Mutual Constraint Research Articles

Mechanical Properties Optimization and Microstructure Analysis of Pure Copper Heterostructured Laminates via Rolling

Heterogeneous metallic structures constitute a novel class of materials with excellent mechanical properties. However, the existing process for obtaining heterostructures from a single material does not meet large-scale industrial requirements. In this study, a pure copper heterostructured laminate (HSL) composed of a surface elongated-grain layer and a central equiaxed-grain layer was fabricated by rolling bonding and annealing. To study the effect of the interface on the mechanical properties of gradient-structured materials, both laminate metal composite (LMC) and non-composite laminate (NCL) were fabricated by cold-rolling pretreatment of the center layer (60% reduction) and cold-rolling bonding of the whole blank (67% reduction). Then, the HSL was obtained by controlling the post-annealing regimes, the microstructure of each layer was optimized, and a larger degree of microstructural heterogeneities, such as grain size, misorientation angle, and grain orientation, was obtained, which resulted in obvious mechanical differences. Tensile tests of the HSL, surface layer, center layer, and NCL specimens revealed that the HSL annealed at 300 °C for 1 h had a significantly higher strength than the center layer and a higher elongation than the surface layer. The HSL had a tensile strength and elongation at fracture of 278.08 MPa and 46.2%, respectively, indicating a good balance of strength and plasticity. The improved properties were primarily attributed to the strengthening or strain hardening due to the inhomogeneous deformation of the heterogeneous layers in the laminate and the mutual constraint acquired by the distinct layers with strong mechanical differences. The HSL had an interfacial bonding strength of 178.5 MPa, which played a vital role in the coordinated deformation of the heterogeneous layers. This study proposes an HSL design method that effectively simplifies the process of obtaining heterostructures in homogeneous materials by controlling the cumulative deformation of the surface and center layers.

Learning personalized drug features and differentiated drug-pair interaction information for drug–drug interaction prediction

Multi-drug combination therapies are increasingly used for complex diseases but carry risks of harmful drug interactions. Effective drug–drug interaction prediction (DDIP) is essential for assessing risks among numerous drug pairs. Most DDIP methods involve two main steps: drug representation and drug pair interaction extraction, respectively challenged by the loss of personalized drug information and the need for differentiated interaction data, and are rarely studied. Specifically, personalized drug information refers to the distinct features of each drug. These properties can be easily confused by neighboring information during graph propagation. This issue is especially prominent in drug interaction graphs with long-tail distributions, which poses challenges for personalized drug information learning. Furthermore, it is crucial to learn interactions with differentiation in order to identify diverse drug relationships. Some methods simply concatenate drug features, often ignoring the differences of different drug relationships, while other methods based on substructures rely on professional pharmacological knowledge and are computationally complex. To address these issues, we propose a novel method, learning personalized Drug Features and differentiated Drug-Pair interaction information for drug–drug interaction prediction (DFPDDI). This approach employs a contrastive learning network with edge-aware augmentations and mutual information estimators to capture personalized drug features across various graph distributions. Furthermore, it applies a mutual information constraint to drug-pair representations, enhancing the accuracy of interaction predictions by better distinguishing between different types of drug relationships. The results evaluated on three public datasets demonstrate competitive performance compared to baselines. It also shows potential for accurate predictions, particularly in imbalanced-distribution graphs.

Research on the interaction between tourism management and rural cultural inheritance and protection

Taking Funing County, Wenshan Zhuang and Miao Autonomous Prefecture, Yunnan Province as an example, this paper studies the interaction between tourism management and rural cultural heritage and protection. Firstly, the importance of tourism management and the protection of rural cultural heritage was introduced, and the significance of studying the interaction in the context of Funing County was expounded. Secondly, the role of tourism management in the inheritance and protection of rural culture was discussed, including the characteristics of the discipline and the goal of talent training. Then, the application of rural cultural heritage protection in tourism management is analyzed, including tourism planning and development, tourism marketing, tourism human resource management, etc. Then, the interactive relationship between tourism management and the inheritance and protection of rural culture was deeply analyzed, including mutual promotion and mutual constraint. Through empirical case analysis, this paper elaborates the current situation and existing problems of tourism management and rural cultural heritage protection in Funing County, and puts forward corresponding countermeasures and suggestions, including policies and regulations, talent training and market development. Finally, the research results are summarized, the shortcomings of the research are pointed out, and the direction of future research is proposed.

