基于贝叶斯理论的恢复力模型参数识别方法

提出了基于贝叶斯理论的恢复力模型参数识别方法,该方法考虑了模型误差的影响,结合实测滞回曲线数据,不仅可以得到模型参数的最有可能值,而且可以得到模型参数的定量的不确定性。以密肋复合墙体在低周反复荷载作用下所得滞回曲线为例,提出了可考虑刚度降低、捏拢滑移及极限荷载后强度降低现象的恢复力模型,建立了基于贝叶斯理论的恢复力模型参数识别计算框架,推导得到了模型参数的负对数似然函数,据此可得到模型参数的最有可能值及协方差矩阵。对标准密肋复合墙体预制试件和现浇试件的恢复力模型参数进行了识别,将根据模型参数最有可能值得到的滞回曲线及根据模型参数最有可能值及协方差矩阵得到的骨架曲线,与相应的实测值进行了对比,验证了所提方法的可行性及识别结果的合理性,更新的模型参数概率分布可用于后续的抗震风险评估。     

优化方法用于模型校核

在工程实际中，常常会遇到需要选择本构模型和校准参数的模型校核问题。文中采用优化方法实现了复杂模型校核的自动化。文中，首先将各种本构模型映射为整数编号，并根据这些编号来激活相应的本构模型和本构模型参数，解决了本构模型的参数化问题；另外，根据模型校核问题的特点选择了试验结果和计算结果相对误差的平方和作为目标函数。通过这些步骤，用实现了优化方法模型校核自动化。     

自由度缩聚的柔度损伤诊断法

为了提高结构在模型自由度缩聚情况下的损伤识别结果的精度，本文推导了基于改进Guyan 缩聚法的结构振动方程式．通过求解振动特征方程，利用其特征值和特征向量构建结构缩聚后的柔度矩阵表达式，并引入结构缩聚后的柔度曲率矩阵差和柔度曲率矩阵变化率两个损伤指标，将引入的新损伤指标应用于平面桁架的损伤识别．研究表明：不管是单损伤还是多损伤，仅仅需要一阶模态参数，利用其引入的新损伤指标就可以精确地识别出损伤杆单元位置．即使在高强度噪音的影响下，也保证了其损伤识别结果的精确性．验证了本文基于改进的Guyan 缩聚法推导出的损伤指标具有较好的损伤定位性能和较高的抗噪性能．     

A modified time domain subspace method for nonlinear identification based on nonlinear separation strategy

Nonlinear factors existing in engineering structures have drawn considerable attention, and nonlinear identification is a competent technique to understand the dynamic characteristics of nonlinear structures. Therefore, in this paper, a novel nonlinear separation subspace identification (NSSI) algorithm based on subspace algorithm and nonlinear separation strategy is proposed to conduct nonlinear parameter identification of nonlinear structures. For the proposed NSSI algorithm, the low-level excitation test is firstly conducted to obtain the transfer matrix in the linear response formula. Then, the obtained transfer matrix is used in the high-level excitation test to calculate the nonlinear response part by the proposed nonlinear separation strategy, and the subspace algorithm is utilized to identify the nonlinear parameter on the modified state-space model including only the nonlinear part. The proposed NSSI algorithm can reduce the coupling error caused by simultaneously processing both the large number part (corresponding to the linear part) and small number part (corresponding to the nonlinear part) in the traditional nonlinear subspace identification (NSI) algorithm. At last, two numerical experiments are given to validate the effectiveness of the developed novel nonlinear identification method. Furthermore, some influence factors are discussed to show the stability of the identification algorithm, and some comparisons between the proposed NSSI method and traditional NSI method are also conducted to demonstrate the advantages of the novel method.     

Instrumental variables approach to identification of a class of MIMO Wiener systems

A new approach to identification of multi-input multi-output (MIMO) Wiener systems using the instrumental variables method is presented. It is assumed that static nonlinear elements are invertible and their inverse characteristics can be expressed or approximated by polynomials of known orders. It is also assumed that the linear part of the Wiener system can be represented by a matrix polynomial form. Based on these assumptions, the Wiener system is transformed introducing a new parameterization and its parameters are estimated using a linear-in-parameters model. To solve the problem of non-consistency of least squares parameter estimates, an instrumental variables method is employed. A numerical example is included to show the effectiveness and the practical feasibility of the presented approach.     

