out to be tedious, hindering users from constructing hierarchical table visualizations with many data insights efficiently. We propose InsigHTable, a mixed-initiative and insight-driven hierarchical table transformation and visualization system. We first define data insights within hierarchical tables, which consider the hierarchical structure in the table headers. Since hierarchical table visualization construction is a sequential decision-making process, InsigHTable integrates a deep reinforcement learning framework incorporating an auxiliary rewards mechanism. This mechanism addresses the challenge of sparse rewards in constructing hierarchical table visualizations. Within the deep reinforcement learning framework, the agent continuously optimizes its decision-making process to create hierarchical table visualizations to uncover more insights by collaborating with analysts. We demonstrate the usability and effectiveness of InsigHTable through two case studies and sets of experiments. The results validate the effectiveness of the deep reinforcement learning framework and show that InsigHTable can facilitate users to construct hierarchical table visualizations and understand underlying data insights.

We demonstrate MaskSearch, a system designed to accelerate queries over databases of image masks generated by machine learning models. MaskSearch formalizes and accelerates a new category of queries for retrieving images and their corresponding masks based on mask properties, which support various applications, from identifying spurious correlations learned by models to exploring discrepancies between model saliency and human attention. This demonstration makes the following contributions:(1) the introduction of MaskSearch's graphical user interface (GUI), which enables interactive exploration of image databases through mask properties, (2) hands-on opportunities for users to explore MaskSearch's capabilities and constraints within machine learning workflows, and (3) an opportunity for conference attendees to understand how MaskSearch accelerates queries over image masks.

While multi-axis additive manufacturing is found to be a good solution to the inherent limitations of conventional 2.5D additive manufacturing, it is a much more sophisticated process. Among different existing multi-axis process planning algorithms, we are interested in those based on a scalar field, in which print slices are the iso-surfaces of a scalar field embedded in the 3D model. In this paper, we propose a partitioned-based print sequence planning algorithm and an adaptive slicing algorithm, which together determine a complete multi-axis printing process for an arbitrary solid model. The first algorithm iteratively subdivides the model into a set of components such that a collision-free print sequence can be established among the components. The second algorithm then extracts print slices from each component such that all these slices satisfy the self-support condition. Since an arbitrary model may not satisfy both the self-support and collision-free requirements, we also define certain critical printability rules at the beginning to check whether a given input model with its associated scalar field is printable. The generated print slices and print sequence by the proposed two algorithms are guaranteed to be printable. Furthermore, a shorter total fabrication time and a better surface quality are achieved. Physical experiments of four test models are performed on a homebuilt multi-axis FDM printer, whose results verify the capabilities of the proposed algorithms. © 2023 Elsevier Ltd

The development of functional materials for tumor immunogenicity enhancement is desirable for overcoming the low therapeutic efficiency and easy metastasis during tumor treatments. Herein, the thermoresponsive nanoparticles composed of photothermal agent (PTA) and click reactive reagent are developed for enhanced immunotherapy application. A Ni-bis(dithiolene)-containing PTA with intense near-infrared absorption and efficient photothermal conversion is developed for thermoresponsive nanoparticles construction. The generated heat by encapsulated PTA further induces the phase transition of thermoresponsive nanoparticles with the release of chemotherapy reagent to react with the amino groups on functional proteins, realizing PTT and chemotherapy simultaneously. Moreover, the immunogenic cell death (ICD) of cancer cells evoked by PTT could be further enhanced by the released reactive reagent. As a result, the synergistic effect of photothermal treatment and reaction-mediated chemotherapy can suppress the growth of a primary tumor, and the evoked ICD could further activate the immune response with the suppression of a distant tumor. This synergistic treatment strategy provides a reliable and promising approach for cancer immunotherapy in clinic. © 2023 American Chemical Society.

Annular parts are widely utilized in high-end manufacturing, which are often manufactured by ring forging to meet the requirement on mechanical properties. However, residual stress (RS) is inevitably introduced in the forging process, which affects the working performance and reduces the service life of the part. Therefore, to counter the determent of RS, it is first necessary to accurately obtain it. Different from sheet or structural parts, the RS distribution in an annular part is more complex. Existing methods based on RS measurement and finite element prediction did not incorporate the RS prior, so the accuracy is influenced by the simplified assumption on RS distribution. This paper proposes a new RS inferencing method for annular parts based on the use of deformation force and physical prior. Firstly, the RS distribution prior of the annular blank after heat treatment is obtained by finite element simulation and the clustering algorithm. Then, the prior is employed to calculate the coefficient matrix to reflect the relationship between the deformation force and the RS. Finally, the Tikhonov regularization method is used to solve the deformation force-RS inverse problem. The effectiveness and feasibility of the proposed method are verified in the simulation and experiment, respectively, which shows that the integration of RS prior distribution is able to reduce the ill-condition of the solving process to obtain an accurate solution of RS distribution in an annular part.

Flank milling of a surface means to use the linear side of a tool to machine the surface while subject to all the necessary constraints such as collision-free and maximum cusp height. For a general freeform surface, it is always necessary to require multiple passes of a tool path to machine the surface. Because of the much-enlarged material removal rate by flank milling, at least at the semi-finish stage, it is preferred over the conventional machining type of point-milling, especially for those heavy and large industrial parts such as centrifugal impellers and turbine blades. Unfortunately, due to the muchworsened tool-surface collision condition in flank milling, for a complex part surface, it is extremely difficult to determine a correct collision-free multi-pass flank milling tool path while satisfying the maximum cusp height requirement. In this paper, we present an algorithm for this problem. The crux of our method is the establishment of a collision-conscious vector field on an offset surface of the part surface, which encapsulates degrees of both local gouging and global accessibility. This offset mesh is further segmented such that each patch will have only a curl-free vector field. The level-set method is then utilized to fit a scalar field on each patch whose iso-curves naturally form a set of initial tool cutter contact rulings, which are further optimized and propagated to constitute a stripification of the patch comprising a set of quad strips, while respecting the constraints of no-overcut, collision-free, and maximum cusp height. Finally, for each patch, a multi-pass flank milling tool path is generated based on the rulings of the optimized quad strips. Both computer simulation and physical cutting experiments have been carried out, and the experimental results give a positive confirmation on the correctness and effectiveness of the proposed method.

