Physical information-enhanced graph neural network for predicting phase separation

Although phase separation is a ubiquitous phenomenon, the interactions between multiple components make it difficult to accurately model and predict. In recent years, machine learning has been widely used in physics simulations. Here,we present a physical information-enhanced graph neural network(PIENet) to simulate and predict the evolution of phase separation. The accuracy of our model in predicting particle positions is improved by 40.3% and 51.77% compared with CNN and SVM respectively. Moreover, we design an order parameter based on local density to measure the evolution of phase separation and analyze the systematic changes with different repulsion coefficients and different Schmidt numbers.The results demonstrate that our model can achieve long-term accurate predictions of order parameters without requiring complex handcrafted features. These results prove that graph neural networks can become new tools and methods for predicting the structure and properties of complex physical systems.

Dissipativity Analysis of Neural Networks with Time-varying Delays

A new definition of dissipativity for neural networks is presented in this paper. By constructing proper Lyapunov functionals and using some analytic techniques, sufficient conditions are given to ensure the dissipativity of neural networks with or without time-varying parametric uncertainties and the integro-differential neural networks in terms of linear matrix inequalities. Numerical examples are given to illustrate the effectiveness of the obtained results.

A Clustering-tree Topology Control Based on the Energy Forecast for Heterogeneous Wireless Sensor Networks

How to design an energy-efficient algorithm to maximize the network lifetime in complicated scenarios is a critical problem for heterogeneous wireless sensor networks (HWSN). In this paper, a clustering-tree topology control algorithm based on the energy forecast (CTEF) is proposed for saving energy and ensuring network load balancing, while considering the link quality, packet loss rate, etc. In CTEF, the average energy of the network is accurately predicted per round (the lifetime of the network is denoted by rounds) in terms of the difference between the ideal and actual average residual energy using central limit theorem and normal distribution mechanism, simultaneously. On this basis, cluster heads are selected by cost function (including the energy, link quality and packet loss rate) and their distance. The non-cluster heads are determined to join the cluster through the energy, distance and link quality. Furthermore, several non-cluster heads in each cluster are chosen as the relay nodes for transmitting data through multi-hop communication to decrease the load of each cluster-head and prolong the lifetime of the network. The simulation results show the efficiency of CTEF. Compared with low-energy adaptive clustering hierarchy (LEACH), energy dissipation forecast and clustering management (EDFCM) and efficient and dynamic clustering scheme (EDCS) protocols, CTEF has longer network lifetime and receives more data packets at base station. © 2014 Chinese Association of Automation.

A Novel Hybrid Clustering Based Transmission Protocol for Wireless Body Area Networks

Wireless sensor networks are a collection of intelligent sensor devices that are connected to one another and have the capability to exchange information packets amongst themselves.In recent years,this field of research has become increasingly popular due to the host of useful applications it can potentially serve.A deep analysis of the concepts associated with this domain reveals that the two main problems that are to be tackled here are throughput enhancement and network security improvement.The present article takes on one of these two issues namely the throughput enhancement.For the purpose of improving network productivity,a hybrid clustering based packet propagation protocol has been proposed.The protocol makes use of not only clustering mechanisms of machine learning but also utilizes the traditional forwarding function approach to arrive at an optimum model.The result of the simulation is a novel transmission protocol which significantly enhances network productivity and increases throughput value.

Dynamics of high-frequency modulated waves in a nonlinear dissipative continuous bi-inductance network

This paper presents intensive investigation of dynamics of high frequency nonlinear modulated excitations in a damped bimodal lattice. The effects of the dissipation are considered through a linear dissipation coefficient whose evolution in terms of the carrier wave frequency is checked. There appears that the dissipation coefficient increases with the carrier wave frequency. In the linear limit and for high frequency waves, study of the asymptotic behavior of plane waves reveals the existence of two additional regions in the dispersion curve where the modulational phenomenon is observed compared to the lossless line. Based on the multiple scales method exploited in the continuum approximation using an appropriate decoupling ansatz for the voltage of the two different cells, it appears that the motion of modulated waves is described by a dissipative complex Ginzburg–Landau equation instead of a Korteweg–de Vries equation. We also show that this amplitude wave equation admits envelope and hole solitons in the high frequency mode. From basic sources, we design a programmable electronic generator of complex signals with desired characteristics, which delivers signals exploited as input waves for all our numerical simulations. These simulations are performed in the LTspice software that uses realistic components and give the results that corroborate perfectly our analytical predictions.