界面连接刚度参数辨识的子结构分析法

以试验模态参数为基础,提出一种通过特征方程反问题辨识子结构界面连接刚度参数的子结构分析法。新方法以子结构动柔度矩阵特征方程为基础,建立求解界面结点内力和位移的方程,从而由子结构内部结点可测自由度上的位移用广义逆理论估计界面结点内力和位移。并通过迭代修正内部结点可测自由度上的试验值,以提高界面内力和位移的估计精度。最后通过连接子结构刚度矩阵建立的平衡方程求解相应的刚度参数。文中以太阳电池阵板间铰链副刚度参数辨识为例,将铰链副简化为两端结点各有6个自由度的弹簧连接元,考虑到自由度之间的耦合,推导了连接元的刚度矩阵。用上述方法辨识了铰链副6个自由度的刚度参数,得到满意的辨识结果。     

Identification of material parameters for inelastic constitutive models: Stochastic simulations for the analysis of deviations

For parameter identification a distance function between the measured and the simulated data has to be minimized. Therefore, the influence of three different norms used in the definition of such a distance function is investigated. The nonlinear optimization problem is solved using a modified random search algorithm originally proposed by Price (1978). Next a stochastic model for the generation of artificial test data is presented. This model is used for a stochastic simulation of test data (constant strain rate tension with relaxation and creep). From these artificial data the material parameters of the model of Chan, Bodner and Lindholm are identified. To measure the quality of the identified material parameters their mean values and empirical standard deviations are computed. Furthermore, the coefficients of the empirical correlation matrix for the material parameters are computed. The model responses for tensile tests with the parameter vector generated from all tests and with the estimated parameters (from stochastic simulations) differ not considerably. However, for the creep tests the different parameter estimations lead to quite different model responses. Received October 22, 1999     

平流层飞艇气动参数辨识

进入21世纪以来，随着科技的飞速发展，世界上掀起了研究和开发平流层平台的热潮。飞艇作为平流层平台可以实现无线通信、空间观测、大气测量以及军事侦查等目的。本文首先将飞艇所受的气动力分成由于来流速度产生的定常气动力和飞艇转动引起的非定常气动力两部分，通过理论分析建立了飞艇的气动力模型，从而得到需要辨识的气动参数。其次建立了以浮心为原点的六自由度非线性动力学模型和一种基于混合遗传算法的气动力系数辨识方法——混合遗传算法（遗传算法＋单纯型法）与极大似然法相结合的方法，并利用该方法对飞艇的气动参数进行辨识。通过仿真结果验证了该方法实用性和有效性。最后通过对气动参数的准确值与辨识值的分析比较，得出各个参数对飞艇运动性能的影响情况。     

改进Guyan～递推减缩技术

本文基于广义逆技术和逐级逼近思想,既建立了精度显著提高的改进Guyan减缩法,又提出了“物理型”近似减缩技术方面精度最好的准动态k级速推减缩公式,然而这些方法的执行时间却与其它“同一级”方法大抵相同．因此说,本方为工程界提供了至今综合效果最好的一类准动态模型减缩技术。     

多种群遗传算法在PBX本构模型参数识别中的应用

利用多种群并行结构对标准遗传算法SGA进行并行化处理,引入移民算子和精华种群形成多种群遗传算法MPGA,并设计了自适应交叉和变异概率对算法的收敛速度进行改进。结合ABAQUS软件和改进的多种群遗传算法,建立了材料本构模型参数识别方法。采用该方法对PBX炸药黏弹性损伤本构模型参数进行了模拟识别,并同基于标准遗传算法的参数识别方法进行了比较。结果证明,基于改进多种群遗传算法IMPGA的方法对克服算法未成熟收敛有显著的效果,识别结果更稳定。同时该方法的收敛速度更快,寻优能力更强,适合复杂非线性问题的优化,此方法可以被应用到其他材料本构模型的参数识别中。     

基于遗传算法的水轮机振动参数识别

通过对水轮机现场测试数据的频域分析，建立了测点位移响应的参数正模型。并基于遗传算法，建立了在时域内识别结构振动参数的数值方法，采用该方法对水轮机位移响应正模型的待定参数识别进行了研究。计算结果表明，所建立的基于遗传算法的参数识别方法具有良好的稳定性、抗观测噪声能力和较高的识别精度，能够反演出比较符合实际情况的位移响应振动模型。这就为荷载识别、结构优化和预报减震效果等后续工作提供了可靠的实施前提。     