We introduce VOCALExplore, a system designed to support users in building domain-specific models over video datasets. VOCALExplore supports interactive labeling sessions and trains models using user-supplied labels. VOCALExplore maximizes model quality by automatically deciding how to select samples based on observed skew in the collected labels. It also selects the optimal video representations to use when training models by casting feature selection as a rising bandit problem. Finally, VOCALExplore implements optimizations to achieve low latency without sacrificing model performance. We demonstrate that VOCALExplore achieves close to the best possible model quality given candidate acquisition functions and feature extractors, and it does so with low visible latency (~1 second per iteration) and no expensive preprocessing.

We introduce EQUI-VOCAL: a new system that automatically synthesizes queries over videos from limited user interactions. The user only provides a handful of positive and negative examples of what they are looking for. EQUI-VOCAL utilizes these initial examples and additional ones collected through active learning to efficiently synthesize complex user queries. Our approach enables users to find events without database expertise, with limited labeling effort, and without declarative specifications or sketches. Core to EQUI-VOCAL's design is the use of spatio-temporal scene graphs in its data model and query language and a novel query synthesis approach that works on large and noisy video data. Our system outperforms two baseline systems -- in terms of F1 score, synthesis time, and robustness to noise -- and can flexibly synthesize complex queries that the baselines do not support., Comment: This is an extended technical report for the following paper: "Enhao Zhang, Maureen Daum, Dong He, Brandon Haynes, Ranjay Krishna, and Magdalena Balazinska. EQUI-VOCAL: Synthesizing Queries for Compositional Video Events from Limited User Interactions. PVLDB, 16(11): 2714-2727, 2023. doi:10.14778/3611479.3611482"

We study the $Z_{cs}(3985)$ and $Z_{cs}(4000)$ exotic states in the decays of $\Lambda_b$ baryons through a molecular scenario. In the final state interaction, the $\Lambda_b\to \Lambda_c^+ D_s^{(*)-}$ decays are followed by the $\Lambda_c^+ D_s^{(*)-}$ to $Z^-_{cs}p$ rescatterings via exchange of a $D^{(*)}$ meson. We predict a branching fraction of $(3.1^{+1.4}_{-2.6})\times 10^{-4}$ for $\Lambda_b\to Z^-_{cs}p$, which can be measured in the $\Lambda_b\to J/\psi K^{(*)-}p$ decay. This study proposes a new approach to test the molecular model, and guides future experimental searches for the $Z_{cs}(3985)$ and $Z_{cs}(4000)$., Comment: 13 pages, 3 figures

Machine learning tasks over image databases often generate masks that annotate image content (e.g., saliency maps, segmentation maps, depth maps) and enable a variety of applications (e.g., determine if a model is learning spurious correlations or if an image was maliciously modified to mislead a model). While queries that retrieve examples based on mask properties are valuable to practitioners, existing systems do not support them efficiently. In this paper, we formalize the problem and propose MaskSearch, a system that focuses on accelerating queries over databases of image masks while guaranteeing the correctness of query results. MaskSearch leverages a novel indexing technique and an efficient filter-verification query execution framework. Experiments with our prototype show that MaskSearch, using indexes approximately 5% of the compressed data size, accelerates individual queries by up to two orders of magnitude and consistently outperforms existing methods on various multi-query workloads that simulate dataset exploration and analysis processes.

Ferroptosis is associated with the occurrence and development of many diseases, which is the result of an imbalance in cellular metabolism and oxidation−reduction balance. Therefore, it is an effective therapeutic strategy that simultaneously regulating the intracellular oxidation−reduction system. Herein, a click reaction of alkynylamide with thiol groups in the presence of amine or in PBS (pH = 7.4) is developed, which can react efficiently with thiol substances, such as cysteine (Cys), glutathione (GSH), and bovine serum albumin (BSA). Notably, MBTB-PA, an aggregation-induced emission (AIE) photosensitizer with an alkynylamide unit, is synthesized and its intracellular behavior is visualized in situ by fluorescence imaging, demonstrating its excellent ability to target the endoplasmic reticulum. Furthermore, MBTB-PA reacted with proteins in tumor cells, consumed reducing substances, and triggered intracellular oxidative stress, resulting in cell death. Based on this reaction therapy strategy, click reaction is combined with photodynamic therapy to achieve effective killing of tumor cells by simultaneously raising the intracellular oxidative state and reducing the reductive state. This work not only develops an application of click reaction of alkynamide with thiol in bioconjugation and anti-tumor therapy, but also provides feasible ideas for organic reactions in the exploration of organisms. © 2023 Wiley-VCH GmbH.

Continuous multi-axis printing has drawn more and more attention owing to its unique advantages of eliminating the need of support (at least in theory) when printing complex models with overhang features and significantly reducing the stair-step effect of the conventional three-axis printing. For continuous multi-axis printing, the printing layers are typically not planar but curved, so to capitalize the extra DOFs provided by the rotary axes of the printer. When a printing layer is concave (i.e., its curvature vector points inward at the printing point), the potential nozzle-layer local gouging becomes a serious concern, which is especially problematic for metal printing wherein the size of the nozzle head is usually large. Unfortunately, this convexity issue has received little attention in the existing curved slicing algorithms, making them prone to the local gouging quandary. In this paper, we propose a novel curved layer decomposition algorithm based on the use of (only the positive sides of) ellipsoidal surfaces. To facilitate the generation of ellipsoidal printing layers for an arbitrary freeform solid model with a complex topology (e.g., having a non-zero genus number), the powerful algebraic technique of skeletonization is employed by our algorithm to first decompose the model into a suitable set of sub-entities, for each of which a series of ellipsoidal printing layers are then generated, wherein the parameters of ellipsoidal layers are adaptively adjusted to satisfy the necessary constraints such as support-free and other requirements. Preliminary experiments in both computer simulation and physical printing of the proposed algorithm are conducted, and the results give a positive validation on the effectiveness and feasibility of the proposed methodology. © 2021 Elsevier Ltd