Explicit Knowledge Graph Reasoning for Conversational Recommendation

Traditional recommender systems estimate user preference on items purely based on historical interaction records, thus failing to capture fine-grained yet dynamic user interests and letting users receive recommendation only passively. Recent conversational recommender systems (CRSs) tackle those limitations by enabling recommender systems to interact with the user to obtain her/his current preference through a sequence of clarifying questions. Recently, there has been a rise of using knowledge graphs (KGs) for CRSs, where the core motivation is to incorporate the abundant side information carried by a KG into both the recommendation and conversation processes. However, existing KG-based CRSs are subject to two defects: (1) there is a semantic gap between the learned representations of utterances and KG entities, hindering the retrieval of relevant KG information; (2) the reasoning over KG is mostly performed with the implicitly learned user interests, overlooking the explicit signals from the entities actually mentioned in the conversation. To address these drawbacks, we propose a new CRS framework, namely, the Knowledge Enhanced Conversational Reasoning (KECR) model. As a user can reflect her/his preferences via both attribute- and item-level expressions, KECR jointly embeds the structured knowledge from two levels in the KG. A mutual information maximization constraint is further proposed for semantic alignment between the embedding spaces of utterances and KG entities. Meanwhile, KECR utilizes the connectivity within the KG to conduct explicit reasoning of the user demand, making the model less dependent on the user’s feedback to clarifying questions. As such, the semantic alignment and explicit KG reasoning can jointly facilitate accurate recommendation and quality dialogue generation. By comparing with strong baselines on two real-world datasets, we demonstrate that KECR obtains state-of-the-art recommendation effectiveness, as well as competitive dialogue generation performance.

Mn‐doped Pb(In1/2Nb1/2)O3‐Pb(Mg1/3Nb2/3)O3‐PbTiO3 relaxor ferroelectric ceramics for high‐power piezoelectric applications

AbstractAchieving comprehensive improvement in both piezoelectric coefficient (d33) and mechanical quality factor (Qm) has become a key issue in the research of high‐power piezoelectric ceramics. Here, we focused on Mn‐doped Pb(In1/2Nb1/2)O3‐Pb(Sc1/2Nb1/2)O3‐PbTiO3 (PIN‐PSN‐PT) relaxor ferroelectric ceramics, where Mn‐doping is employed to enhance Qm. The effects of doping concentration and PT content on phase structure, ferroelectric, dielectric, and electromechanical properties were systematically studied. The well radius matching between Mn ions and B‐site ions facilitates the ion substitution and significantly increases the concentration of oxygen vacancies, thereby enhancing the domain wall pinning effect. A significantly weakened mutual constraint relationship between d33 and Qm is observed in the 2 mol% Mn‐doped PIN‐PSN‐PT ceramics with tetragonal phase, which exhibits Qm exceeding 3000 while maintaining d33 higher than 320 pC/N and Curie temperature higher than 300°C, leading to the vibration velocity over 1.15 m/s. Rayleigh analysis reveals the higher intrinsic piezoelectric response in high PT components compared to that in low PT components, which was thought to be the origin for the weakened mutual constraint relationship between d33 and Qm in tetragonal Mn‐doped PIN‐PSN‐PT ceramics.

