东北大学秦皇岛分校

Temporal Knowledge Graph Entity Alignment (TKGEA) is vital for knowledge fusion. Current embedding-based methods rely on numerical time encoding, which fails to capture periodic temporal patterns. Additionally, the high heterogeneity of temporal knowledge graphs (such as unbalanced structures and sparse temporal evidence) makes embedding similarity scores unreliable when facing conflicting signals. Simultaneously, recent Large Language Model (LLM) approaches often ignore temporal pattern modeling and lack structured constraints from embedding models, leading to unstable inferences. Consequently, both existing paradigms struggle with complex temporal dynamics and structural heterogeneity in TKGEA. To solve the above problems, this paper proposes a Pattern-Aware Gating mechanism for Entity Alignment, denoted as PAGEA. The method learns a robust entity representation with structural, temporal and semantic decoupling in a heterogeneous temporal knowledge graph based on periodic temporal coding and adaptive pattern-aware gating mechanism. On this basis, PAGEA introduces a reasoning stage based on pattern evidence constraints, and selectively uses large language models for semantic reasoning for low-confidence entity pairs. By transforming the structured temporal versus relational evidence extracted during the pattern-aware alignment process into interpretable text, the model is able to obtain more reliable alignment results in the constrained inference space. Experimental results on multiple heterogeneous time and cross-language entity alignment benchmark datasets show that PAGEA outperforms existing SOTA methods with an average Hits@1 improvement of 2.29%, verifying the effectiveness of the model.

Differential Evolution (DE) has established itself as a leading population-based stochastic optimization technique, widely acclaimed for its conceptual simplicity and effectiveness in handling continuous real-parameter problems. Notwithstanding its broad applicability, the canonical DE framework and many refined variants continue to grapple with two longstanding issues: premature convergence and erosion of population diversity, which become particularly pronounced in complex, multi-modal optimization scenarios. These shortcomings often originate from a suboptimal trade-off between exploratory and exploitative behaviors, ineffective mutation operators during stagnant phases, and diversity preservation approaches that tend to disrupt promising search trajectories. To mitigate these limitations, we present the Dynamic Hybrid Adaptive Differential Evolution (DMADE) algorithm, which introduces three principal innovations: first, a dual-phase parameter adaptation mechanism that employs Gaussian-inspired base components and adaptive perturbations to dynamically regulate exploration-exploitation balance; second, a centroid-driven mutation strategy that utilizes population distribution features to rejuvenate trapped solutions; and third, a diversity enhancement technique grounded in potential energy theory, incorporating dynamic interaction sensing and gradient-based relocation. Empirical studies conducted on the CEC2014, CEC2017, and CEC2022 test suites indicate that DMADE consistently outperforms state-of-the-art DE algorithms in terms of solution precision, convergence rate, and algorithmic stability. The method’s robustness is further substantiated through extensive statistical testing and component-wise ablation experiments, affirming its efficacy in overcoming core challenges prevalent in contemporary evolutionary optimization.

High-strength low-alloy (HSLA) strip combines high strength with excellent formability, yet its constraints make cold rolling more challenging, leading to frequent shape defects during production. To investigate and control the causes of such defects, a multistand 3D continuous rolling simulation model was developed based on industrial data, integrating the widthwise property variations of HSLA thin strip with a data genetic mechanism. The relative errors of the calculated rolling force and center thickness for each rolling pass model are controlled within +/- 7% and +/- 0.15%, respectively. Based on this model, the differentiated regulation mechanisms of work roll bending (WRB) and intermediate roll bending (IRB) on the shape evolution of HSLA strip were elucidated. WRB regulation shows clear stage-dependent behavior: weak in early stands with relative thickness and length differences within +/- 2.5, nonlinear in intermediate stands driven by genetic inheritance effects, and saturating in final stands. IRB improves flatness in intermediate stands by enhancing center reduction and reducing edge-to-center differences but has a weaker overall effect. WRB directly affects work roll deflection and exit strip crown, while IRB may cause overcompensation in the final stand. Overall, the low deviation flatness zones for all forest stands are concentrated within the range of approximate to 100-250 kN for WRB and 100-300 kN for IRB.

The high energy consumption and spatiotemporal thermal asymmetry of data center cooling systems have become critical bottlenecks constraining their green and sustainable development. Traditional point-type temperature sensors suffer from insufficient spatial coverage, while conventional feedback control strategies exhibit delayed responses and limited adaptability under dynamic workloads. To address these challenges, this study proposes a real-time thermal symmetry management framework for data centers based on distributed fiber optic temperature sensing and model predictive control (MPC). The proposed system employs Brillouin scattering-based distributed sensing to continuously acquire high-density temperature measurements from thousands of points along a single optical fiber, enabling fine-grained perception of the three-dimensional thermal field. On this basis, a hybrid prediction model integrating thermodynamic physical equations with a Temporal Convolutional Network-Bidirectional Gated Recurrent Unit (TCN-BiGRU) deep neural network is developed to achieve accurate and stable spatiotemporal temperature forecasting. Furthermore, a symmetry-aware MPC controller is designed with the dual objectives of minimizing cooling energy consumption and suppressing thermal field deviations, thereby restoring temperature uniformity through rolling-horizon optimization. Experimental validation in a production data center demonstrates that the distributed sensing system achieves a measurement deviation of 0.12 degrees C, while the hybrid prediction model attains a root mean square error of 0.41 degrees C, representing a 26.8% improvement over baseline methods. The MPC-based control strategy reduces daily cooling energy consumption by 14.4%, improves the power usage effectiveness (PUE) from 1.58 to 1.47, and significantly enhances both thermal symmetry and operational safety. The Thermal Symmetry Index (TSI) decreased from 0.060 to 0.035, indicating a 41.7% improvement in spatial temperature distribution uniformity. The TSI is defined as the ratio of spatial temperature standard deviation to mean temperature, where lower values indicate better thermal uniformity; TSI < 0.03 represents excellent symmetry, 0.03-0.05 indicates good symmetry, and TSI > 0.08 suggests significant asymmetry requiring intervention. These results provide an effective and practical solution for intelligent operation, energy-efficient control, and low-carbon transformation of next-generation green data centers.

The inevitable graphitization of bituminous coal during pyrolysis inhibits its application as a precursor for carbon anodes in sodium ion batteries. Heterogeneous crosslinking modification is an effective pathway to enhance its degree of disorder. In this paper, carbon dots (CDs) were employed as a regulator to modulate the structure of bituminous coal-based carbon. The C(O)-O groups formed through cross-linking reactions between oxygencontaining groups of CDs and bituminous coal suppress the long-range ordered arrangement of carbon microcrystals during coal pyrolysis, thereby enhancing the disorder of coal-based carbon. The test results reveal that the obtained BCDs-75 possesses expanded interlayer spacing, abundant closed pores and surface defects. As an anode for SIBs, BCDs-75 releases an exceptional initial charge capacity of 302 mAh g- 1 at 0.1 A g- 1, with an initial coulombic efficiency of 65.5%. The combined electrochemical measurement and carbon structure characterization indicate that the disordered structure and closed pores increase the plateau capacity of BCDs-75 anode, while the increase of surface oxygen-containing groups enhance its slope capacity.