The emergence of multi-axis printing systems provides a new solution to print complex parts. However, the current process planning methods fail to balance the support-free requirement, the collision-free condition, as well as the mechanical performance at the same time when printing parts with complicated features. This paper proposes a new curved layer slicing and continuous printing path planning method that considers these factors. The proposed method makes use of two mutually orthogonal unit vector fields embedded on the tetrahedral mesh of the part: the filament orientation vector field and the printing orientation vector field, which indicate the filament orientation and the nozzle orientation at each position during the printing process respectively. A quantitative optimization model is established to compute the two optimal unit vector fields. Then a monotonically increasing scalar distance field is generated by integrating along the optimal printing orientation vector field, and the isosurfaces of the distance field can be used to naturally slice the part into curved layers. Finally, at each isosurface, a continuous printing path is planned along the optimal filament orientation field by interpolating isolines of a surface embedded scalar distance field. Both the simulation and physical experiments were conducted to confirms the effectiveness of the proposed method and its advantage over the existing methods in balance different factors including support-free condition, collision-free requirement, and mechanical performance.

Trochoidal (TR) milling is a popular means for slotting operation. Attributing to its unique circular-shaped path pattern, TR milling avoids the full tool-workpiece engagement, which helps reduce the cutting heat accumulation and hence slow down the tool wear. While traditionally TR milling is only used for machining 2.5D cavities, it has now been extended to machining genuine 3D curved cavities under the realm of five-axis machining. However, since for a typical five-axis machine tool the rotary axes have a much larger moment of inertia than the three linear axes, to reduce both the total machining time and the consumed electric energy (for driving the machine tool), it is desirable to minimize the use of the two rotary axes (particularly the one with the larger moment of inertia) when planning a TR tool path for a given 3D cavity. Nevertheless, due to the newness of five-axis TR machining, there has no published reports on this subject. In this paper, we present a five-axis TR tool path planning algorithm for machining an arbitrary 3D curved cavity, which will consider the kinematical characteristics of the five-axis machine tool and try to minimize the use of the rotary axis with the largest moment of inertia, while tending to all the required constraints such as the threshold on the cutting force. Both computer simulation and physical cutting experiments of the proposed method have been conducted, and the results give a preliminary confirmation on the feasibility and advantages of the proposed method.

Continuous multi-axis printing has drawn more and more attention owing to its unique advantages of eliminating the need of support (at least in theory) when printing complex models with overhang features and significantly reducing the stair-step effect of the conventional three-axis printing. For continuous multi-axis printing, the printing layers are typically not planar but curved, so to capitalize the extra DOFs provided by the rotary axes of the printer. When a printing layer is concave (i.e., its curvature vector points inward at the printing point), the potential nozzle-layer local gouging becomes a serious concern, which is especially problematic for metal printing wherein the size of the nozzle head is usually large. Unfortunately, this convexity issue has received little attention in the existing curved slicing algorithms, making them prone to the local gouging quandary. In this paper, we propose a novel curved layer decomposition algorithm based on the use of (only the positive sides of) ellipsoidal surfaces. To facilitate the generation of ellipsoidal printing layers for an arbitrary freeform solid model with a complex topology (e.g., having a non-zero genus number), the powerful algebraic technique of skeletonization is employed by our algorithm to first decompose the model into a suitable set of sub-entities, for each of which a series of ellipsoidal printing layers are then generated, wherein the parameters of ellipsoidal layers are adaptively adjusted to satisfy the necessary constraints such as support-free and other requirements. Preliminary experiments in both computer simulation and physical printing of the proposed algorithm are conducted, and the results give a positive validation on the effectiveness and feasibility of the proposed methodology. © 2021 Elsevier Ltd

We demonstrate Tensor Query Processor (TQP): a query processor that automatically compiles relational operators into tensor programs. By leveraging tensor runtimes such as PyTorch, TQP is able to: (1) integrate with ML tools (e.g., Pandas for data ingestion, Tensorboard for visualization); (2) target different hardware (e.g., CPU, GPU) and software (e.g., browser) backends; and (3) end-to-end accelerate queries containing both relational and ML operators. TQP is generic enough to support the TPC-H benchmark, and it provides performance that is comparable to, and often better than, that of specialized CPU and GPU query processors.

The huge demand for computation in artificial intelligence (AI) is driving unparalleled investments in hardware and software systems for AI. This leads to an explosion in the number of specialized hardware devices, which are now offered by major cloud vendors. By hiding the low-level complexity through a tensor-based interface, tensor computation runtimes (TCRs) such as PyTorch allow data scientists to efficiently exploit the exciting capabilities offered by the new hardware. In this paper, we explore how database management systems can ride the wave of innovation happening in the AI space. We design, build, and evaluate Tensor Query Processor (TQP): TQP transforms SQL queries into tensor programs and executes them on TCRs. TQP is able to run the full TPC-H benchmark by implementing novel algorithms for relational operators on the tensor routines. At the same time, TQP can support various hardware while only requiring a fraction of the usual development effort. Experiments show that TQP can improve query execution time by up to 10$\times$ over specialized CPU- and GPU-only systems. Finally, TQP can accelerate queries mixing ML predictions and SQL end-to-end, and deliver up to 9$\times$ speedup over CPU baselines.

