基于递归最小二乘支持向量机的动态系统建模研究

针对递归最小二乘支持向量机的递归性易导致建模中偏微分方程组求解困难的问题,提出用解析法求解偏微分方程组,实现了完整的递归最小二乘支持向量机模型．首先分析了各参数的相关性,然后推导出偏微分方程的解析表达式并求解．仿真实例表明,在动态系统建模中,该模型的性能比常用的串并联模型以及现有不完整递归最小二乘支持向量机模型的精度更高、性能更好．     

原子力显微镜中微悬臂梁分布参数系统的Hammerstein模型

为提高原子力显微镜(atomic force microscope,AFM)中微悬臂梁分布参数模型的精度,本文提出了包含非线性时空特性的改进模型,在此基础上简化控制器的结构.首先加入非线性补偿项修正传统分布参数模型;然后采用Karhunen-Loève(K–L)方法提取系统主导空间基函数,实现系统输出的时空变量分离.利用求解得到的时间系数和系统激励,建立系统时域Hammerstein模型,使系统无限维偏微分方程模型转化为时域有限维常微分方程形式,控制器的设计无需考虑空间耦合的影响;最后,利用最小二乘支持向量机结合奇异值分解法辨识模型中的参数.与传统分布参数模型进行仿真和实验结果比较,验证了方法的有效性.     

一种基于Pade近似的频域辨识与频域模型降阶新方法

研究了基于积分最小二乘指标的SISO时滞系统频域辨识与频域模型降阶问题.通过采用Pade近似,将积分最小二乘指标推广到可以处理时滞系统的情形.深入分析了Pade近似所引入误差,揭示了采用Pade近似的可行性与有效性.所提出的方法能够用最小二乘类算法高效求解,无须困难的非线性优化.仿真验证了所提出方法的有效性.     

基于递归最小二乘支持向量机的动态系统建模研究

针对递归最小二乘支持向量机的递归性易导致建模中偏微分方程组求解困难的问题,提出用解析法求解偏微分方程组,实现了完整的递归最小二乘支持向量机模型．首先分析了各参数的相关性,然后推导出偏微分方程的解析表达式并求解．仿真实例表明,在动态系统建模中,该模型的性能比常用的串并联模型以及现有不完整递归最小二乘支持向量机模型的精度更高、性能更好．

     

一类非线性逆系统的加权最小二乘支持向量机辨识方法

文中依据T-S模型的思想,提出了一种加权最小二乘支持向量机辨识算法.它采用模糊c均值(FCM)聚类确定规则数目,通过Gauss型函数将原输入输出空间分成若干子空间,在子空间中使用最小二乘支持向量机(LS-SVM)拟合获得子模型,然后由一个权重机制合成这些子模型,得到系统的模型.文中使用该方法去辨识关键反馈变量难以获得的非线性逆系统.为了得到这类逆系统的有效建模数据,采用了联合逆系统方法.仿真结果表明,加权最小二乘支持向量机辨识方法是有效的,它能够实现这类非线性逆系统的辨识,而且拟合误差平稳,波动幅度小,拟合精度和泛化能力都较好.     

基于支持向量机的迟滞系统建模及性能研究

迟滞系统广泛存在于各工程领域,但由于迟滞非线性系统的不确定性、状态不可测等特性,因此迟滞系统在建模方面存在一定的困难。针对上述问题,提出了一种采用最小二乘支持向量回归机的解决方案,对系统进行建模方法的研究,并利用粒子群算法、量子粒子群算法等对最小二乘支持向量机中的惩罚参数γ和核函数参数σ的组合进行优化,以提高模型性能及泛化能力。仿真结果表明,利用粒子群优化算法的最小二乘支持向量回归机对迟滞系统的模型仿真可以得到较好的结果。     

基于MATLAB的最小二乘法参数辨识与仿真

本文介绍了基于MATLAB/Simulink的使用最小二乘法进行参数辨识的设计与仿真方法.首先简述参数辨识的概念和最小二乘法的基本原理,然后介绍如何采用Simulink建立系统的仿真对象模型和运用MATLAB的M语言编写最小二乘递推算法,最后结合实例给出相应的仿真结果和分析.本文的仿真方法克服了传统编程语言仿真时繁杂、难度高、周期长的缺点.     

基于正交函数的立式淬火炉控制系统参数辨识

大型立式淬火炉体积庞大,工况复杂,炉内温度分布呈本征非均匀性.为了获得温度控制高精度和高均匀性提出参数辨识算法,包括求解正交函数正、反向积分运算矩阵,以块脉冲函数为基函数利用正交函数变换将由偏微分方程描述的分布参数系统模型转化为最小二乘形式的代数方程.辨识过程中考虑了大型立式淬火炉温度分布参数系统模型边界条件和初始条件的影响,提高了参数辨识精度,算法计算量小且保持了系统的空间分布特性.     

