arXiv subjects

Yu Mei

Publications and source records attributed to Yu Mei.

At least 37 records · Page 2Linked to original sources

arXiv2025

Existence and convergence of the Beris-Edwards system with general Landau-de Gennes energy

In this paper, we investigate the Beris-Edwards system for both biaxial and uniaxial $Q$-tensors with a general Landau-de Gennes energy density depending on four non-zero elastic constants. We prove existence of the strong solution of the Beris-Edwards system for uniaxial $Q$-tensors up to a maximal time. Furthermore, we prove that the strong solutions of the Beris-Edwards system for biaxial $Q$-tensors converge smoothly to the solution of the Beris-Edwards system for uniaxial $Q$-tensors up to its maximal existence time.

math.AP↗

arXiv2025

Stability of Large-Amplitude Viscous Shock Under Periodic Perturbation for 1-d Viscoelasticity with Non-Convex Constitutive Relations

This paper investigates the large-time behavior of the viscous shock profile for the one-dimensional system of viscoelasticity, subject to initial perturbations that approach space-periodic functions at far fields. We specifically address the case with non-convex constitutive stress relations and non-degenerate Lax's shock. Under the assumptions of suitably small initial perturbations satisfying a zero-mass type condition, we prove that the solution of the system converges to a viscous shock profile with a shift, which is partially determined by the space-periodic perturbation. Notably, our result imposes no amplitude restrictions on the viscous shock waves. This work extends the result of Kawashima-Matsumura (\textit{Commun. Pure Appl. Math.} \textbf{47}, 1994) by simultaneously handling both large-amplitude shocks and space-periodic perturbations, while also generalizing the result of Huang-Yuan (\textit{Commun. Math. Phys.} \textbf{387}, 2021) by allowing for a non-convex constitutive relation. The key ingredient of proof is decomposing the large-amplitude shock wave into small-amplitude shocks and, for each, introducing suitable transform and weight functions to counteract the adverse effects of non-convex constitutive relations encountered during weighted energy estimates on the system in effective velocity and deformation gradient variables.

math.AP↗

arXiv2025

Global strong solution of the 3D compressible liquid crystal flows with density-dependent viscosity and large velocity

This paper concerns the Cauchy problem of three-dimensional compressible liquid crystal flows with density-dependent viscosity. When the viscosity coefficients $μ_1(ρ),μ_2(ρ)$ are power functions of the density with the power larger than $1$, it is proved that the system exists a unique global strong solution as long as the initial density is sufficiently large and $L^3$-norm of the derivative of the initial director is sufficiently small. This is the first result concerning the global strong solution for three-dimensional compressible liquid crystal flows without smallness of velocity.

math.AP↗

arXiv2025

Learning-Based Modeling of Soft Actuators Using Euler Spiral-Inspired Curvature

Soft robots, distinguished by their inherent compliance and continuum structures, present unique modeling challenges, especially when subjected to significant external loads such as gravity and payloads. In this study, we introduce an innovative data-driven modeling framework leveraging an Euler spiral-inspired shape representations to accurately describe the complex shapes of soft continuum actuators. Based on this representation, we develop neural network-based forward and inverse models to effectively capture the nonlinear behavior of a fiber-reinforced pneumatic bending actuator. Our forward model accurately predicts the actuator's deformation given inputs of pressure and payload, while the inverse model reliably estimates payloads from observed actuator shapes and known pressure inputs. Comprehensive experimental validation demonstrates the effectiveness and accuracy of our proposed approach. Notably, the augmented Euler spiral-based forward model achieves low average positional prediction errors of 3.38%, 2.19%, and 1.93% of the actuator length at the one-third, two-thirds, and tip positions, respectively. Furthermore, the inverse model demonstrates precision of estimating payloads with an average error as low as 0.72% across the tested range. These results underscore the potential of our method to significantly enhance the accuracy and predictive capabilities of modeling frameworks for soft robotic systems.

cs.RO↗

arXiv2024

A Link-Based Flow Model with Turn-Level Queue Transmission and Time-Varying Free-Flow Speed for Urban Road Networks

Macroscopic link-based flow models are efficient for simulating flow propagation in urban road networks. Existing link-based flow models described traffic states of a link with two state variables of link inflow and outflow and assumed homogeneous traffic states within a whole link. Consequently, the turn-level queue length change within the link can not be captured, resulting in underrepresented queue spillback. Moreover, a constant link free-flow speed was assumed to formulate models, restricting their applicability in modeling phenomena involving time-varying free-flow speed. This study proposed a new link-based flow model by introducing an additional state variable of link queue inflow and adapting the link outflow to be free-flow speed-dependent. In our model, the vehicle propagation within each link is described by the link inflow, queue inflow, and outflow, which depends on the link free-flow speed changes. A node model is further defined to capture the presence of signal control and potential queue spillback, which estimates the constrained flow propagation between adjacent road segments. Simulation experiments were conducted on a single intersection and a network with consecutive intersections to verify the proposed model performance. Results demonstrate the predictive power of the proposed model in predicting traffic operations of intersections with multiple turning movements and time-varying free-flow speed. Our model outperforms the baseline link-based flow model and preserves the computational tractability property of link-based flow models.