Pulsar timing arrays (PTAs) have the potential to detect Nanohertz gravitational waves (GWs) that are usually generated by the individual inspiraling supermassive black hole binaries (SMBHBs) in the galactic centers. The GW signals as cosmological standard sirens can provide the absolute cosmic distances, thereby can be used to constrain the cosmological parameters. In this paper, we analyze the ability of future SKA-era PTAs to detect the existing SMBHBs candidates assuming the root mean square of timing noise $\sigma_t=20\ {\rm ns}$, and use the simulated PTA data to constrain the interacting dark energy (IDE) models with energy transfer rate $Q = \beta H\rho_c$. We find that, the future SKA-era PTAs will play an important role in constraining the IDE cosmology. Using only the mock PTA data consisting of 100 pulsars, we obtain $\sigma(H_0)=0.239\ {\rm km} \ {\rm s}^{-1} {\rm Mpc}^{-1}$ and $\sigma(\Omega_m)=0.0103$ in the I$\Lambda$CDM model, which are much better than the results from the Planck TT, TE, EE+lowE. However, the PTA data cannot provide a tight constraint on the coupling parameter $\beta$ compared with Planck, but the data combination of Planck+PTA can provide a rather tight constraint, i.e., $\sigma(\beta)=0.00232$, since the PTA data could break the parameter degeneracies inherent in CMB. In the I$w$CDM model, we obtain $\sigma(\beta)=0.00137$ and $\sigma(w)=0.0492$ from the Planck+PTA data combination. In addition, we find that with the increase of the number of pulsars in PTA, the constraint results from the Planck+PTA will be further improved to some extent. We show that the observations of Nanohertz GWs with future SKA-era PTAs will provide a powerful tool for exploring the nature of dark energy and measuring the coupling between dark energy and dark matter.

Precise control of machining deformation is crucial for improving the manufacturing quality of structural aerospace components. In the machining process, different batches of blanks have different residual stress distributions, which pose a significant challenge to machining deformation control. In this study, a reinforcement learning method for machining deformation control based on a meta-invariant feature space was developed. The proposed method uses a reinforcement-learning model to dynamically control the machining process by monitoring the deformation force. Moreover, combined with a meta-invariant feature space, the proposed method learns the internal relationship of the deformation control approaches under different stress distributions to achieve the machining deformation control of different batches of blanks. Finally, the experimental results show that the proposed method achieves better deformation control than the two existing benchmarking methods. © 2022, The Author(s).

When machining a complex freeform part, using a non-spherical tool could significantly improve the machining efficiency, as one can adaptively adjust the tool posture to maximize its contact area with the part surface. However, since adjusting the tool posture requires changing the tool orientation, a five-axis machine tool is needed, which is extremely expensive as compared to a conventional three-axis machine tool. Moreover, for a complex freeform surface with high curvature variation, to match its curvature change, the tool axis has to drastically change accordingly, thus inducing high velocity and acceleration on the machine tool's rotary axes. To address these issues, in this paper we propose a partition-based 3 + 2-axis strategy for machining a general complex freeform surface with a non-spherical tool. As only a finite small number of distinct tool orientations are needed for 3 + 2-axis machining, an indexed three-axis machine tool suffices, thus relieving the need of an expensive five-axis machine tool. In addition, the much-increased rigidity of the three linear axes of the machine tool will greatly improve the kinematics and dynamics of the machine tool and thus enhance the machining accuracy. Experiments in both computer simulation and physical machining are carried out, whose results confirm that, when compared to using a conventional spherical cutter, by using a non-spherical cutter and adaptively adjusting the contacting tool posture and the feed direction, significant improvement in machining efficiency could be achieved, e.g., more than 50% achieved in our experiments.

The emergence of multi-axis printing systems provides a new solution to print complex parts. However, the current process planning methods fail to balance the support-free requirement, the collision-free condition, as well as the mechanical performance at the same time when printing parts with complicated features. This paper proposes a new curved layer slicing and continuous printing path planning method that considers these factors. The proposed method makes use of two mutually orthogonal unit vector fields embedded on the tetrahedral mesh of the part: the filament orientation vector field and the printing orientation vector field, which indicate the filament orientation and the nozzle orientation at each position during the printing process respectively. A quantitative optimization model is established to compute the two optimal unit vector fields. Then a monotonically increasing scalar distance field is generated by integrating along the optimal printing orientation vector field, and the isosurfaces of the distance field can be used to naturally slice the part into curved layers. Finally, at each isosurface, a continuous printing path is planned along the optimal filament orientation field by interpolating isolines of a surface embedded scalar distance field. Both the simulation and physical experiments were conducted to confirms the effectiveness of the proposed method and its advantage over the existing methods in balance different factors including support-free condition, collision-free requirement, and mechanical performance.

Tooth contact analysis (TCA) is a crucial step in evaluating the working performances of gear designs. According to the meshing process, or the equation of meshing, TCA is usually numerically implemented to find contact points on tooth surfaces with a highly nonlinear model, which contains five independent equations and five variables. It is computationally expensive to solve this model, especially for spatial gears with a complicated tooth geometry such as face gears. To solve this problem, we propose a novel method to simplify the optimization model to two variables. Specifically, the rotation relationship between the face gear and pinion is analyzed in a kinematic view to mathematically eliminate three parameters in the meshing equation. The geometric characteristic of face gear drives is also studied to constrain the parameter range at a reasonable computation cost. Furthermore, this optimization model is solved by a two-stage algorithm with high stability and efficiency, realized by the particle swarm optimization (PSO) algorithm and the Quasi-Newton method, respectively. Examples are given to verify the efficiency and stability of the proposed method. © 2022

Trochoidal (TR) milling is a popular means for slotting operation. Attributing to its unique circular-shaped path pattern, TR milling avoids the full tool-workpiece engagement, which helps reduce the cutting heat accumulation and hence slow down the tool wear. While traditionally TR milling is only used for machining 2.5D cavities, it has now been extended to machining genuine 3D curved cavities under the realm of five-axis machining. However, since for a typical five-axis machine tool the rotary axes have a much larger moment of inertia than the three linear axes, to reduce both the total machining time and the consumed electric energy (for driving the machine tool), it is desirable to minimize the use of the two rotary axes (particularly the one with the larger moment of inertia) when planning a TR tool path for a given 3D cavity. Nevertheless, due to the newness of five-axis TR machining, there has no published reports on this subject. In this paper, we present a five-axis TR tool path planning algorithm for machining an arbitrary 3D curved cavity, which will consider the kinematical characteristics of the five-axis machine tool and try to minimize the use of the rotary axis with the largest moment of inertia, while tending to all the required constraints such as the threshold on the cutting force. Both computer simulation and physical cutting experiments of the proposed method have been conducted, and the results give a preliminary confirmation on the feasibility and advantages of the proposed method.