利用递推增广矩阵辨识加速度计的参数

介绍了加速度计的基本工作原理和结构模型,将加速度计的微分方程转换成差分方程,在传统最小二乘法辨识的基础上,采用递推增广矩阵的辨识方法对加速度计的参数模型进行辨识.通过Matlab对其仿真,得到被辨识参数的估计值与曲线图,说明采用递推增广矩阵辨识方法辨识系统参数具有辨识速度快、辨识精度高、辨识结果准确等特点.     

随机连续系统的连续时间最小二乘辨识的数值实现

讨论了随机连续系统的连续时间最小二乘（ＣＴＬＳ）辨识的数值实现及仿真，首先回顾了随机连续系统的ＣＴＬＳ辨识法和理论分析结果，然后基于数值积分技术和求解常微分方程的数值解欧拉法和龙格－库塔法，给出了ＣＴＬＳ法的两种数值实现方法，仿真结果显示出此方法的有效性。     

基于Matlab的板式换热器动态特性建模与仿真

为了正确预测板式换热器的动态特性，在合理假设的基础上，根据流道和换热平板的质量、能量守恒方程，建立了无量纲动态仿真数学模型。考虑了流体沿流动方向的导热扩散特性、换热平板的金属蓄热以及沿径向的导热对出口温度瞬态响应的影响。基于Matlab的数值计算基础，对无量纲动态仿真数学模型的空间维变量的一阶导数项采用向后差分，二阶导数项采用中心差分，然后采用Matlab的ODE求解器进行求解，得到了阶跃扰动和频率扰动下的出口温度的响应曲线。     

一种电热式MEMS微镜多自由度模型解析方法

结合微镜驱动器结构,通过引入热导和弹簧刚体模型分析了多驱动器间热传导和相互作用力导致的形变原因,并建立了多自由度微镜的数学模型.仿真与实验数据对比说明：建立的模型与实验结果具有较好的一致性,能够预测4个驱动器的输入响应特性.     

Rapid Simulation of Laser Processing of Discrete Particulate Materials

The objective of this paper is to develop a computational model and corresponding solution algorithm to enable rapid simulation of laser processing and subsequent targeted zonal heating of materials composed of packed, discrete, particles. Because of the complex microstructure, containing gaps and interfaces, this type of system is extremely difficult to simulate using continuum-based methods, such as the Finite Difference Time Domain Method or the Finite Element Method. The computationally-amenable model that is developed captures the primary physical events, such as reflection and absorption of optical energy, conversion into heat, thermal conduction through the microstructure and possible phase transformations. Specifically, the features of the computational model are (1) a discretization of a concentrated laser beam into rays, (2) a discrete element representation of the particulate material microstructure and (3) a discrete element transient heat transfer model that accounts for optical (laser) energy propagation (reflection and absorption), its conversion into heat, the subsequent conduction of heat and phase transformations involving possible melting and vaporization. A discrete ray-tracking algorithm is developed, along with an embedded, staggered, iterative solution scheme, which is needed to calculate the optical-to-thermal conversion, particle-to-particle conduction and phase-transformations, implicitly. Numerical examples are given, focusing on concentrated laser beams and the effects of surrounding material conductivity, which draws heat away from the laser contact zone, thus affecting the targeted material state.     

Anti-diffusive methods for hyperbolic heat transfer

The hyperbolic heat transfer equation is a model used to replace the Fourier heat conduction for heat transfer of extremely short time duration or at very low temperature. Unlike the Fourier heat conduction, in which heat energy is transferred by diffusion, thermal energy is transferred as wave propagation at a finite speed in the hyperbolic heat transfer model. Therefore methods accurate for Fourier heat conduction may not be suitable for hyperbolic heat transfer. In this paper, we present two anti-diffusive methods, a second-order TVD-based scheme and a fifth-order WENO-based scheme, to solve the hyperbolic heat transfer equation and extend them to two-dimension, including a nonlinear application caused by temperature-dependent thermal conductivity. Several numerical examples are applied to validate the methods. The current solution is compared in one-dimension with the analytical one as well as the one obtained from a high-resolution TVD scheme. Numerical results indicate that the fifth-order anti-diffusive method is more accurate than the high-resolution TVD scheme and the second-order anti-diffusive method in solving the hyperbolic heat transfer equation.     

热色液晶瞬态测量全表面换热系数的技术研究

建立了基于瞬态实验技术的风洞和测试系统,用HSV色彩模型对颜色进行描述,得到颜色-温度关系;用半无限大导热理论对瞬态导热过程进行换热系数求解.测量了平板表面的换热系数分布,并与理论解进行了对比,两者相吻合.     

