热膜式空气质量流量传感器动态非线性建模

热膜式空气质量流量传感器响应速度较快,测量范围宽,可靠性高,广泛应用于发动机空燃比的测量。但是,这种传感器的动态非线性问题影响了进气量测量的精度。采用基于多幅值阶跃信号的Hammerstein模型和两步辨识方法描述了传感器的特性。通过动态标定实验,得到不同幅值的阶跃输入、输出数据。根据幅值,用最小二乘拟合方法确定非线性系数,由输入数据和非线性系数得到静态非线性环节的输出。根据输出数据,运用线性自回归方法得到动态线性环节的传递函数。这样就建立起块联形式的动态非线性模型,有利于传感器动态非线性的校正。建模结果表明,动态非线性模型比线性模型可更准确地描述传感器的特性。     

基于BP神经网络的非线性分类器研究

BP神经网络作为非线性问题的良好解决方案,广泛应用于自动控制、模式识别、数据压缩等领域。本文基于多层BP神经网络分析了其在模式分类方面的特性,重点介绍了使用BP算法为核心的非线性分类器的性能和应用,并通过一个两类问题的实例具体评估了这种非线性分类器的实际性能。结果表明,采用BP神经网络的非线性分类器能够较好地处理简单非线性分类问题,但在识别精度、运算速度等方面仍然有待改善。     

非线性隔振系统动力学特性分析的FFT多谐波平衡法

鉴于非线性隔振系统响应的多谐波特性,单一谐波平衡法不能广泛适用于系统隔振性能评估。本文提出基于主导频率及多谐波的FFT多谐波平衡法,针对隔振系统的阶跃响应多解特性,倍周期和拟周期响应特性做了深入探讨。据此这些特性对非线性隔振传递率定义做了修正。达芬系统算例表明,采用改进方法求解的非线性运动响应及隔振传递率的准确率显著提高。此方法广泛适用于具有定常响应的非线性动力系统,为非线性系统隔振的理论分析提供有效工具。     

基于非线性状态估计（NSET）的手写体数字识别系统设计

手写体数字识别技术在财会、气象、邮政等领域有着广泛的应用，探究数字识别新方法并付诸实践，可以产生巨大的社会和经济效益。本文从传统的非线性状态估计（NonlinearStateEstimateTechnology，NSET）技术出发，将这种一般应用于设备运行状态监测领域的方法应用于手写体数字识别，并在美国邮政服务（UnitedStatesPostalService，USPS）手写数字识别库上进行测试。测试结果表明，该方法简单易行，运算效率高，对于解决非线性和高维模式识别问题凸显出其适应性强的特点。     

非线性梁压电分阶最优减振控制

提出非线性的分阶最优控制方法,并将其应用于梁的非线性振动压电减振控制。建立梁压电减振系统动力学模型,导出减振系统的非线性动力学运动微分方程,利用摄动法,实现非线性微分方程的线性化。将各阶线性方程解耦,化为状态空间方程。设计非线性分阶控制器,对减振系统进行分阶最优控制。仿真算例验证这种控制方法的有效性。     

光学薄膜非线性参数的快速测量与分析

提出一种测量非线性光学薄膜非线性参数的方法，测量了ＺｎＳｅ非晶态薄膜的非线性折射率和消光系数。在此方法中，实验数据的计算与处理由一套计算机软件完成。对这种测量方法的精确性及实用性进行了分析和讨论。     

隐式中点法对于非线性阻尼结构的稳定性

用能量守恒的方法证明了隐式中点法对于非线性指数阻尼的结构动力方程为数值稳定。工程中常用的双线性本构模型作为这种指数模型的特殊情况,同样满足数值稳定性条件。为了验证证明过程的可靠性,对一个单自由度体系和两个多自由度结构进行动力非线性计算分析,对比不同时间增量步的计算结果。从而给出了针对这种非线性动力方程计算的稳定的数值积分方法,为动力计算数值稳定性提供理论基础。     

一类非线性振动自适应控制的神经网络方法

针对一类非线性振动系统，本文详细阐述了非线性振动自适应控制的神经网络方法。首先提出一类非线性系统的控制模型与自适应控制策略；然后介绍了神经网络控制器的模型，进而导出了基于神经网络的振动自适应控制算法，数字仿真结果表明了这种方法的有效性。     