Continuous multi-axis printing has drawn more and more attention owing to its unique advantages of eliminating the need of support (at least in theory) when printing complex models with overhang features and significantly reducing the stair-step effect of the conventional three-axis printing. For continuous multi-axis printing, the printing layers are typically not planar but curved, so to capitalize the extra DOFs provided by the rotary axes of the printer. When a printing layer is concave (i.e., its curvature vector points inward at the printing point), the potential nozzle-layer local gouging becomes a serious concern, which is especially problematic for metal printing wherein the size of the nozzle head is usually large. Unfortunately, this convexity issue has received little attention in the existing curved slicing algorithms, making them prone to the local gouging quandary. In this paper, we propose a novel curved layer decomposition algorithm based on the use of (only the positive sides of) ellipsoidal surfaces. To facilitate the generation of ellipsoidal printing layers for an arbitrary freeform solid model with a complex topology (e.g., having a non-zero genus number), the powerful algebraic technique of skeletonization is employed by our algorithm to first decompose the model into a suitable set of sub-entities, for each of which a series of ellipsoidal printing layers are then generated, wherein the parameters of ellipsoidal layers are adaptively adjusted to satisfy the necessary constraints such as support-free and other requirements. Preliminary experiments in both computer simulation and physical printing of the proposed algorithm are conducted, and the results give a positive validation on the effectiveness and feasibility of the proposed methodology. © 2021 Elsevier Ltd

We show an improvement in extraction of remotely sensed phenology in the Qinghai-Tibet Plateau (QTP). The improvement includes multiple preprocessing, newly proposed absolute growth rate and relative growth rate models based on botany and phenology, and selection of the appropriate growth rate model at different growing stages. We refer to this model as the tempo-differentially selected growth rate model (TDSGM). The newly developed TDSGM is a comprehensive and accurate remote sensing phenological extraction model without manual intervention, which is better than the current mainstream methods. Results show that: (1) influence of elevation and longitude on the QTP phenology has been more effective than that of latitude and vegetation types; (2) length of the growing season (LOS) was in a shortening trend in the long term, even it showed a slight extension in recent years in the short term; (3) the shortest LOS was identified in the area with an altitude of 3000 to 5500 m, which accounts for 90% of the total area of QTP. High instability of LOS was concentrated in meadows at 4000 to 5000 m; and (4) most previous studies on remotely sensed phenology showed that LOS was more related to the start of the growing season. However, we show that the LOSs of four out of seven QTP vegetation types were closely related to the end of the growth season, consistent with 1455 ground observations.

Multi-axis machining has been widely used in the manufacturing industry and the multi-axis milling process is one typical machining technique owing to its flexibility and accuracy, which could be classified as point milling and flank milling. The process planning of multi-axis milling on complex parts is of great significance to manufacturing efficiency and accuracy. Nevertheless, the complexities of parts such as the non-height field, deep cavity, non-zero genus features, and weak structures of workpieces limit the effectiveness of conventional point or flank milling methods. In this thesis, aiming at these issues, we propose three process planning methods for multi-axis machining of complex parts. First, we present a novel volumetric field slicing method for rough and finish machining on a part with complex features such as deep cavities, non-zero genus as well as weak structures. For the roughing operation of point milling, rather than simply taking Z-level parallel planes as intermediate machining layers, we propose to use curved machining layers, which will eliminate the severe staircase effect on the in-process workpiece as suffered by the Z-level method, and hence significantly stabilize the cutting force and reduce the susceptibility to dynamic problems e.g., chattering. To facilitate the determination of the desired curved machining layers, an elaborate algorithm is presented to construct a geodesic distance field embedded in the volume-to-remove whose iso-surfaces are then naturally used as the machining layers, which are assumed to be smooth and have a uniform thickness. In terms of finishing, aiming at improving the stiffness of the in-process workpiece, and also facilitated by the prescribed geodesic distance field, we employed a new machining strategy of alternating between the roughing and finishing processes. Finally, collision-free tool paths are devised to machine the generated machining layers. Next, we present a quasi-developable approximation-ba

Five-axis flank milling is widely used due to its higher cutting efficiency and better surface finish quality compared with point milling. In recent years, the use of flank milling has been further extended from single-pass milling to multi-pass milling. However, the existing multi-pass flank milling methods focus only on a simple parametric surface and suffer from extraneous machining problems due to the varied cutting width. Moreover, as they ignore the tool's effective cutting length, they are in general incapable of handling other related constraints such as the hard constraints of no-overcut and interference-free. In this paper, we propose a new algorithm of planning multi-pass flank milling for a complex freeform surface mesh. In our method, the design surface mesh is naturally partitioned based on a tangent vector field and each partitioned patch is approximated by a set of piecewise quasi-developable quad strips, on which semi-finishing flank milling tool paths of a given constant cutting width are generated tending to all the required constraints such as no-overcut. The powerful algebraic technique of level-set method is utilized to generate a scalar field on each patch whose iso-lines will serve as the equi-distant cutter contact rulings of the initial quad strips. An elaborated energy-based unified framework is then presented that will optimize the signed quad strips to increase their developability while satisfying all the machining constraints, thus further reducing the flanking machining error with all the hard constraints (e.g., no-overcut) upheld. Ample physical cutting experiments are performed, whose results convincingly confirm the advantages of the proposed method. © 2021 Elsevier Ltd

A critical task in multi-pass process planning for the five-Axis machining of complicated parts is to determine the intermediate surfaces for rough machining. Traditionally, the intermediate surfaces are simply parallel Z-level planes, and the machining is of the simplest three-Axis type. However, for complicated parts, this so-called Z-level method lacks flexibility and causes isolated islands on layers, which require extraneous air movements by the tool. Moreover, the in-process workpiece machined according to the Z-level method suffers from the staircase effect, which often induces unstable dynamic problems on the tool-spindle system. In this paper, we propose a new method of planning a five-Axis machining process for a complicated freeform solid part. In our method, the intermediate surfaces are no longer planar but curved, and they are intrinsically influenced by the convex hull of the part. The powerful algebraic tool of geodesic distance field is utilized to generate the desired intermediate surfaces, for which collision-free five-Axis machining tool paths are then planned. In addition, we propose a novel idea of alternating between the roughing and finishing machining operations, which helps improve the stiffness of the in-process workpiece. Ample physical cutting experiments are performed, and the experimental results convincingly confirm the advantages of our method. © 2021 Georg Thieme Verlag. All rights reserved.