A Novel Boundary Control Solution for Unstable Heat Conduction Systems Based on Active Disturbance Rejection Control

In this study, a new boundary control scheme is proposed for a class of unstable heat conduction systems based on active disturbance rejection control and frequency domain analysis methods. The unstable heat conduction systems studied here corresponds to a thin rod with surface heat loss and internal heat generation. An active disturbance rejection boundary controller is designed to stabilize the unstable heat conduction systems. Using the frequency domain analysis method, the Nyquist stability criterion for distributed parameter systems, the relative stability indices (gain margin, phase margin, and exponential stability speed) can be obtained for performance evaluation, and a constructive parameter‐tuning method for controller is proposed. Finally, it is proved that an active disturbance rejection boundary controller can stabilize the unstable heat conduction systems with one unstable pole, and the simulation results demonstrate that the temperature profile of the whole rod has good convergence properties under both Dirichlet and Neumann boundary control.     

Measurement technique for the thermal properties of thin-film diaphragms embedded in calorimetric flow sensors

In diaphragm-based micromachined calorimetric flow sensors, convective heat transfer through the test fluid competes with the spurious heat shunt induced by the thin-film diaphragm where heating and temperature sensing elements are embedded. Consequently, accurate knowledge of thermal conductivity, thermal diffusivity, and emissivity of the diaphragm is mandatory for design, simulation, optimization, and characterization of such devices. However, these parameters can differ considerably from those stated for bulk material and they typically depend on the production process. We developed a novel technique to extract the thermal thin-film properties directly from measurements carried out on calorimetric flow sensors. Here, the heat transfer frequency response from the heater to the spatially separated temperature sensors is measured and compared to a theoretically obtained relationship arising from an extensive two-dimensional analytical model. The model covers the heat generation by the resistive heater, the heat conduction within the diaphragm, the radiation loss at the diaphragm’s surface, and the heat sink caused by the supporting silicon frame. This contribution summarizes the analytical heat transfer analysis in the microstructure and its verification by a computer numerical model, the measurement setup, and the associated thermal parameter extraction procedure. Furthermore, we report on measurement results for the thermal conductivity, thermal diffusivity, and effective emissivity obtained from calorimetric flow sensor specimens featuring dielectric thin-film diaphragms made of plasma enhanced chemical vapor deposition silicon nitride.     

Finite element modeling of multi-pass welding and shaped metal deposition processes

This paper describes the formulation adopted for the numerical simulation of the shaped metal deposition process (SMD) and the experimental work carried out at ITP Industry to calibrate and validate the proposed model. The SMD process is a novel manufacturing technology, similar to the multi-pass welding used for building features such as lugs and flanges on fabricated components (see Fig. 1a and b). A fully coupled thermo-mechanical solution is adopted including phase-change phenomena defined in terms of both latent heat release and shrinkage effects. Temperature evolution as well as residual stresses and distortions, due to the successive welding layers deposited, are accurately simulated coupling the heat transfer and the mechanical analysis. The material behavior is characterized by a thermo-elasto-viscoplastic constitutive model coupled with a metallurgical model. Nickel super-alloy 718 is the target material of this work. Both heat convection and heat radiation models are introduced to dissipate heat through the boundaries of the component. An in-house coupled FE software is used to deal with the numerical simulation and an ad-hoc activation methodology is formulated to simulate the deposition of the different layers of filler material. Difficulties and simplifying hypotheses are discussed. Thermo-mechanical results are presented in terms of both temperature evolution and distortions, and compared with the experimental data obtained at the SMD laboratory of ITP.     

Numerical solution of integrodifferential heat conduction equation for a nonlocal medium

A heat conduction model is proposed based on the relations of rational thermodynamics of irreversible processes, taking into account the nonlocal nature of the medium and the finite speed of heat propagation. In the one-dimensional case, the numerical solution of the integral-differential equation of heat conduction is obtained by the finite-element method.     

A system-level model for a silicon thermal flow sensor

A system-level model with lumped parameters for a thermal flow sensor is presented. The model is built with 13 circuit cells consisting of thermal resistors and thermal capacitors in SPICE. The circuit cell originates from the heat conduction equation using the Finite Differential Method, including the 2-D thermal conduction cell, the convection cell, and the thermal capacity in the chip. Based on the thermal model of the flow sensor, the 2-D temperature distribution of the chip can be calculated with SPICE in both the constant power mode (CP) and constant temperature difference mode (CTD). As an example, the system level model of the thermal anemometer in the CTD mode was established in PSPICE. Wind tunnel test was carried out to verify the system model, and show a reasonable agreement with the simulation results, with an error less than 8%.