A novel fractional-order memristive Hopfield neural network for traveling salesman problem and its FPGA implementation

This paper proposes a novel fractional-order memristive Hopfield neural network (HNN) to address traveling salesman problem (TSP). Fractional-order memristive HNN can efficiently converge to a globally optimal solution, while conventional HNN tends to become stuck at a local minimum in solving TSP. Incorporating fractional-order calculus and memristors gives the system long-term memory properties and complex chaotic characteristics, resulting in faster convergence speeds and shorter average distances in solving TSP. Moreover, a novel chaotic optimization algorithm based on fractional-order memristive HNN is designed for the calculation process to deal with mutual constraint between convergence accuracy and convergence speed, which circumvents random search and diminishes the rate of invalid solutions. Numerical simulations demonstrate the effectiveness and merits of the proposed algorithm. Furthermore, Field Programmable Gate Array (FPGA) technology is utilized to implement the proposed neural network.

All-composite honeycomb-enhanced corrugated hybrid structures to improve flexural responses

An all-composite corrugated hybrid structure was proposed. The Nomex honeycomb was filled to enhance carbon fiber reinforced polymer (CFRP) corrugated sandwich against mono-corrugation flexural collapse. Three-point bending behaviors of the honeycomb enhanced corrugated sandwich (HECS), taking account of the effects of honeycomb density, loading position, and honeycomb proportion, were investigated. The bearing capacity of the HECS is greatly increased. The mutual constraint between the web and honeycomb exhibits different damage modes from those of corrugated and honeycomb sandwich. The peak load and elastic stiffness of the HECS are 1.37 times and 1.4 times higher than that of unenhanced corrugated sandwich, respectively. Numerical and analytical models were able to consistently predict the deflection and stiffness of the HECS.

A probabilistic framework to evaluate seismic resilience of substations based on three-stage uncertainty

Electrical substation systems exhibit vulnerability after an earthquake. It results in significant economic losses because of the uncertainty of the functional state and recovery process. To quantitatively assess the uncertainty of substations, this paper proposes a probability-based seismic resilience assessment framework. This framework considers uncertainty in three stages: functional state, functional recovery, and resource constraints. Based on the functional characteristics of substation systems, a directed acyclic graph was constructed, and the Monte Carlo algorithm was used to obtain the functional state matrix, thereby establishing a probability-based functional state model. Moreover, a dynamic post-earthquake functional recovery analysis framework was built based on recovery efficiency metrics, and functional recovery paths were obtained through iterative simulations. A stepped functional recovery curve was developed based on resource constraints. Three-stage uncertainty parameters were transformed into resilience features for quantitative assessment. By analysing seismic resilience of a typical 220 kV step-down substation, probability distributions of functional recovery curves and confidence intervals for seismic resilience metrics were obtained. Notably, mutual constraint relationships among resource conditions were identified.

Disruption of polar order in lead zirconate titanate by composition-modulated artificial superlattice

Lead zirconate titanate (Pb (Zrx, Ti1−x)O3: PZT) is a well-known ferroelectric compound, in which long-range polar order is usually developed. In the present study, it was clarified by distortion-corrected atomic-scale scanning transmission electron microscopy imaging that long-range polar order is disrupted in PZT by utilizing composition-modulated superlattice. Shape of unit cell was unusual both in the Pb(Zr0.65Ti0.35)O3 (PZT65) and Pb(Zr0.30Ti0.70)O3 (PZT30) layers, which was due to mutual in-plane lattice constraint. By taking account of this, first-principles calculations clarified that multiple directions can be energetically favorable for lead-ion displacement, which explains a reason why long-range polar order was disrupted.

Algebraic Control Strategy for Solving the Mutual Exclusion Constraint Problem in a Network of Timed Event Graphs With Disturbance Inputs: Application to an Assembly Process

Algebraic Control Strategy for Solving the Mutual Exclusion Constraint Problem in a Network of Timed Event Graphs With Disturbance Inputs: Application to an Assembly Process