In additive manufacturing, infill structures are commonly used to reduce the weight and cost of a solid part. Currently, most infill structure generation methods are based on the conventional 2.5-axis printing configuration, which, although able to satisfy the self-supporting condition on the infills, suffer from the well-known stair-case effect on the finished surface and the need of extensive support for overhang features. In this paper, based on the emerging continuous multi-axis printing configuration, we present a new lattice infill structure generation algorithm, which is able to achieve the self-supporting condition for both the infills and the boundary surface of the part. The algorithm critically relies on the use of three mutually orthogonal geodesic distance fields that are embedded in the tetrahedral mesh of the solid model. The intersection between the iso-geodesic distance surfaces of these three geodesic distance fields naturally forms the desired lattice of infill structure, while the density of the infills can be conveniently controlled by adjusting the iso-values. The lattice infill pattern in each curved slicing layer is trimmed to conform to an Eulerian graph so to generate a continuous printing path, which can effectively reduce the retractions of the nozzle during the printing process. In addition, to cater to the collision-free requirement and to improve the printing efficiency, we also propose a printing sequence optimization algorithm for determining a collision-free order of printing of the connected lattice infills, which seeks to reduce the air-move length of the nozzle. Ample experiments in both computer simulation and physical printing are performed, and the results give a preliminary confirmation of the advantages of our methodology. © 2020 Elsevier Ltd

In additive manufacturing, infill structures are commonly used to reduce the weight and cost of a solid part. Currently, most infill structure generation methods are based on the conventional 2.5-axis printing configuration, which, although able to satisfy the self-supporting condition on the infills, suffer from the well-known stair-case effect on the finished surface and the need of extensive support for overhang features. In this paper, based on the emerging continuous multi-axis printing configuration, we present a new lattice infill structure generation algorithm, which is able to achieve the self-supporting condition for both the infills and the boundary surface of the part. The algorithm critically relies on the use of three mutually orthogonal geodesic distance fields that are embedded in the tetrahedral mesh of the solid model. The intersection between the iso-geodesic distance surfaces of these three geodesic distance fields naturally forms the desired lattice of infill structure, while the density of the infills can be conveniently controlled by adjusting the iso-values. The lattice infill pattern in each curved slicing layer is trimmed to conform to an Eulerian graph so to generate a continuous printing path, which can effectively reduce the retractions of the nozzle during the printing process. In addition, to cater to the collision-free requirement and to improve the printing efficiency, we also propose a printing sequence optimization algorithm for determining a collision-free order of printing of the connected lattice infills, which seeks to reduce the air-move length of the nozzle. Ample experiments in both computer simulation and physical printing are performed, and the results give a preliminary confirmation of the advantages of our methodology. © 2020 Elsevier Ltd

A critical task in multi-pass process planning for the five-Axis machining of complicated parts is to determine the intermediate surfaces for rough machining. Traditionally, the intermediate surfaces are simply parallel Z-level planes, and the machining is of the simplest three-Axis type. However, for complicated parts, this so-called Z-level method lacks flexibility and causes isolated islands on layers, which require extraneous air movements by the tool. Moreover, the in-process workpiece machined according to the Z-level method suffers from the staircase effect, which often induces unstable dynamic problems on the tool-spindle system. In this paper, we propose a new method of planning a five-Axis machining process for a complicated freeform solid part. In our method, the intermediate surfaces are no longer planar but curved, and they are intrinsically influenced by the convex hull of the part. The powerful algebraic tool of geodesic distance field is utilized to generate the desired intermediate surfaces, for which collision-free five-Axis machining tool paths are then planned. In addition, we propose a novel idea of alternating between the roughing and finishing machining operations, which helps improve the stiffness of the in-process workpiece. Ample physical cutting experiments are performed, and the experimental results convincingly confirm the advantages of our method. © 2021 Georg Thieme Verlag. All rights reserved.

We study gravity wave production and baryogenesis at the electroweak phase transition, in a real singlet scalar extension of the Standard Model, including vector-like top partners to generate the CP violation needed for electroweak baryogenesis (EWBG). The singlet makes the phase transition strongly first-order through its coupling to the Higgs boson, and it spontaneously breaks CP invariance through a dimension-5 contribution to the top quark mass term, generated by integrating out the heavy top quark partners. We improve on previous studies by incorporating updated transport equations, compatible with large bubble wall velocities. The wall speed and thickness are computed directly from the microphysical parameters rather than treating them as free parameters, allowing for a first-principles computation of the baryon asymmetry. The size of the CP-violating dimension-5 operator needed for EWBG is constrained by collider, electroweak precision, and renormalization group running constraints. We identify regions of parameter space that can produce the observed baryon asymmetry or observable gravitational (GW) wave signals. Contrary to standard lore, we find that for strong deflagrations, the efficiencies of large baryon asymmetry production and strong GW signals can be positively correlated. However we find the overall likelihood of observably large GW signals to be smaller than estimated in previous studies. In particular, only detonation-type transitions are predicted to produce observably large gravitational waves., Comment: 29 pages, 10 figures; v2: GW spectrum corrected, conclusions changed, references added

