基于神经网络与光纤传感阵列的结构状态监测方法

以探测机敏复合材料与结构内应变、应力及损伤等物理状态的位置信息为目的．提出了一种新颖的可内埋于材料与结构内作为结构状态监测用的强度调制型光纤传感阵列网络，并采用人工神经网络来处理传感阵列输出的并行分布式传感信号，阐述了适用的Ｋｏｈｏｎｅｎ网模型及其变化形式，给出了仿真实验结果     

Composite and filament models for the mechanical behaviour of elastomeric materials

In this work we present a composite model, which combines the approach of Poisson's function with the filament theory and requires three material parameters. We also suggest the form for a strain-energy function that approximates the constitutive equations of the composite model. Furthermore, a simple asymptotic analysis allows us to reduce the number of material constants to only two, thus, forming a new filament model. The predictive capability of the two models to reproduce the mechanical behaviour of elastomeric materials in deformation experiments is evaluated against the extensive data of Kawabata et al. (Macromolecules 14 (1981) 154). The models give excellent agreement in not only uniaxial and equibiaxial but also non-equibiaxial extension. Although being rather more simplistic in comparison with some successful network models involving non-Gaussian chain statistics, the two models conform much more closely to the classical experimental data of Treloar (Trans. Faraday Soc. 40 (1944) 59).     

神经网络算法及其在GPS导航中的应用

GPS导航解算中常采用最小二乘算法。由于导航解算中存在观测异常（粗差），采用的模型精度很难满足用户需求，而高动态用户需求的精度却不断提高，为此文中提出一种利用神经网络的非线性逼近能力，在顾及观测信息权阵的条件下，利用Hopfieid神经网络进行GPS导航解算的新算法。该算法具有很好的抗差性和自适应性，能较好地抑制观测粗差对导航解的影响。计算结果证明了该算法的有效性。     

Necessary conditions for shear thickening in associating polymer networks

We study shear-thickening phenomena observed in thermoreversible gels of telechelic associating polymers on the basis of the transient network theory. We introduce nonlinear models for the elongational properties of the middle chain; the tension along middle chains contains a nonlinear term with an amplitude A that diverges if the chain is fully stretched. Models for the dissociation rate of end chains from network junctions are also introduced; the dissociation rate is coupled to the middle-chain tension by a coupling constant g. The balance between the nonlinearity strength and coupling intensity determines the necessary condition for thickening. A strong nonlinearity in the tension caused by a large A leads to thickening if the dissociation rate of the end chains is weakly coupled to the middle-chain tension. In contrast, if it is strongly coupled, the end chains easily dissociate upon a small nonlinear elongation of the middle chains, thus leading to thinning. We present theoretical phase diagrams showing the boundary of the thickening/thinning transition in the A–g plane, and we show the existence of a critical point for thickening in these diagrams.     

损伤弹性材料的有效模量及本构关系

考虑了微裂纹之间的相互作用，利用随机熔断丝网络（Ｒａｎｄｏｍｆｕｓｅｎｅｔｗｏｒｋ）作为模型，模拟了脆性材料在外载荷作用下，其有效模量随损伤的演变行为，并考察了系统含损伤的本构关系。发现在损伤初期，有效模量和损伤具有线性下降关系；而在后期，由于损伤局部化，有效模量迅速下降，提出了一个有效模量随损伤变化的非线性关系。讨论了利用网络模型来研究材料含损伤本构关系的可行性和优越性。     

Rheological properties of liver actin solutions

G-actin prepared from pork liver and purified to give a product of 95–98% purity was polymerized to F-actin in a rotational dynamic viscometer. The rheological properties were investigated during the course of polymerisation and after polymerisation. G andG measured at frequencies around 0.1 Hz increased strongly during the first 3 hours of the reaction and then slowly approached a constant value in the range of 10–50 mPa. The rise during polymerisation was more than two orders of magnitude. When a solution of F-actin was subjected to steady shear at a high shear rate for a short period of time and subsequently dynamic measurements were performed, then a considerably smallerG (about 70% smaller) was observed immediately after cessation of the steady shear.G then increased with time and approached its initial value.When measuring the viscosity of F-actin solutions as a function of shear rate, a strong shear thinning effect was seen which did not vanish even at shear rates as low as 0.001 s^–1. Even at this low shear rates, a pronounced yield maximum was observed before reaching the steady state. Oscillatory experiments showed a remarkably weak frequency dependence ofG. The results imply that F-actin solutions are largely structured forming a weak temporary network which can be easily destroyed by application of high shear rates. It seems most likely that the destruction occurs by a reversible, shear induced depolymerisation process.     

埋入压电元件的自诊断智能结构的理论分析与实验研究

将压电元件埋设于复合材料层板结构中，可实现结构应变分布的在线监测。本文提出一种采用压电应变传感器阵列和人工神经网络模型的自诊断方法，对局部埋入压电应变传感元件的平板结构进行了分析，采用神经网络模型根据压电传感器组的输出识别结构承载位置和大小，对该方法进行了数值模拟和实验验证。     

Pore-scale network model for drainage-dominated three-phase flow in porous media

Drainage displacements in three-phase flow under strongly wetting conditions are completely described by a simple generalisation of well understood two-phase drainage mechanisms. As in two-phase flow, the sequence of throat invasions in three-phase flow is determined by fluid connectivity and threshold capillary pressure for the invading interface. Flow through wetting and intermediate spreading films is important in determining fluid recoveries and the progress of the displacement in three-phase flow. Viscous pressure drops associated with flow through films give rise to multiple filling and emptying of pores. A three-phase, two-dimensional network model based on the pore-scale fluid distributions and displacement mechanisms reported by Øren et al. and which accounts for flow through both wetting and intermediate fluid films is shown to correctly predict all the important characteristics of three-phase flow observed in glass micromodel experiments.     

Pore-Level Modeling of Gas and Condensate Flow in Two- and Three-Dimensional Pore Networks: Pore Size Distribution Effects on the Relative Permeability of Gas and Condensate

We present a mechanistic model of retrograde condensation processes in two- and three-dimensional capillary tube networks under gravitational forces. Condensate filling-emptying cycles in pore segments and gas connection–isolation cycles are included. With the pore-level distribution of gas and condensate in hand, we determine their corresponding relative permeabilities. Details of pore space and displacement are subsumed in pore conductances. Solving for the pressure field in each phase, we find a single effective conductance for each phase as a function of condensate saturation. Along with the effective conductance for the saturated network, the relative permeability for each phase is calculated. Our model porous media are two- and three-dimensional regular networks of pore segments with distributed size and square cross-section. With a Monte Carlo sampling we find the optimum network size to avoid size effects and then we investigate the effect of network dimensionality and pore size distribution on the relative permeabilities of gas and condensate.