eess.SY↗

arXiv2024

Contextualizing MLP-Mixers Spatiotemporally for Urban Data Forecast at Scale

Spatiotemporal traffic data (STTD) displays complex correlational structures. Extensive advanced techniques have been designed to capture these structures for effective forecasting. However, because STTD is often massive in scale, practitioners need to strike a balance between effectiveness and efficiency using computationally efficient models. An alternative paradigm based on multilayer perceptron (MLP) called MLP-Mixer has the potential for both simplicity and effectiveness. Taking inspiration from its success in other domains, we propose an adapted version, named NexuSQN, for STTD forecast at scale. We first identify the challenges faced when directly applying MLP-Mixers as seriesand window-wise multivaluedness. To distinguish between spatial and temporal patterns, the concept of ST-contextualization is then proposed. Our results surprisingly show that this simple-yeteffective solution can rival SOTA baselines when tested on several traffic benchmarks. Furthermore, NexuSQN has demonstrated its versatility across different domains, including energy and environment data, and has been deployed in a collaborative project with Baidu to predict congestion in megacities like Beijing and Shanghai. Our findings contribute to the exploration of simple-yet-effective models for real-world STTD forecasting.

cs.LG↗

arXiv2024

Global strong solution of the 3D inhomogeneous liquid crystal flows with density-dependent viscosity and large velocity

This paper concerns the initial boundary value problem of three-dimensional inhomogeneous incompressible liquid crystal flows with density-dependent viscosity. When the viscosity coefficient $μ(ρ)$ is a power function of the density with the power larger than $1$, that is $μ(ρ)=μρ^α$ with $α>1$, it is proved that the system exists a unique global strong solution as long as the initial density is sufficiently large and $L^3$-norm of the derivative of the initial director is sufficiently small. This is the first result concerning the global strong solution for three-dimensional inhomogeneous liquid crystal flows without smallness of velocity.

math.AP↗

arXiv2024

Irregular Traffic Time Series Forecasting Based on Asynchronous Spatio-Temporal Graph Convolutional Network

Accurate traffic forecasting is crucial for the development of Intelligent Transportation Systems (ITS), playing a pivotal role in modern urban traffic management. Traditional forecasting methods, however, struggle with the irregular traffic time series resulting from adaptive traffic signal controls, presenting challenges in asynchronous spatial dependency, irregular temporal dependency, and predicting variable-length sequences. To this end, we propose an Asynchronous Spatio-tEmporal graph convolutional nEtwoRk (ASeer) tailored for irregular traffic time series forecasting. Specifically, we first propose an Asynchronous Graph Diffusion Network to capture the spatial dependency between asynchronously measured traffic states regulated by adaptive traffic signals. After that, to capture the temporal dependency within irregular traffic state sequences, a personalized time encoding is devised to embed the continuous time signals. Then, we propose a Transformable Time-aware Convolution Network, which adapts meta-filters for time-aware convolution on the sequences with inconsistent temporal flow. Additionally, a Semi-Autoregressive Prediction Network, comprising a state evolution unit and a semi-autoregressive predictor, is designed to predict variable-length traffic sequences effectively and efficiently. Extensive experiments on a newly established benchmark demonstrate the superiority of ASeer compared with twelve competitive baselines across six metrics.

cs.LG↗

arXiv2024

Design and Nonlinear Modeling of a Modular Cable Driven Soft Robotic Arm

We propose a novel multi-section cable-driven soft robotic arm inspired by octopus tentacles along with a new modeling approach. Each section of the modular manipulator is made of a soft tubing backbone, a soft silicon arm body, and two rigid endcaps, which connect adjacent sections and decouple the actuation cables of different sections. The soft robotic arm is made with casting after the rigid endcaps are 3D-printed, achieving low-cost and convenient fabrication. To capture the nonlinear effect of cables pushing into the soft silicon arm body, which results from the absence of intermediate rigid cable guides for higher compliance, an analytical static model is developed to capture the relationship between the bending curvature and the cable lengths. The proposed model shows superior prediction performance in experiments over that of a baseline model, especially under large bending conditions. Based on the nonlinear static model, a kinematic model of a multi-section arm is further developed and used to derive a motion planning algorithm. Experiments show that the proposed soft arm has high flexibility and a large workspace, and the tracking errors under the algorithm based on the proposed modeling approach are up to 52$\%$ smaller than those with the algorithm derived from the baseline model. The presented modeling approach is expected to be applicable to a broad range of soft cable-driven actuators and manipulators.