Re-decoupling the classification branch in object detectors for few-class scenes

Few-class object detection is a critical task in numerous scenes, such as autonomous driving and intelligent surveillance. The current researches mainly focus on the correlation or decoupling between classification and regression subtasks in object detection. However, they rarely take advantage of the potential of re-decoupling the classification subtask. In this paper, we propose to re-decouple the classification branch in object detection for few-class scenes by reducing multi-classification features to multiple binary-classification features. Since multi-classification losses cannot supervise the network to learn decoupled binary-classification features, we introduce a single-class loss to supervise decoupled multiple binary-classification branches. In particular, we propose a basic feature degradation head (FD-Head) structure that decouples the classification branches and applies binary-classification loss to encourage each branch to learn only the degraded single-class features. In addition, based on the mutual exclusion between classes, we propose a mutual exclusion constraint (FD-Head-M) module to constrain the scores of all classes, promoting the detector performance. Finally, we replace the original convolution with more powerful feature extraction modules to form the enhanced FD-Head (FD-Head-E). Notably, our method can be used as a universal module and embedded into the existing object detectors to boost their performance. When applying our method to typical object detectors, it experimentally achieves performance gains of 1.2–2.2%, 1.7–2.5% on the KITTI-3, SeaShips datasets respectively. When using ResNet50 as the backbone network, our method gains an accuracy of 45.9% on the MS COCO dataset.

M 3 Rec: A Context-aware Offline Meta-level Model-based Reinforcement Learning Approach for Cold-Start Recommendation

Reinforcement learning (RL) has shown great promise in optimizing long-term user interest in recommender systems. However, existing RL-based recommendation methods need a large number of interactions for each user to learn the recommendation policy. The challenge becomes more critical when recommending to new users who have a limited number of interactions. To that end, in this paper, we address the cold-start challenge in the RL-based recommender systems by proposing a novel context-aware offline meta-level model-based reinforcement learning approach for user adaptation. Our proposed approach learns to infer each user's preference with a user context variable that enables recommendation systems to better adapt to new users with limited contextual information. To improve adaptation efficiency, our approach learns to recover the user choice function and reward from limited contextual information through an inverse reinforcement learning method, which is used to assist the training of a meta-level recommendation agent. To avoid the need for online interaction, the proposed method is trained using historically collected offline data. Moreover, to tackle the challenge of offline policy training, we introduce a mutual information constraint between the user model and recommendation agent. Evaluation results show the superiority of our developed offline policy learning method when adapting to new users with limited contextual information. In addition, we provide a theoretical analysis of the recommendation performance bound.

Design and Implementation of a Four-Unit Array Piezoelectric Bionic MEMS Vector Hydrophone.

High-performance vector hydrophones have been gaining attention for underwater target-monitoring applications. Nevertheless, there exists the mutual constraint between sensitivity and bandwidth of a single hydrophone. To solve this problem, a four-unit array piezoelectric bionic MEMS vector hydrophone (FPVH) was developed in this paper, which has a cross-beam and a bionic fish-lateral-line-nerve-cell-cilia unit array structure. Simulation analysis and optimization in the design of the bionic microstructure have been performed by COMSOL 6.1 software to determine the structure dimensions and the lead zirconate titanate (PZT) thin film distribution. The FPVH was manufactured using MEMS technology and tested in a standing wave bucket. The results indicate that the FPVH has a sensitivity of up to -167.93 dB@1000 Hz (0 dB = 1 V/μPa), which is 12 dB higher than that of the one-unit piezoelectric MEMS vector hydrophone (OPVH). Additionally, the working bandwidth of the FPVH reaches 20 Hz~1200 Hz, exhibiting a good cosine curve with an 8-shape. This work paves a new way for the development of multi-unit piezoelectric vector hydrophones for underwater acoustic detectors.

Breaking the Mutual Constraint between Polarization and Voltage Resistance with Nanograined High-Entropy Ceramic.