Measurements of time-delay cosmography of lensed quasars can provide an independent probe to explore the expansion history of the late-time Universe. In this paper, we employ the time-delay cosmography measurements from seven lenses (here abbreviated as the TD data) to constrain interacting dark energy (IDE) models. We mainly focus on the scenario of vacuum energy (with $w=-1$) interacting with cold dark matter, and consider four typical cases of the interaction form $Q$. When the TD data alone are employed, we find that the IDE models with $Q\propto \rho_{\rm de}$ seem to have an advantage in relieving the $H_{0}$ tension between the cosmic microwave background (CMB) and TD data. When the TD data are added to the CMB$+$BAO$+$SN$+H_0$ data, we find that: (i) the coupling parameter $\beta$ in all the considered IDE models is positive within 1$\sigma$ range, implying a mild preference for the case of cold dark matter decaying into dark energy; (ii) the IDE model with $Q = \beta H_{0} \rho_{\rm c}$ slightly relieves the $S_8$ tension, but the other considered IDE models further aggravate this tension; (iii) the Akaike information criteria of the IDE models with $Q \propto \rho_{\rm c}$ are lower than that of the $\Lambda$CDM model, indicating that these IDE models are more preferred by the current mainstream data. We conclude that the considered IDE models have their own different advantages when the TD data are employed, and none of them can achieve good scores in all aspects., Comment: 8 pages, 3 figures

Measurements of time-delay cosmography of lensed quasars can provide an independent probe to explore the expansion history of the late-time Universe. In this paper, we employ the time-delay cosmography measurements from seven lenses (here abbreviated as the TD data) to constrain interacting dark energy (IDE) models. We mainly focus on the scenario of vacuum energy (with $w=-1$) interacting with cold dark matter, and consider four typical cases of the interaction form $Q$. When the TD data alone are employed, we find that the IDE models with $Q\propto \rho_{\rm de}$ seem to have an advantage in relieving the $H_{0}$ tension between the cosmic microwave background (CMB) and TD data. When the TD data are added to the CMB$+$BAO$+$SN$+H_0$ data, we find that: (i) the coupling parameter $\beta$ in all the considered IDE models is positive within 1$\sigma$ range, implying a mild preference for the case of cold dark matter decaying into dark energy; (ii) the IDE model with $Q = \beta H_{0} \rho_{\rm c}$ slightly relieves the $S_8$ tension, but the other considered IDE models further aggravate this tension; (iii) the Akaike information criteria of the IDE models with $Q \propto \rho_{\rm c}$ are lower than that of the $\Lambda$CDM model, indicating that these IDE models are more preferred by the current mainstream data. We conclude that the considered IDE models have their own different advantages when the TD data are employed, and none of them can achieve good scores in all aspects., Comment: 8 pages, 3 figures

We study gravity wave production and baryogenesis at the electroweak phase transition, in a real singlet scalar extension of the Standard Model, including vector-like top partners to generate the CP violation needed for electroweak baryogenesis (EWBG). The singlet makes the phase transition strongly first-order through its coupling to the Higgs boson, and it spontaneously breaks CP invariance through a dimension-5 contribution to the top quark mass term, generated by integrating out the heavy top quark partners. We improve on previous studies by incorporating updated transport equations, compatible with large bubble wall velocities. The wall speed and thickness are computed directly from the microphysical parameters rather than treating them as free parameters, allowing for a first-principles computation of the baryon asymmetry. The size of the CP-violating dimension-5 operator needed for EWBG is constrained by collider, electroweak precision, and renormalization group running constraints. We identify regions of parameter space that can produce the observed baryon asymmetry or observable gravitational (GW) wave signals. Contrary to standard lore, we find that for strong deflagrations, the efficiencies of large baryon asymmetry production and strong GW signals can be positively correlated. However we find the overall likelihood of observably large GW signals to be smaller than estimated in previous studies. In particular, only detonation-type transitions are predicted to produce observably large gravitational waves., Comment: 29 pages, 10 figures; v2: GW spectrum corrected, conclusions changed, references added

Five-axis flank milling is widely used due to its higher cutting efficiency and better surface finish quality compared with point milling. In recent years, the use of flank milling has been further extended from single-pass milling to multi-pass milling. However, the existing multi-pass flank milling methods focus only on a simple parametric surface and suffer from extraneous machining problems due to the varied cutting width. Moreover, as they ignore the tool's effective cutting length, they are in general incapable of handling other related constraints such as the hard constraints of no-overcut and interference-free. In this paper, we propose a new algorithm of planning multi-pass flank milling for a complex freeform surface mesh. In our method, the design surface mesh is naturally partitioned based on a tangent vector field and each partitioned patch is approximated by a set of piecewise quasi-developable quad strips, on which semi-finishing flank milling tool paths of a given constant cutting width are generated tending to all the required constraints such as no-overcut. The powerful algebraic technique of level-set method is utilized to generate a scalar field on each patch whose iso-lines will serve as the equi-distant cutter contact rulings of the initial quad strips. An elaborated energy-based unified framework is then presented that will optimize the signed quad strips to increase their developability while satisfying all the machining constraints, thus further reducing the flanking machining error with all the hard constraints (e.g., no-overcut) upheld. Ample physical cutting experiments are performed, whose results convincingly confirm the advantages of the proposed method. © 2021 Elsevier Ltd

A critical task in multi-pass process planning for the five-Axis machining of complicated parts is to determine the intermediate surfaces for rough machining. Traditionally, the intermediate surfaces are simply parallel Z-level planes, and the machining is of the simplest three-Axis type. However, for complicated parts, this so-called Z-level method lacks flexibility and causes isolated islands on layers, which require extraneous air movements by the tool. Moreover, the in-process workpiece machined according to the Z-level method suffers from the staircase effect, which often induces unstable dynamic problems on the tool-spindle system. In this paper, we propose a new method of planning a five-Axis machining process for a complicated freeform solid part. In our method, the intermediate surfaces are no longer planar but curved, and they are intrinsically influenced by the convex hull of the part. The powerful algebraic tool of geodesic distance field is utilized to generate the desired intermediate surfaces, for which collision-free five-Axis machining tool paths are then planned. In addition, we propose a novel idea of alternating between the roughing and finishing machining operations, which helps improve the stiffness of the in-process workpiece. Ample physical cutting experiments are performed, and the experimental results convincingly confirm the advantages of our method. © 2021 Georg Thieme Verlag. All rights reserved.

