非线性二维导热反问题的混沌-正则化混合解法

考虑热传导系数随温度变化,建立了非线性二维稳态导热反问题数值计算模型。并把混沌优化方法和梯度正则化方法相结合,构成一种混沌-正则化混合算法求该计算模型的全局解。以热传导系数随温度线性变化为例,由布置在结构边界上的观测点温度信息确定了结构材料热传导系数及其随温度变化规律。结果表明混合算法计算结果与初值无关,具有很好的全局寻优性能,而且计算量远比经典遗传算法和单纯采用混沌优化方法小。     

二维瞬态热传导的PDDO分析北大核心CSCD

采用近场动力学微分算子(peridynamic differential operator, PDDO)理论求解了二维瞬态热传导问题.将热传导方程和边界条件由其局部微分形式重构为非局部积分形式,引入Lagrange乘数法,采用变分原理的概念,建立了二维瞬态热传导问题的非局部分析模型.通过误差与收敛性分析,与其他数值方法计算结果进行比较,验证了本模型的准确性.在此基础上,将本模型应用于计算不规则边界板和内部含微缺陷(裂纹和圆孔)板的二维瞬态热传导问题.结果表明该方法计算精度高、适用范围广、具有较好的收敛性,为计算二维瞬态热传导问题提供了新的思路.     

热防护服-空气-皮肤热传导模型及其解析解

建立了高温环境下热防护服-空气-皮肤的热传导模型.利用热传导时,层合界面间温度相等和热流量连续的条件,结合微分思想,用分离变量法推导了微小时间段内模型热传导的解析解,然后通过循环得到整个时域内的解析解.利用求得的解析解分析了在80 ℃的环境温度下模型各位置温度和热流密度的变化情况,以及在不同环境温度下皮肤表面温度变化和热损伤情况.该求解方法可用来分析一般层合结构传热问题,计算结果对热防护服的设计和效果评价具有一定的参考意义.     

瞬态热传导的奇异边界法及其MATLAB实现

基于动力学问题时间依赖基本解的奇异边界法是一种无网格边界配点法.该方法引入源点强度因子的概念从而避免了基本解的源点奇异性,具有数学简单、编程容易、精度高等优点.将该方法用于瞬态热传导问题的数值模拟,运用MATLAB实现该问题的数值研究,并创建相应的MATLAB工具箱.针对二维和三维瞬态热传导问题,进行了基于反插值技术和经验公式的奇异边界法MATLAB算例实现.针对支撑圆坯低温瞬态温度场的模拟结果表明,瞬态热传导奇异边界法的MATLAB工具箱具有简单、方便、精确可靠的优点.研究成果有助于发展瞬态热传导的奇异边界法,并为瞬态热传导问题的数值分析和仿真提供了一种简单高效的模拟工具.     

在模糊随机因素影响下的多层圆筒结构的稳态热传导问题的区间数解法

在多层圆筒结构稳态热传导分析中,根据给定固体壁两侧表面温度总传热量公式,首先推导出当边界温度为随机变量情况下总传热量函数统计参数的均值和方差;然后推导出在导热系数为模糊数,边界温度为随机数下的总传热量的区间表达式．通过比较可以知道由区间数算法得到的区间最大,由概率统计算法得到的区间最小．并给出了两者的相对误差公式．最后引进粗糙集中的上、下近似集,提出用一个参数来统一定义模糊和随机区间进行稳态结构的热传导分析．     

城市火灾次数与气象因素的Adaptive-Lasso分析

为研究城市火灾次数与气象因素的关系,以天津市的火灾数据为例,建立月度火灾次数与温度、风速、降雨量、日照、湿度等相关的各月气象因素及上月火灾起数间的线性回归模型.通过Adaptive-Lasso方法对上述变量进行选择并估计其参数,并对所得模型进行了分析比较,同时预测分析了2008年上半年火灾发生数.研究表明,通过Adaptive-Lasso方法建立的线性回归模型能够更好地预测火灾次数.     

四段式有限元法分析热耦合的流体-固体相互作用问题

给出了一种流(体)-热-结构综合的分析方法,固体中的热传导耦合了粘性流体中的热对流,因而在固体中产生热应力.应用四段式有限元法和流线逆风Petrov-Galerkin法分析热粘性流动,应用Galerkin法分析固体中的热传导和热应力.应用二阶半隐式Crank-Nicolson格式对时间积分,提高了非线性方程线性化后的计算效率.为了简化所有有限元公式,采用3节点的三角形单元,对所有的变量:流体的速度分量、压力、温度和固体的位移,使用同阶次的插值函数.这样做的主要优点是,使流体-固体介面处的热传导连接成一体.数个测试问题的结果表明,这种有限元法是有效的,且能加深对流(体)-热-结构相互作用现象的理解.     