Modal parameter identification based on ARMAV and state–space approaches

An accurate prediction for the response of civil and mechanical engineering structures subject to ambient excitation requires the information of dynamic properties of these structures including natural frequencies, damping ratios and mode shapes. Since the excitation force is not available as a measured signal, we need to develop techniques which are capable of accurately extracting the modal parameters from output-only data. This article presents the results of modal parameter identification using two time-domain methods as follows: the autoregressive moving average vector (ARMAV) method and the state–space method. These methods directly work with the recorded time signals and allow the analysis of structures where only the output is measured, while the input is unmeasured and unknown. The equivalence between ARMAV and state–space approaches for the problem of modal parameter identification of vibrating systems is shown in the article. Using only the singular value decomposition of a block Hankel matrix of sample covariances, it is shown that these two approaches give identical modal parameters in the case where the block Hankel matrix has full row rank. The time-domain modal identification algorithms have a serious problem of model order determination: when extracting structural modes these algorithms always generate spurious modes. A modal indicator to differentiate spurious and structural modes is presented. Numerical and experimental examples are given to show the effectiveness of the ARMAV or state–space approaches in modal parameter identification using response data only.     

Synchronization of unknown chaotic delayed competitive neural networks with different time scales based on adaptive control and parameter identification

In this paper, we investigate the synchronization problems of delayed competitive neural networks with different time scales and unknown parameters. A simple and robust adaptive controller is designed such that the response system can be synchronized with a drive system with unknown parameters by utilizing Lyapunov stability theory and parameter identification. Our synchronization criteria are easily verified and do not need to solve any linear matrix inequality. This research also demonstrates the effectiveness of application in secure communication. Numerical simulations are carried out to illustrate the main results.     

Entropy of microstructure

Two points are made in this paper: first, energy of random structures is not determined uniquely by any finite set of the characteristics of microstructure. The information lost is characterized by entropy of microstructure; it describes the scattering of the values of energy. Therefore, entropy of microstructure is a key thermodynamic parameter in phenomenological modeling of the behavior of random structures. Second, mathematical modeling of a random structure is based on the construction of its probabilistic measure; a way to select the probabilistic measure from the experimental data is outlined. The corresponding probabilistic measure is remarkably similar to that of classical statistical mechanics, though the underlying physics is quite different. After the probabilistic measure is chosen, the entropy of microstructure can be found from the analysis of the homogenization problem. Entropy of microstructure is computed in two example problems. Applications to phenomenological modeling of work hardening are discussed.     

Development of an approach to constitutive modelling of concrete: Isotropic damage coupled with plasticity

The paper presents an approach to constitutive modelling of concrete using damage mechanics and plasticity theory. The thermodynamic formulation, and parameter identification of a non-local coupled damage-plasticity model are presented in this study. The particular focus is the calibration of model parameters. It is shown that both the local parameters and the parameters governing the non-local interaction can be determined from experimental data reliably and consistently. A novel procedure is developed for parameter identification, using the separation of total dissipation energy into additive parts corresponding to different dissipation mechanisms. The relationship between the local and non-local parameters is also addressed, helping to obtain model responses consistent with the fracture energy of the material. The application of the model and the calibration procedure proposed in this study to the numerical failure analysis of concrete structures is illustrated through a series of real structural tests, showing both the performance of the model and the consistency of the proposed calibration procedure.     

Time domain parameters identification of foundation-structure interaction system

The time domain parameter laenuncauon memoa oi me iounuauon-structure interaction system is presented. On the basis of building the computation mode and the motion equation of the foundation-structure interaction system, the system parameter identification method was established by using the extended Kalman filter （EKF） technique and taking the unknown parameters in the system as the augment state variables. And the time parameter identification process of the foundation-structure interaction system was implemented by using the data of the layer foundation-storehouse interaction system model test on the large vibration platform. The computation result shows that the established parameter identification method can induce good parameter estimation.     