In additive manufacturing, infill structures are commonly used to reduce the weight and cost of a solid part. Currently, most infill structure generation methods are based on the conventional 2.5-axis printing configuration, which, although able to satisfy the self-supporting condition on the infills, suffer from the well-known stair-case effect on the finished surface and the need of extensive support for overhang features. In this paper, based on the emerging continuous multi-axis printing configuration, we present a new lattice infill structure generation algorithm, which is able to achieve the self-supporting condition for both the infills and the boundary surface of the part. The algorithm critically relies on the use of three mutually orthogonal geodesic distance fields that are embedded in the tetrahedral mesh of the solid model. The intersection between the iso-geodesic distance surfaces of these three geodesic distance fields naturally forms the desired lattice of infill structure, while the density of the infills can be conveniently controlled by adjusting the iso-values. The lattice infill pattern in each curved slicing layer is trimmed to conform to an Eulerian graph so to generate a continuous printing path, which can effectively reduce the retractions of the nozzle during the printing process. In addition, to cater to the collision-free requirement and to improve the printing efficiency, we also propose a printing sequence optimization algorithm for determining a collision-free order of printing of the connected lattice infills, which seeks to reduce the air-move length of the nozzle. Ample experiments in both computer simulation and physical printing are performed, and the results give a preliminary confirmation of the advantages of our methodology. © 2020 Elsevier Ltd

Background: As plant secondary metabolites, polyphenols have gained more attention with increasing market demand due to their potential benefit for health. However, the low uptake rate and target delivery efficiency of polyphenols towards target sites such as organs, tissues and cells limit their applications. Polyphenols possess binding affinities with proteins via non-covalent and (or) covalent interactions, which provide a strategy for engineering as polyphenol-protein complexes to protect them from oxidation and enzymatic hydrolysis during gastrointestinal digestion. Polyphenol engineering via polyphenol-protein interaction changes the physical and chemical characteristics of polyphenols, thereby protecting polyphenols from oxidation and enzymatic hydrolysis during gastrointestinal digestion and improving their uptake rate, target specific delivery and biological activity. Scope and approach: This review aims to describe the mechanisms underlying engineering polyphenols via polyphenol-protein interaction, as well as their effect on the antioxidative, anti-inflammatory and anti-cancer activities of polyphenols. Especially, it focuses on polyphenols functional enhancement for improved intestinal health by engineering with proteins. Key finding and conclusions: Polyphenol stability in gastrointestinal tract, uptake, target specific delivery, bioavailability and biological activity can be enhanced by engineering with proteins via polyphenol-protein interactions. The potential applications of the engineering polyphenol-protein complex for health benefit are specifically addressed, but their safety needs to be assessed carefully before developing as functional food ingredients.

We design, implement, and evaluate DeepEverest, a system for the efficient execution of interpretation by example queries over the activation values of a deep neural network. DeepEverest consists of an efficient indexing technique and a query execution algorithm with various optimizations. We prove that the proposed query execution algorithm is instance optimal. Experiments with our prototype show that DeepEverest, using less than 20% of the storage of full materialization, significantly accelerates individual queries by up to 63x and consistently outperforms other methods on multi-query workloads that simulate DNN interpretation processes., Comment: This is an extended technical report for the following paper: "DeepEverest: Accelerating Declarative Top-K Queries for Deep Neural Network Interpretation. PVLDB, 15(1): 98 - 111, 2021. doi:10.14778/3485450.3485460"

Ground segmentation of point clouds remains challenging because of the sparse and unordered data structure. This paper proposes the GSECnet - Ground Segmentation network for Edge Computing, an efficient ground segmentation framework of point clouds specifically designed to be deployable on a low-power edge computing unit. First, raw point clouds are converted into a discretization representation by pillarization. Afterward, features of points within pillars are fed into PointNet to get the corresponding pillars feature map. Then, a depthwise-separable U-Net with the attention module learns the classification from the pillars feature map with an enormously diminished model parameter size. Our proposed framework is evaluated on SemanticKITTI against both point-based and discretization-based state-of-the-art learning approaches, and achieves an excellent balance between high accuracy and low computing complexity. Remarkably, our framework achieves the inference runtime of 135.2 Hz on a desktop platform. Moreover, experiments verify that it is deployable on a low-power edge computing unit powered 10 watts only., Comment: 6 pages, 5 figures, code is available at https://github.com/SAMMiCA/GSECnet

We present a new video storage system (VSS) designed to decouple high-level video operations from the low-level details required to store and efficiently retrieve video data. VSS is designed to be the storage subsystem of a video data management system (VDBMS) and is responsible for: (1) transparently and automatically arranging the data on disk in an efficient, granular format; (2) caching frequently-retrieved regions in the most useful formats; and (3) eliminating redundancies found in videos captured from multiple cameras with overlapping fields of view. Our results suggest that VSS can improve VDBMS read performance by up to 54%, reduce storage costs by up to 45%, and enable developers to focus on application logic rather than video storage and retrieval.

Supramolecular assembly and PEGylation (attachment of a polyethylene glycol polymer chain) of peptides can be an effective strategy to develop antimicrobial peptides with increased stability, antimicrobial efficacy and hemocompatibility. However, how the self-assembly properties and PEGylation affect their lipid membrane interaction is still an unanswered question. In this work, we use state-of-the-art small angle X-ray and neutron scattering (SAXS/SANS) together with neutron reflectometry (NR) to study the membrane interaction of a series of multidomain peptides, with and without PEGylation, known to self-assemble into nanofibers. Our approach allows us to study both how the structure of the peptide and the membrane are affected by the peptide–lipid interactions. When comparing self-assembled peptides with monomeric peptides that are not able to undergo assembly due to shorter chain length, we found that the nanofibers interact more strongly with the membrane. They were found to insert into the core of the membrane as well as to absorb as intact fibres on the surface. Based on the presented results, PEGylation of the multidomain peptides leads to a slight net decrease in the membrane interaction, while the distribution of the peptide at the interface is similar to the non-PEGylated peptides. Based on the structural information, we showed that nanofibers were partially disrupted upon interaction with phospholipid membranes. This is in contrast with the considerable physical stability of the peptide in solution, which is desirable for an extended in vivo circulation time.