可变形孔隙介质中热、水耦合力学问题的代数多格子分析方法

研究了孔隙介质中包括热和质量传递的全耦合多相流问题的代数多格子分析方法。数学模型包括质量、线性矩、能量平衡方程和本构方程,以位移、毛细压力、汽压和温度为基本变量,模型中采用了考虑毛细压力关系的修正有效应力概念,并考虑相变、热传导、对流和潜热交换(汽化-冷凝),气相是由易混合的干空气和水蒸气组成,视为理想气体。考题显示出较高的计算效率。     

热传导(对流-扩散)方程源项识别的粒子群优化算法

提出了利用粒子群优化(PSO)算法反演热传导方程与对流-扩散方程源项的一种新方法,在已有文献方法的基础上,求解出这两类方程正问题的解析解,再把源项识别问题转化为最优化问题,结合粒子群优化算法寻优求解.通过数值模拟与统计检验,结果表明,此方法可快速有效地实现热传导方程与对流-扩散方程源项的识别,并可推广应用到其它数学物理方程的源项或参数的反演识别.     

金属材料热处理过程中的瞬态温度场与解态相变的数值计算方法

本文首先将Kirchhoff变换推广到导热系数为温度的多项式的非定常非线性热传导问题.并用分析方法确定热传导问题的边界条件.其次提出以孕育期叠加法并引用线性混合法则来模拟金属热处理过程的多相瞬态相变,较为简便地确定相变的开始时间、相变的种类及相变组织的数量.最后利用三维双重边界元法分析工件多种形式的热处理全过程.算例的数值计算结果表明本文方法行之有效.     

主动冷却点阵夹层防热结构温度响应计算模型北大核心CSCD

针对点阵夹层结构主动热防护问题,建立了夹层结构面板和芯体导热与冷却剂对流耦合的非稳态传热理论模型,利用有限体积法离散控制方程并在MATLAB中进行了迭代求解.模型首次考虑了面板与夹芯杆之间的收缩热阻,并利用分离变量法得到了收缩热阻的近似解析解.基于单胞模型和周期性边界条件,模拟得到了模型所需的表面对流传热系数h_(b)和h_(fin).最后,选取多单胞计算工况进行数值模拟和理论模型对比,并讨论了收缩热阻对模型预测精度的影响.结果表明:理论模型能够准确预测夹层结构及内部流体的温度变化,理论与仿真之间的最大误差不超过1%;随着外加热流密度不断增大,忽略收缩热阻使得计算结果造成的误差不断增大;与数值模拟相比,理论模型可显著地减少计算时间并节省计算资源,尤其适用于非均匀、非稳态复杂热载荷下点阵夹层结构的温度响应计算.     

Coupled boundary element method and finite difference method for the heat conduction in laser processing

This paper is presented as a way to model transient heat conduction in a 3-D axisymmetric case where large rates of heat fluxes are applied on the surfaces as done in the case of laser processing. This would result in large temperature gradients in a small area irradiated by the laser on the incident surface that could also reach melting and subsequent vaporization. BEM can handle large fluxes very easily and it also can be formulated if needed to incorporate the moving boundary problem in a unique manner while on the other hand FDM is a fast and efficient method. For these reasons a coupled BEM–FDM method is formulated to simulate the heat conduction process. In the BEM method linear elements for the boundary and quadratic elements for the domain were used. The integrals in BEM were integrated in time using the asymptotic expansion for the modified Bessel functions in the Green’s function. To further improve the accuracy, special techniques were employed in the spatial integration. As for the FDM formulation, a flux conservation scheme with a 4th order formula for the fluxes was used. The FDM and BEM were coupled at the interface by the temperature from the FDM formulation being imposed on the BEM and the flux from the BEM being utilized by the FDM elements near to the interface. To advance in time, the Crank–Nicholson scheme was used on the FDM directly and due to coupling indirectly on the BEM. The relative errors for the simulation of constant and variable flux cases demonstrate the successful nature of the numerical model.     

Spatial decay estimates for the Maxwell-Cattaneo equations with mixed boundary conditions

In this paper the authors derive spatial decay bounds for the temperature and heat flux as defined by the Generalized Maxwell-Cattaneo equations for heat conduction in a semi-infinite cylinder when the temperature and the tangential components of the heat flux vector vanish on the lateral surface of the cylinder. The results here supplement those previously found by the authors [5] when the heat flux vector was assumed to be zero on the lateral surface but no condition was imposed on the temperature.     