航空电推进电机多层波浪形拓扑及散热设计方法

针对航空电推进电机的散热问题,提出一种基于多层波浪形散热拓扑的航空电推进电机高效散热设计方法。航空电推进电机采用多层波浪形散热拓扑,建立电机等效热网络模型,确定等效热阻、对流换热系数等重要参数,完成电机温度精确计算,并通过CFD仿真验证所建立电机等效热网络模型的准确性和有效性。以此为基础,对比分析传统散热翅和多层波浪形散热拓扑对电机功率密度的影响。基于多层波浪形散热拓扑的等效热网络模型,采用遗传学习粒子群优化(GL-PSO)算法,完成航空电推进电机高效散热优化设计。优化结果表明:相比于原始方案,优化方案的机壳质量减轻15.1%,整个电机的功率密度提升0.06 kW/kg。

基于高低温分区热管网络的高轨通信卫星散热能力提升方法

针对我国高轨通信卫星不断提升的散热需求,提出采用基于高低温分区热管网络的热设计方法。在分析热管网络散热能力的基础上,进行设备分区布局设计、分区热管网络散热面设计和分区热管网络布局设计。该方法可在保证卫星仪器设备工作温度、不降低热设计可靠性、不增加分系统重量及功率需求的前提下,将卫星散热能力提升15%~20%。文章还结合近些年通信卫星研制中的经验和教训,归纳给出高低温分区热管网络设计中的注意事项,以期为相关工程设计提供有效指导。

基于事件触发的网络化系统输出反馈耗散控制

针对事件触发机制下网络化系统的输出反馈耗散控制问题进行研究。首先,引入事件触发机制来减少有限带宽的影响,节约系统资源;其次,考虑网络化系统中存在的外部干扰,利用输入时滞法建立了基于事件触发机制的时滞系统模型;然后,构造一种时滞相关的Lyapunov-Krasovskii泛函,利用改进的积分处理方法推导出了保证闭环系统严格耗散性能的充分条件,并采用锥补线性化算法求出输出反馈控制器增益和事件触发参数矩阵;最后,通过数值仿真验证了所提方法的有效性与优越性。

基于卷积神经网络的水下湍流探测技术

针对水下湍流的复杂性和多变性对水下航行器性能和姿态控制产生的挑战,提出使用卷积神经网络来测量水下湍流的温差耗散率XT。首先,采用功率谱反演法和惠更斯-菲涅尔原理仿真生成了受水下湍流影响的散斑图像数据集。随后,利用卷积神经网络提取这些受湍流影响的散斑图像中的特征信息,并对温差耗散率XT进行估计。最后,通过现场实验数据集验证了所提出方法的可行性。实验结果表明,所提出的神经网络在实地实验数据集和模拟仿真数据集上表现出相似的分类精度和损失曲线,其测量准确率分别为98.8%和99.2%。这一研究为水下环境监测和资源勘探领域提供了重要的参考,对于光学图像处理和湍流研究等相关领域具有实际意义。

基于FNM技术的风冷电子设备散热性能分析

为了快速有效地分析电子设备的散热性能,提出一种CFD(Computational Fluid Dynamics,计算流体力学)分析和流动阻力网络分析的耦合方法,基于低计算成本CFD分析的结果,开发了流动阻力网络模型,通过使用初步CFD分析结果开发的流动阻力网络模型,预测电子设备中每个流道中的流量分布。该方法对采用风冷的电子设备的热设计具有很好的设计指导意义。

基于热阻网络的高性能贴装设备温控设计和仿真分析

阐述利用热阻网络对高性能半导体贴装设备温控系统进行应用分析,通过热阻网络的构建能够快速评估出温控台隔热性能、降温速率等性能,给出有效的设计指导建议。通过仿真验证,显示优化后的温控台散热性能均达到设计指标要求,且与热阻网络分析结果接近。利用热阻网络分析方法可快速收敛设计方案。

Ultrafast battery heat dissipation enabled by highly ordered and interconnected hexagonal boron nitride thermal conductive composites

Heat dissipation involved safety issues are crucial for industrial applications of the high-energy density battery and fast charging technology.While traditional air or liquid cooling methods suffering from space limitation and possible leakage of electricity during charge process,emerging phase change materials as solid cooling media are of growing interest.Among them,paraffin wax(PW)with large latent heat capacity and low cost is desirable for heat dissipation and thermal management which mainly hindered by their relatively low thermal conductivity and susceptibility to leakage.Here,highly ordered and interconnected hexagonal boron nitride(h-BN)networks were established via ice template method and introduced into PW to enhance the thermal conductivity.The composite with 20 wt%loading amount of h-BN can guarantee a highly ordered network and exhibited high thermal conductivity(1.86 W m^(-1) K^(-1))which was 4 times larger compared with that of random dispersed h-BN involved PW and nearly 8 times larger compared with that of bare PW.The optimal thermal conductive composites demonstrated ultrafast heat dissipation as well as leakage resistance for lithium-ion batteries(LIBs),heat generated by LIBs can be effectively transferred under the working state and the surface temperature kept 6.9℃ lower at most under 2–5℃ continuous charge-discharge process compared with that of bare one which illustrated great potential for industrial thermal management.

Dissipative-based adaptive neural control for nonlinear systems

A dissipative-based adaptive neural control scheme was developed for a class of nonlinear uncertain systems with unknown nonlinearities that might not be linearly parameterized. The major advantage of the present work was to relax the requirement of matching condition, i.e., the unknown nonlinearities appear on the same equation as the control input in a state-space representation, which was required in most of the available neural network controllers. By synthesizing a state-feedback neural controller to make the closed-loop system dissipative with respect to a quadratic supply rate, the developed control scheme guarantees that the L2-gain of controlled system was less than or equal to a prescribed level. And then, it is shown that the output tracking error is uniformly ultimate bounded. The design scheme is illustrated using a numerical simulation.