cs.RO↗

arXiv2024

From Words to Molecules: A Survey of Large Language Models in Chemistry

In recent years, Large Language Models (LLMs) have achieved significant success in natural language processing (NLP) and various interdisciplinary areas. However, applying LLMs to chemistry is a complex task that requires specialized domain knowledge. This paper provides a thorough exploration of the nuanced methodologies employed in integrating LLMs into the field of chemistry, delving into the complexities and innovations at this interdisciplinary juncture. Specifically, our analysis begins with examining how molecular information is fed into LLMs through various representation and tokenization methods. We then categorize chemical LLMs into three distinct groups based on the domain and modality of their input data, and discuss approaches for integrating these inputs for LLMs. Furthermore, this paper delves into the pretraining objectives with adaptations to chemical LLMs. After that, we explore the diverse applications of LLMs in chemistry, including novel paradigms for their application in chemistry tasks. Finally, we identify promising research directions, including further integration with chemical knowledge, advancements in continual learning, and improvements in model interpretability, paving the way for groundbreaking developments in the field.

cs.LG↗

arXiv2023

A Fully Data-Driven Approach for Realistic Traffic Signal Control Using Offline Reinforcement Learning

The optimization of traffic signal control (TSC) is critical for an efficient transportation system. In recent years, reinforcement learning (RL) techniques have emerged as a popular approach for TSC and show promising results for highly adaptive control. However, existing RL-based methods suffer from notably poor real-world applicability and hardly have any successful deployments. The reasons for such failures are mostly due to the reliance on over-idealized traffic simulators for policy optimization, as well as using unrealistic fine-grained state observations and reward signals that are not directly obtainable from real-world sensors. In this paper, we propose a fully Data-Driven and simulator-free framework for realistic Traffic Signal Control (D2TSC). Specifically, we combine well-established traffic flow theory with machine learning to construct a reward inference model to infer the reward signals from coarse-grained traffic data. With the inferred rewards, we further propose a sample-efficient offline RL method to enable direct signal control policy learning from historical offline datasets of real-world intersections. To evaluate our approach, we collect historical traffic data from a real-world intersection, and develop a highly customized simulation environment that strictly follows real data characteristics. We demonstrate through extensive experiments that our approach achieves superior performance over conventional and offline RL baselines, and also enjoys much better real-world applicability.

cs.AI↗

arXiv2023

Asymptotic stability of shock profiles and rarefaction waves to the Navier-Stokes-Poisson system under space-periodic perturbations

This paper concerns with the large-time behaviors of the viscous shock profile and rarefaction wave under initial perturbations which tend to space-periodic functions at infinities for the one-dimensional compressible Navier-Stokes-Poisson equations. It is proved that: (1) for the viscous shock with small strength, if the initial perturbation is suitably small and satisfies a zero-mass type condition, then the solution tends to background viscous shock with a constant shift as time tends to the infinity, and the shift depends on both the mass of the localized perturbation, and the space-periodic perturbation; (2) for the rarefaction wave, if the initial perturbation is suitably small, then the solution tends to background rarefaction wave as time tends to infinity. The proof is based on the delicate constructions of the quadratic ansatzes, which capture the infinitely many interactions between the background waves and the periodic perturbations, and the energy method in Eulerian coordinates involving the effect of self-consistent electric field. Moreover, an abstract lemma is established to distinguish the non-decaying terms and good-decaying terms from the error terms of the equations of the quadratic ansatzes, which will be benefit to constructing the ansatzes and simplifying calculations for other non-localized perturbation problems, especially those with complicatedly coupling physical effects.

math.AP↗

arXiv2022

Vanishing viscosity limits for the free boundary problem of compressible viscoelastic fluids with surface tension

We consider the free boundary problem of compressible isentropic neo-Hookean viscoelastic fluid equations with surface tension. Under the physical kinetic and dynamic conditions proposed on the free boundary, we investigate regularities of classical solutions to viscoelastic fluid equations in Sobolev spaces which are uniform in viscosity and justify the corresponding vanishing viscosity limits. The key ingredient of our proof is that the deformation gradient tensor in Lagrangian coordinates can be represented as a parameter in terms of flow map so that the inherent structure of the elastic term improves the uniform regularity of normal derivatives in the limit of vanishing viscosity. This result indicates that the boundary layer does not appear in the free boundary problem of compressible viscoelastic fluids which is different to the case studied by the second author for the free boundary compressible Navier-Stokes system.