非线性光学材料聚二乙炔及其衍生物的制备及应用

非线性光学材料聚二乙炔及其衍生物的研究已引起国内外的广泛关注。本文总结了近年来聚二乙炔及其衍生物的合成方法 ,综述了聚二乙炔及其生物在非线性光学方面的研究进展 ,并且对聚二乙炔及其衍生物在非线性光学领域的进一步应用提出了展望。     

基于FPGA的非线性校正设计方案研究

OFDM非线性校正技术是现代通信系统广泛采用的调制方式，但其信号具有较高的信号峰均比而导致功率放大器HPA的非线性失真较为严重。本文简单介绍了常用的非线性校正方法，重点针对现有的方法本文提出了采用了基于FPGA非线性校正方案的实现。本方案具有集成度高、灵活性强、收敛速度快等优点。     

A new approach on vibration analysis of locally nonlinear stiffness and damping system

The nonlinear force induced by spring and damping of 2-degree-of-freedom locally nonlinear vibrating system is regarded as applied force, and its mathematical model is established in this paper. Then impulse response temporal method of linear vibrating system is applied in the system, the response of locally nonlinear vibrating system is obtained by convolution integration between unit impulse response of corresponding linear system and equivalent nonlinear force, and numerical simulation of the model is attained. Finally, the feasibility of the new method on the domain of locally nonlinear vibrating system is verified by comparing the results, which supplies a new method to solve approximately vibration response of locally nonlinear vibrating systems.     

Analysis of Power Magnetic Components With Nonlinear Static Hysteresis: Proper Orthogonal Decomposition and Model Reduction

We applied the proper orthogonal decomposition (POD) method to extract reduced-order models to efficiently solve nonlinear electromagnetic problems governed by Maxwell's equations with nonlinear hysteresis at low frequency (10 kHz), called static hysteresis, discretized by a finite-element method. We used a new domain-wall-motion hysteresis model for Power MAgnetic Components (POMACs) in the finite-element potential formulation via an efficient implicit-inverse model calculation. We propose a rational method for the selection of snapshots employed in the POD, used in conjunction with a fixed-point method for the solution of nonlinear POMAC problems. The reduced simulation time and great flexibility of the reduced-order models, as applied to nonlinear POMAC systems, suggest that the procedure can be applied to other electromagnetic problems with nonlinear hysteresis     

Fast Fourier nonlinear vibration analysis

We present an implementation of the multi-harmonic balance method (MHB) where intensive use of the Fast Fourier Transform algorithm (FFT) is made at all stages of calculations. The MHB method is not modified in essence, but computations are organized to obtain a very attractive method that can be applied systematically on general nonlinear vibration problems. The resulting nonlinear algebraic problem is solved by a particular implementation of a continuation method. Nonlinear vibration results are analyzed a posteriori by a Floquet method to determine their stability. The technique is applied on a series of problems of different nature, demonstrating the robustness and flexibility of the approach.     

预测复合材料非线性有效介电系数的割线方法

提出了预测复合材料非线性电位移和电场强度关系的一种解析方法,该方法基于各向材料的割线介电常数,将非线性问题转化成一系列线性问题来求解。该方法适用于任意各向异性复合材料和组分材料的非线性性质,而常用的Stroud和Hui的模型只适用于各向同性复合材料和组分材料的弱线性。证明了本文方法具有Ponte Castaneda提出的变分结构。计算结果表明,当基体非线性较小时,本文模型的预测与Stroud和Hui的模型一致,但当基体非线性系数增大时,本文模型能给出合理的预测结果,而Stroud和Hui的模型则会超出基体和夹杂的性能范围。      

基于GNAR模型和Itakura距离的结构非线性损伤识别方法

为了有效地解决时域非线性损伤的识别问题,提出了基于线性/非线性自回归一般表达式模型(general expression for linear and nonlinear autoregressive model,GNAR模型)和Itakura距离的损伤识别方法。描述了GNAR模型的基本理论,给出了基本的模型定阶和参数估计方法。采用剪枝算法对GNAR模型结构进行了优化选取,并提出了以Itakura距离作为损伤指标的非线性损伤识别方法。采用3层框架非线性损伤实验验证了该方法的有效性,并将提出的方法运用于输电塔钢架模型的非线性损伤识别实验中。结果表明:提出的结构非线性损伤识别方法对框架和输电塔钢架结构的非线性损伤识别效果良好,且环境变化对其识别结果影响较小,由该方法计算得到的损伤层损伤概率明显大于未损伤层,有利于高效地确定非线性损伤源的位置。     

Applications of a reduction method for reanalysis to nonlinear dynamic analysis of framed structures