Nonlinear continuous bi-inductance electrical line with dissipative elements:Dynamics of the low frequency modulated waves

The dynamics of modulated waves in a nonlinear bi-inductance transmission line with dissipative elements are examined.We show the existence of two frequency modes and carry out intensive investigations on the low frequency mode.Thanks to the multiple scales method,the behavior of these waves is investigated and the dissipative effects are analyzed.It appears that the dissipation coefficient increases with the carrier wave frequency.In the continuous approximation,we derive that the propagation of these waves is governed by the complex Ginzburg-Landau equation instead of the Korteweg-de-Vries equation as previously established.Asymptotic studies of the dynamics of plane waves in the line reveal the existence of three additional regions in the dispersion curve where the modulational phenomenon is observed.In the low frequency mode,we demonstrate that the network allows the propagation of dark and bright solitons.Numerical findings are in perfect agreement with the analytical predictions.

Robust networked H_∞ synchronization of nonidentical chaotic Lur'e systems

We mainly investigate the robust networked H~ synchronization problem of nonidentical chaotic Lur＇e systems. In the design of the synchronization scheme, some network characteristics, such as nonuniform sampling, transmission- induced delays, and data packet dropouts, are considered. The parameters of master-slave chaotic Lur＇e systems often allow differences. The sufficient condition in terms of linear matrix inequality （LMI） is obtained to guarantee the dissipative synchronization of nonidentical chaotic Lur＇e systems in network environments. A numerical example is given to illustrate the validity of the proposed method.

Tree Network Formation in Poisson Equation Models and the Implications for the Maximum Entropy Production Principle

This paper presents not only practical but also instructive mathematical models to simulate tree network formation using the Poisson equation and the Finite Difference Method (FDM). Then, the implications for entropic theories are discussed from the viewpoint of Maximum Entropy Production (MEP). According to the MEP principle, open systems existing in the state far from equilibrium are stabilized when entropy production is maximized, creating dissipative structures with low entropy such as the tree-shaped network. We prepare two simulation models: one is the Poisson equation model that simulates the state far from equilibrium, and the other is the Laplace equation model that simulates the isolated state or the state near thermodynamic equilibrium. The output of these equations is considered to be positively correlated to entropy production of the system. Setting the Poisson equation model so that entropy production is maximized, tree network formation is advanced. We suppose that this is due to the invocation of the MEP principle, that is, entropy of the system is lowered by emitting maximal entropy out of the system. On the other hand, tree network formation is not observed in the Laplace equation model. Our simulation results will offer the persuasive evidence that certifies the effect of the MEP principle.

导流式输水管网消能装置设计与影响因素研究

为了保障山区输水管网运行安全,设计导流式输水管网消能装置。装置由上壳体、消能空腔和下壳体组成,设置入口端和出口端,消能空腔内设有均匀间隔的消能板和导流孔。采用Fluent数值模拟和验证试验对试验方法进行验证,设置3种入口流速、3种导流孔径比例和有无导流片开展全因素试验,并对2种导流孔直径进行消能率对比试验。结果表明:在保证过流能力下,入口流速和导流孔径均对消能率起主导作用,入口流速越大,即流量越大,消能率越好。消能率与导流孔径负相关,导流孔径越小越有利于消能。当基础孔径相同时,为同时满足过流能力且确保消能达到较好效果,建议选择导流孔径比例保持不变布置方式。入口流速为1.0 m/s时,局部水头损失占总水头损失的96.3%,所以当计算总水头损失时,可以忽略沿程水头损失。当入口流速小于4.0 m/s时,选择不安装导流片,达到4.0 m/s时,有无导流片消能率基本持平,大于5.0 m/s后,选用安装导流片消能效果更优。

永磁同步电机拓扑结构对散热性能的影响

为了探究永磁同步电机拓扑结构变化对电机散热性能的影响,以30 kW电动叉车用水冷型内置式永磁同步电机为例,建立改进的等效热网络模型,并利用电机的有限元温度场模型计算验证其准确性。采用热网络模型,计算不同定转子结构下电机的温度场,分析定子齿宽、定子槽深、永磁体夹角等拓扑结构改变对电机散热性能的影响规律,为永磁同步电机的运行性能优化提供参考。

基于耗散结构理论的主流意识形态网络传播

基于耗散结构理论,主流意识形态的网络传播通常具有开放性、非平衡性、非线性和随机涨落等特征。强化主流意识形态网络传播,坚持开放性,把握其开放的度和界;坚持非平衡性,找寻其秩序生成归因;坚持非线性,引导其由无序向有序发展;坚持随机涨落,创新其传播方式,进而增强主流意识形态网络传播的引领力、公信力和感染力,牢牢掌握网络意识形态工作主导权。