Spatial decay estimates for the Maxwell-Cattaneo equations with mixed boundary conditions

In this paper the authors derive spatial decay bounds for the temperature and heat flux as defined by the Generalized Maxwell-Cattaneo equations for heat conduction in a semi-infinite cylinder when the temperature and the tangential components of the heat flux vector vanish on the lateral surface of the cylinder. The results here supplement those previously found by the authors [5] when the heat flux vector was assumed to be zero on the lateral surface but no condition was imposed on the temperature.Received: February 7, 2002; revised: June 3, 2002     

Frequency-domain regression method for estimating CTF models of building multilayer constructions

The conduction transfer function (CTF) models of heat conduction through building constructions are widely used in building energy analysis and heating, ventilation, and air conditioning (HVAC) system design and simulation. A frequency-domain regression (FDR) method is developed to estimate the CTF model of a building construction from its theoretical response characteristics in this paper. First, the theoretical response characteristics are calculated simply by numerical matrix multiplication. The CTF coefficients are then obtained by simply solving a set of linear equations. The tests and comparisons have shown that CTF models obtained by the FDR method are completely equivalent to those found by the methods currently available. The FDR method can provide a short series of CTF coefficients, and the models by this method have a high accuracy in calculating heat gain/loss through building constructions. The FDR method is very simple and straightforward, has high computational speeds, and needs less computational efforts. Therefore, the FDR method is much easier and convenient to use and implement in building energy analysis and HVAC system simulation programs.     

Simulation of thermal disturbances with finite wave speeds using a high order method

The classical heat diffusion theory based on the Fourier’s model breaks down when considering transient heat flow, for short times, extreme thermal gradients or at low temperatures. The hyperbolic heat conduction equation based on the Cattaneo model for the heat flux incorporates a relaxation mechanism in order to gradually adjust to a change in the temperature gradient. A spectral element method is applied for solving the hyperbolic system treating the heat flux as an independent variable in addition to temperature. The numerical solution is based on the time-space least squares spectral method. Numerical examples are included for discussing the effects of the thermal waves.     

Numerical modeling for compartment fire environment under a solid-cone water spray

A mathematical model was developed for simulating the fire environment of a compartment under the action of a solid-cone water spray such as those discharged from a water mist fire suppression system. A smoke layer was induced by a fire in the compartment. The solid-cone water spray was discharged to act on the smoke layer, but not on the burning object. Under this condition of having a stable smoke layer, the compartment was divided into three regions. Region 1 is the upper hot smoke layer, Region 2 is the lower cool air layer and Region 3 is the solid-cone spray. The effects on the smoke layer development due to spray-induced flow were considered on the basis of mass, momentum and heat conservation. Water droplets of the solid-cone spray were divided into four typical classes based on the droplet distribution function.

The parameters including the smoke layer interface height, smoke temperature and air temperature, smoke flow rate through the opening and oxygen concentration in the air layer were investigated under various heat release rates, water application rates and volume mean diameters of the solid-cone spray. Effective hot gas entrainment and water vapor production suggested that the water spray should contain a variety of droplet size. In this way, a compartment fire can be controlled effectively through indirect interaction such as oxygen concentration depletion.     

A superstatistical model for anomalous heat conduction and diffusion

We propose a superstatistical model for anomalous heat conduction and diffusion, which is formulated by the thermal conductivity distribution, overall temperature and heat flux distributions. Our model obeys Fourier's law and the continuity equation at the individual level. The evolution of the thermal conductivity distribution is described by an advection-diffusion equation. We show that the superstatistical model predict anomalous behaviors including the time-dependent effective thermal conductivity and slow long-time asymptotics. The time-dependence of the effective thermal conductivity is determined by the mean square displacement (MSD), which coincides with existing investigations. The superstatistical structure can also be extended into other non-Fourier models including the Cattaneo and fractional-order heat conduction models.     

An inverse hyperbolic heat conduction problem in estimating surface heat flux of a living skin tissue

In this study, an inverse algorithm based on the conjugate gradient method and the discrepancy principle is applied to solve the inverse hyperbolic heat conduction problem in estimating the unknown time-dependent surface heat flux in a living skin tissue from the temperature measurements taken within the tissue. The inverse solutions will be justified based on the numerical experiments in which three different heat flux distributions are to be determined. The temperature data obtained from the direct problem are used to simulate the temperature measurements. The influence of measurement errors upon the precision of the estimated results is also investigated. Results show that an excellent estimation on the time-dependent surface heat flux can be obtained for the test cases considered in this study.