 This paper is concerned with the nonlinear dynamic analysis of framed structures using a reduction method recently proposed by the authors. The reduction method is originally devised for structural static reanalysis and has been applied in optimal design of structures to speed up the design process. For nonlinear dynamic analysis of framed structures, the incremental or iterative equations of motion can be transformed into an algebraic system of equations if appropriate integration methods such as Newmark's method are used to integrate the equations of motion. The resulting algebraic system, referred to as the effective system in this paper, changes during the simulation for a nonlinear dynamic problem. Therefore, from the point of view of solving systems of equations, a nonlinear dynamic problem is very similar to an optimal design problem in that the system of equations changes for both types of problems. Hence, any reanalysis technique can be readily applied to carry out a nonlinear dynamic analysis of structures. As demonstrated from the presented numerical examples, the response obtained by the adopted reduction method is as accurate as that obtained by the Cholesky method, and as estimated from the operation counts involved in the method, it is more efficient than the Cholesky method when the half-band width is greater than about 50. Received 23 March 2000     

Homotopy method of fundamental solutions for solving certain nonlinear partial differential equations

In this study, the homotopy analysis method (HAM) is combined with the method of fundamental solutions (MFS) and the augmented polyharmonic spline (APS) to solve certain nonlinear partial differential equations (PDE). The method of fundamental solutions with high-order augmented polyharmonic spline (MFS–APS) is a very accurate meshless numerical method which is capable of solving inhomogeneous PDEs if the fundamental solution and the analytical particular solutions of the APS associated with the considered operator are known. In the solution procedure, the HAM is applied to convert the considered nonlinear PDEs into a hierarchy of linear inhomogeneous PDEs, which can be sequentially solved by the MFS–APS. In order to solve strongly nonlinear problems, two auxiliary parameters are introduced to ensure the convergence of the HAM. Therefore, the homotopy method of fundamental solutions can be applied to solve problems of strongly nonlinear PDEs, including even those whose governing equation and boundary conditions do not contain any linear terms. Therefore, it can greatly enlarge the application areas of the MFS. Several numerical experiments were carried out to validate the proposed method.     

非线性参数结构系统的参数识别

本对基于时域信息的非线性参数系统参数识别问题进行了研究。首先提出了一种非线性参数系统的分类方法。然后对非线性参数系统的参数化问题进行了讨论：本提出用NEWTON RAPHSON方法进行非线性参数系统识别方程的求解，并用埃特金加速法提高迭代收敛的速度。章的最后以剪力墙参数的识别问题为例演示了本所提方法的有效性。数值算例表明，识别结果的精度和识别耗时对参数初值的选取并不敏感。     

A fast implementation of explicit time-stepping algorithms with the finite element method for a class of nonlinear evolution problems

In this paper a class of nonlinear evolution problems is considered. It is shown that, under special conditions, the application of the product approximation method for nonlinear problems in the finite element method results in constant (i.e. time-independent) matrices. In those cases the amount of computing required to solve these equations with an explicit time-stepping algorithm is decreased considerably compared to the standard Galerkin formulation in which the matrices are time-dependent. The method is applied to two practical two-dimensional problems: the shallow water equations and a nonlinear heat conduction problem.     

Application of the proper orthogonal decomposition for linear and nonlinear structures under transient excitations

Model reduction has become very important in order to save calculation time. In particular, in structural dynamics, computations become very time-consuming when the critical time step of explicit integrators becomes very small. The main focus of this paper is on the application of the Proper Orthogonal Decomposition (POD) method to a structure subjected to transient earthquake loading. It is shown that based on the information of only a small portion of the transient excitation and the structure (“snapshots”), it is possible to assemble a reduced-order model, which yields a very accurate and time-saving approximation of the response to the entire earthquake. The POD reduction method is applied not only to linear, but also to nonlinear structures under earthquake loading. In the linear case, the POD results can be compared to those obtained by the classical method of modal truncation. In the nonlinear case, base isolation systems (friction pendulum systems) are integrated in the structure. Error estimations are applied in order to assess the solution of the POD-reduced system of the linear and the nonlinear systems. The POD can be applied successfully if the snapshots within the chosen time interval describe the main behavior of the system well. In both the linear and nonlinear cases, the approximation of the system as reduced by the POD is very accurate even if only a few POD modes are used. The advantage over the method of Modal Truncation is not only the optimality of the POD modes concerning their associated energy, but also its applicability to nonlinear systems.