Dielectric ceramics with a high energy storage capacity are key to advanced pulsed power capacitors. However, conventional materials face a mutual constraint between polarization strength and the breakdown strength bottleneck. To address this limitation, the concept of nanograined high-entropy ceramics is introduced in this work. By introducing a large number of constituent elements into the A-site of perovskite material lattice, high-entropy (Bi0.2K0.2Ba0.2Sr0.2Ca0.2)TiO3-0.2 'CuO relaxor ceramic with nanoscale grains have been successfully prepared, which breaks the mutual constraint between polarization strength and breakdown strength bottleneck and results a recoverable energy density (Wrec ∼ 6.86 J/cm3) and an efficiency (η ∼ 87.7%) at 670 kV/cm. Moreover, its excellent stability makes it potentially useful under a variety of extreme conditions, including at high temperatures and high/low frequencies. These obtained results demonstrate that this nanograined high-entropy lead-free perovskite ceramic has great potential for energy storage applications.

Mixed noise-guided mutual constraint framework for unsupervised anomaly detection in smart industries

Large-scale sensor and data acquisition systems, integrated with deep learning methodologies, play a pivotal role in enhancing the sustainability and security of smart city environments, exemplifying the critical significance of anomaly detection techniques. Anomaly detection in complex industrial scenarios presents various challenges, such as intricate working environments, limited anomaly samples, and lack of a priori information. Unsupervised anomaly detection based on knowledge distillation enables anomaly detection using only normal samples. However, the similarity in structure between teacher and student models, along with identical input data flow, hampers accurate anomaly detection and localization. To address these issues, we propose MNMC, an unsupervised anomaly detection model consisting of a mixed noise generation module emulating real defects, a mutual constraint module, and an anomaly segmentation module. Firstly, to enhance the student network’s ability to learn robust features, we construct a hybrid noise model comprising dead-leaves noise and perlin noise. This generates features with structural texture and distributional characteristics closer to real anomalies. Secondly, we design a mutual constraint framework to further improve the learning ability of the student network for normal features by constraining representations containing only a single noise. Lastly, for the detection of anomalies at different scales, we propose a new evaluation metric based on equal importance of normal and anomalous regions. Through ablation experiments, we demonstrate the effectiveness of the simulated real defect generation module and the mutual constraints module. Performance experiments on the MVTec dataset show that our method achieves competitive results compared to the current state-of-the-art anomaly detection methods.

Wide-swath and high-resolution whisk-broom imaging and on-orbit performance of SDGSAT-1 thermal infrared spectrometer

Wide-swath and high-resolution Thermal infrared (TIR) data are limited by the mutual constraint between swath and resolution. In this paper, we propose a whisk-broom imaging method with the long-linear-array (LLA) 4-stage TDI detector (a total of 2048 pixels along splicing direction) and high-precision one-dimensional scanning mirror to achieve 30 m resolution and 300 km swath. And it is being implemented in the first SDGs satellite (SATSAT-1) thermal infrared spectrometer (TIS). Specially, the integration time of whisk-broom imaging system is limited by the speed of scanning mirror, and the shorter integration time is a huge challenge to meet the requirement of high sensitivity and dynamic range. The HgCdTe detector has been improved in its annealing process to reduce dark current and noise, and the opto-mechanical system has been regulated at 195 K to reduce instrument self-emission. To address the limitations of small satellite resources, TIS is equipped with a highly integrated and lightweight design, weighing 247.5 kg and consuming 248 watts respectively. As a result of the complexity of imaging mechanisms and the low frequency of deep space observations, positioning error and radiometric calibration resources have increased, and geometric and radiometric calibration methods have been improved. On-orbit test results show that TIS is capable of detecting temperature differences with a NEdT (noise equivalent differential temperature) less than 0.08 K@300 K, an absolute radiometric accuracy of less than 1 K @ 300 K, and MTF of greater than 0.125. Presently, a collection of more than 201,856 scenes covering land and marine and some precious TIR datasets of planes and naval vessels have been obtained and provided for United Nations for the realization of SDGs goals.