Method of reverberation ray matrix for static analysis of planar framed structures composed of anisotropic Timoshenko beam members

Based on the method of reverberation ray matrix(MRRM), a reverberation matrix for planar framed structures composed of anisotropic Timoshenko(T) beam members containing completely hinged joints is developed for static analysis of such structures.In the MRRM for dynamic analysis, amplitudes of arriving and departing waves for joints are chosen as unknown quantities. However, for the present case of static analysis, displacements and rotational angles at the ends of each beam member are directly considered as unknown quantities. The expressions for stiffness matrices for anisotropic beam members are developed. A corresponding reverberation matrix is derived analytically for exact and unified determination on the displacements and internal forces at both ends of each member and arbitrary cross sectional locations in the structure. Numerical examples are given and compared with the finite element method(FEM) results to validate the present model. The characteristic parameter analysis is performed to demonstrate accuracy of the present model with the T beam theory in contrast with errors in the usual model based on the Euler-Bernoulli(EB) beam theory. The resulting reverberation matrix can be used for exact calculation of anisotropic framed structures as well as for parameter analysis of geometrical and material properties of the framed structures.     

基于高次广义柔度灵敏度的结构损伤识别

柔度矩阵可以由结构的低价模态近似计算获得，因此被广泛用于结构的模型修正和损伤识别中。由普通柔度派生而来的广义柔度，可以由低价模态数据更加精确的获得，且随着广义柔度次数的增高其精度越高，往往只需要第一或二阶模态数据即可获得很准确的高次广义柔度。因此，广义柔度灵敏度方法自提出以来受到广泛关注。本文详细研究了基于高次广义柔度灵敏度的损伤识别计算方法，研究中发现，利用广义柔度灵敏度进行损伤识别计算时，并非越高次的广义柔度其识别结果越准确，随着广义柔度次数的增加，损伤识别结果精度呈现出先提高但随后显著降低的趋势。究其原因在于，虽然随着广义柔度次数的增加，广义柔度本身的精度更高，但与之相应的灵敏度方程组系数矩阵的条件数却也显著增大了，即方程组的病态性反而更加严重了，这导致了基于高次广义柔度计算所得的损伤参数的精度反而不如低次广义柔度的情况。因此，本文的研究表明，工程中利用广义柔度进行模型修正或损伤识别时，一般采用一次广义柔度或二次广义柔度即可，且计算中为了克服方程组的病态性和数据噪声的不利影响，本文提出了一种反馈奇异值截断法，能够明显提高计算精度，获得较准确的识别结果。     

基于不完备频响函数的结构损伤统计识别研究

基于不完备频响函数数据,结合概率统计方法,提出了一种能同时考虑模型参数不确定性和测试噪声影响的结构损伤统计识别方法。首先,基于频响函数某一行向量在不同频率下的幅值数据,利用矩阵拉直运算建立了关于损伤系数的确定性识别方程。其次,假设模型参数误差和测试噪声为零均值的高斯随机变量,根据摄动理论,推导了损伤后结构刚度参数的前二阶矩。随后,利用结构损伤前后的概率分布得到了结构损伤存在概率。最后,通过一个平面桁架结构模型验证了本文方法的有效性。数值算例的研究结果表明,损伤单元的损伤存在概率远大于非损伤单元;测试噪声对识别结果的影响比模型参数不确定性的影响更为显著。     

A strain-energy criterion for recognition of identified modes of continuous structural models

In structural modal analysis and modal testing, an important but difficult task is to match the identified natural frequencies and the corresponding modal deflections. This process is called the modal recognition in this paper. There were some treatments towards this problem for the lumped parameter structural models. For the distributed parameter models, however, little research has been reported on the modal recognition problem. In this paper, a strain-energy criterion for modal recognition has been developed. As an example, a distributed parameter model for a two-beam structural system has been formulated, which is expected to simulate the dynamics of a two-arm manipulating system fixed on a shuttle. Transfer matrix method has been used to set up the dynamic equation of the system. The natural frequencies are obtained from the solution of the characteristics equation. Consequently, the mode shape functions are found out analytically.

Strain energy can be viewed as a measure of the structural deformation. When performing modal analysis, we always assume that the structural system is vibrating at a particular natural frequency. The strain energy is, therefore, stored in the deflection caused by such a harmonic motion. The vibration at a particular natural frequency will not produce any strain energy in the other modal components. On the other hand, if a particular mode shape is contributed mostly by the deformation of a specific component of the global structural system, then the great percentage of the total strain energy will be stored in the deformation of that component. Based upon the calculation of the strain energy in the structural components we can find out which component is deformed most and in what motion it is deformed, thereby, the mode shape can be detected. The computer simulation demonstrated that the strain energy indicated an essentially perfect recognition of the identified natural frequencies with the corresponding mode shapes. The creation of the strain-energy criterion consummates the procedure of the distributed parameter modeling, modal identification and parameter estimation.