math.AP↗

arXiv2021

Distributionally robust second-order stochastic dominance constrained optimization with Wasserstein ball

We consider a distributionally robust second-order stochastic dominance constrained optimization problem. We require the dominance constraints hold with respect to all probability distributions in a Wasserstein ball centered at the empirical distribution. We adopt the sample approximation approach to develop a linear programming formulation that provides a lower bound. We propose a novel split-and-dual decomposition framework which provides an upper bound. We establish quantitative convergency for both lower and upper approximations given some constraint qualification conditions. To efficiently solve the non-convex upper bound problem, we use a sequential convex approximation algorithm. Numerical evidences on a portfolio selection problem valid the convergency and effectiveness of the proposed two approximation methods.

math.OC↗

arXiv2021

Eavesdropping Risk Evaluation on Terahertz Wireless Channels in Atmospheric Turbulence

Wireless networks operating at terahertz (THz) frequencies have been proposed as a promising candidate to support the ever-increasing capacity demand, which cannot be satisfied with existing radio-frequency (RF) technology. On the other hand, it likely will serve as backbone infrastructure and could therefore be an attractive target for eavesdropping attacks. Compared with regular RF spectrum, wireless channels in the THz range could be less vulnerable to interceptions because of their high beam directionality and small signal coverage. However, a risk for eavesdropping can still exist due to the multipath effects caused by unintended scattering. In this work, an eavesdropping risk for THz channel passing atmospheric turbulences and producing compromising emissions is investigated from a physical layer perspective. A model combining signal attenuation due to turbulence, gaseous absorption and beam divergence, is developed for prediction of deterministic and probabilistic signal leakages. The secrecy capacity and outage probability of the THz channel are derived and analyzed with respect to variations of the turbulence strength and other channels characteristics. The dependence of the channel performance on the eavesdropper's position is investigated with respect to the maximum safe data transmission rate (MSR) and the signal leakage region. Design results for THz channels are provided to minimize an eavesdropping risk at physical layer.

eess.SP↗

arXiv2021

Secrecy Performance of Terahertz Wireless Links in Rain and Snow

Wireless communication technique at terahertz (THz) frequencies is regarded as the most potential candidate for future wireless networks due to its wider frequency bandwidth and higher data capacity when compared to that employing radio frequency (RF) and millimeter wave (mmWave). Besides, a THz link can achieve higher security at physical layer when it propagates in clear weather due to its higher directionality, which reduces the possibility of eavesdropping attacks. However, under adverse weather conditions (such as water fog, dust fog, rain and snow), the link degradation due to weather particles and gaseous molecules will affect the link secrecy performance seriously. In this work, we present theoretical investigations on physical layer security of a point-to-point THz link in rain and snow with a potential eavesdropper locating outside of the legitimate link path. Signal degradation due to rain/snow, gaseous attenuation and beam divergence are included in a theoretical model to estimate the link performance. Secrecy capacity of the link with carriers at 140, 220 and 340 GHz is calculated and compared. Insecure regions are also presented and the secrecy performance is analyzed. We find that the THz link suffers least eavesdropping attacks in rain and the maximum data transmission rate decreases for higher carrier frequencies in rain and snow.

physics.app-ph↗

arXiv2020

Planning Skip-Stop Transit Service under Heterogeneous Demands

Transit vehicles operating under skip-stop service visit only a subset of the stops residing along a corridor. It is a strategy commonly used to increase vehicle speeds and reduce patron travel times. The present paper develops a continuous approximation model to optimally design a select form of skip-stop service, termed AB-type service. The model accounts for spatially-heterogeneous demand patterns. An efficient heuristic is developed to obtain solutions. These are shown to be near-optimal for a variety of numerical examples. Results also indicate that optimal AB-type designs outperform optimized all-stop service in a variety of cases. The AB-type service is found to be especially competitive when travel demands are high, trip origins are unevenly distributed along a corridor, and patrons have relatively high values of time. In these cases, AB-type service is found to reduce system costs by as much as 8%.

math.OC↗

arXiv2020

Leaf Water Status Monitoring by Scattering Effects at Terahertz Frequencies

The applications of terahertz (THz) radiation for plant water status monitoring require systematic studies on interaction of THz wave and plants. Here, we present theoretical investigations on scattering behavior of THz waves reflected by and transmitting through a plant leaf under different water content. A theoretical model combining integral equation and radiative transfer theory is presented to fit the measured data. Good agreement confirms the availability of the model for water status evaluation when variation of leaf thickness and surface roughness is considered. We investigate the applicability of THz waves for water status monitoring in reflection and transmission geometries under different temperatures, salinities and polarizations.

physics.bio-ph↗