Superior strength-ductility combination resulted from hetero-zone boundary affected region in Cu-Fe layered material

The hetero-zone boundary affected region (HBAR) significantly influences the mechanical behaviors of layered materials, where the deformation mechanisms differ from those in the bulk layers. In this study, three kinds of heterogeneous Cu-Fe layered materials with different interface spacing but identical total thicknesses were prepared. The effects of HBAR and strain partitioning on the tensile behavior of the layered materials were investigated. The results showed that layered materials had enhanced yield strength and uniform elongation with decreasing interface spacing. During tensile deformation, geometrically necessary dislocations (GNDs) were generated at hetero-zone boundaries and piled up near them, resulting in hetero-deformation induced (HDI) strengthening and HDI work hardening. Surface profilometry measurements showed that the Cu and Fe layers exhibited obvious strain partitioning and mutual constraint. With decreasing interface spacing, strain partitioning is enhanced by interlayer constraint, which prevented strain localization at interfaces and thus improved the synergetic deformation of layers. A higher fraction of HBAR can improve the mechanical performance of heterogeneous layered materials. This study deepens our understanding of the relationship between HBAR and strength-ductility synergy and provides some insight into the design of layered materials.

Effects of ion polarizability and oxygen vacancy on microwave dielectric properties of fluorite‐structured Ce1−xCaxO2−x

AbstractX‐ray diffractometer with Rietveld refinement and Raman spectroscopy were used for phase analysis. Ca doping results in decreased εr (24.09–21.52), Q × f (68 914–40 110 GHz), and τf (−50.0 ppm/°C to −60.2 ppm/°C) all decrease, which obviously does not conform to the mutual constraint relationship among the three parameters of microwave dielectric properties. The εr of Ce1−xCaxO2−x ceramics is affected by the ion polarizability and the Ce rattling and valence states, among which the rattling of Ce cations plays a dominant role. The presence of oxygen vacancies in Ce1−xCaxO2−x ceramics explains the decrease in Q × f. The τε, εr, and αL are responsible for the decrease in τf, especially for the ταm‐3αL values.

A cell image segmentation method based on edge feature residual fusion.

In recent years, cancer has seriously damaged human health, and the morphological structure of cells serves as an important basis for cancer diagnosis and grading. Automatic cell segmentation based on deep learning has become an important means of computer-aided pathological diagnosis. Aiming at the existing problems of rough segmentation boundaries and inaccurate segmentation in cell image segmentation, this paper designs a cell image segmentation network model (ERF-TransUNet) based on edge feature residual fusion from the perspective of mutual complementarity and constraint between edge features and object features. The model uses a hybrid architecture of CNN and Transformer to extract multi-scale features from cell images, and adds independent edge feature extraction modules and residual fusion modules to enhance the extraction of edge features and their constraints when fusing with cell object features, improving the accuracy of cell contour positioning. Through experiments on two gland cell datasets, CRAG and Glas, and comparing the segmentation effects with current popular deep learning models, the network model proposed in this paper has achieved good performance in both Dice coefficient and Hausdorff distance, which can effectively improve the segmentation effect of cell images.

Information bottleneck disentanglement based sparse representation for fair classification

Unlike current state-of-the-art methods based on data augmentation and adversarial frameworks to solve the fair classification problem, this paper proposes the Information Bottleneck Disentanglement Based Sparse Representation algorithm to ensure the fairness of extracted features. The algorithm assumes that each sample is collaboratively generated by a categorical latent variable and a sensitive or category-irrelevant latent variable. To ensure that these two latent variables correctly capture their corresponding features, we impose Information Bottleneck Disentanglement constraints on them. Specifically, the Information Bottleneck is used to obtain more relevant information while compressing the amount of information. Simultaneously, a mutual information minimization constraint enhances their independence. Additionally, by employing the sparse structure, we can further disentangle the discriminative and sensitive information under supervision guidance. Therefore, the algorithm can improve the discrimination of the categorical latent variable as much as possible without extracting the sensitive data into it. Extensive experiments on five public and challenging benchmark datasets reveal the effectiveness of our proposed method, demonstrating that it can ensure fairness without sacrificing classification performance.