Parameter estimation of an aeroelastic aircraft using neural networks

Application of neural networks to the problem of aerodynamic modelling and parameter estimation for aeroelastic aircraft is addressed. A neural model capable of predicting generalized force and moment coefficients using measured motion and control variables only, without any need for conventional normal elastic variables or their time derivatives, is proposed. Furthermore, it is shown that such a neural model can be used to extract equivalent stability and control derivatives of a flexible aircraft. Results are presented for aircraft with different levels of flexibility to demonstrate the utility of the neural approach for both modelling and estimation of parameters.     

Design of a diffractive optical element for pattern formation in a bilingual virtual keyboard

Pattern formation is one of the many applications of diffractive optical elements (DOEs) for display. Since DOEs have lightweight and slim nature compared to other optical devices, using them as image projection device in virtual keyboards is suggested. In this paper, we present an approach to designing elements that produce distinct intensity patterns, in the far field, for two wavelengths. These two patterns are images of bilingual virtual keyboard. To achieve this with DOEs is not simple, as they are inherently wavelength specific. Our technique is based on phase periodic characteristic of wavefront using iterative algorithm to design the phase profiles.     

基于粒子群算法的6自由度机械臂动力学模型参数辨识

提出了基于粒子群优化(PSO)算法的工业机器人动力学参数辨识方法。首先利用改进的牛顿-欧拉方法,建立考虑关节摩擦的机械臂线性动力学模型,然后引入PSO算法,建立基于PSO算法的估计未知动力学参数的算法,最后以UR工业机器人为实验对象,通过设计激励轨迹,激励工业机器人关节运动,并对关节运动参数进行采样,实现UR工业机器人的动力学参数估计,并根据力矩预测精度验证动力学模型。实验证明了所提出算法辨识工业机器人动力学模型参数的准确性和有效性。     

Parameter estimation of the generalized extreme value distribution for structural health monitoring

Structural health monitoring (SHM) can be defined as a statistical pattern recognition problem which necessitates establishing a decision boundary for damage identification. In general, data points associated with damage manifest themselves near the tail of a baseline data distribution, which is obtained from a healthy state of a structure. Because damage diagnosis is concerned with outliers potentially associated with damage, improper modeling of the tail distribution may impair the performance of SHM by misclassifying a condition state of the structure. This paper attempts to address the issue of establishing a decision boundary based on extreme value statistics (EVS) so that the extreme values associated with the tail distribution can be properly modeled. The generalized extreme value distribution (GEV) is adopted to model the extreme values. A theoretical framework and a parameter estimation technique are developed to automatically estimate model parameters of the GEV. The validity of the proposed method is demonstrated through numerically simulated data, previously published real sample data sets, and experimental data obtained from the damage detection study in a composite plate.     

Optimal parameters identification and sensitivity study for abrasive waterjet milling model

In this paper we present the work related to the parameters identification for abrasive waterjet milling (AWJM) model that appears as an ill-posed inverse problem. The necessity of studying this problem comes from the industrial milling applications where the possibility to predict and model the final surface with high accuracy is one of the primary tasks in the absence of any knowledge of the model parameters that should be used. The adjoint approach based on corresponding Lagrangian gives the opportunity to find out the unknowns of the AWJM model and their optimal values that could be used to reproduce the required trench profile. Due to the complexity of the nonlinear problem and the large number of the model parameters, we use an automatic differentiation software tool. This approach also gives us the ability to distribute the research on more complex cases and consider different types of model errors and 3D time dependent model with variations of the jet feed speed. This approach gives us a good opportunity to identify the optimal model parameters and predict the surface profile both with self-generated data and measurements obtained from the real production. Considering different types of model errors allows us to receive the results acceptable in manufacturing and to expect the proper identification of unknowns.     

河流水质模型参数估值方法探讨

采用最速下降法、正则化方法和正交化方法对河流水质模型参数进行了优化估算，并从计算方法的适用范围、计算机编程的复杂程度、整体工作量的大小，计算结果的代表性和合理性等方面对这三种方法进行了比较。     

Parameter identification of a class of nonlinear systems with numerical optimization technique

Abstract

A time domain technique for the identification of unknown parameters in the nonlinear system of some given structures is presented. The nonlinearities considered in this paper are in the forms of saturation, deadzone and backlash. The steps of the identification process are first to set the initial or guessed system parameters in a learning model, then apply the recorded input signal of the actual system to the mathematical learning model. By comparing the output errors between the model's and the system's responses, an optimization algorithm can be applied to search for the true parameters of the nonlinear system in order to meet a least mean square error criterion. The computer simulations have demonstrated the feasibility of this numerical technique.     

Structural optimization of a motorcycle chassis by pattern search algorithm

Changes to the technical regulations of the motorcycle racing world classes introduced the new Moto2 category. The vehicles are prototypes that use single-brand tyres and engines derived from series production, supplied by a single manufacturer. The stability and handling of the vehicle are highly dependent on the geometric properties of the chassis. The performance of a racing motorcycle chassis can be primarily evaluated in terms of weight and stiffness. The aim of this work is to maximize the performance of a tubular frame designed for a motorcycle racing in the Moto2 category. The goal is the implementation of an optimization algorithm that acts on the dimensions of the single pipes of the frame and involves the design of an objective function to minimize the weight of the frame by controlling its stiffnesses.     

Assessing and ensuring parameter identifiability for a physically-based strain hardening model for twinning-induced plasticity

Physically-based strain hardening models have become important ingredients in metal forming simulations over the last years, since they allow for the modeling of multi-stage forming processes based on the evolution of physically meaningful internal variables. Although these models are physically-based, there are still many fitting parameters involved which have to be identified from experiments. As a matter of fact, for each physical effect that is included in the model, a separate equation with new fitting parameters is introduced, such that physically-based models tend to contain a large number of fitting parameters. Parameter estimation is often based on the macroscopic response of a specimen which is tested in compression, tension or shear at various strain rates and temperatures. It is not guaranteed that this macroscopic information suffices to estimate parameters in model equations that describe (sub-) microscopic phenomena, since the effect of one parameter on the course of strain hardening can be compensated by other parameters. Since such parameter correlations are hard to detect from the model equations alone, the parameter estimation process may be ill-conditioned, i.e. numerous parameter sets can be found for such models that deliver almost the same minimum value of the error function in the parameter identification process. Given that parameter estimation involves a series of costly experiments, methods are needed that allow for analyzing the identifiability of the model parameters before costly experiments are performed. In this paper, an approach is presented that analyzes model parameter dependencies and quantifies the identifiability of the model parameters. The model considered in this study calculates the flow stress based on the evolution of three dislocation densities and the evolution of deformation twins. The analysis shows that correlations between the model parameters exist and that it is not possible to determine all model parameters based on an experimental set of flow curves in a single curve fitting procedure. An adapted fitting strategy is presented in which fitting is performed step-wise so that in each fitting step, only identifiable parameters are estimated, allowing for successful parameter identification.     

Identification of probabilistic distribution parameters for the mesoscopic stochastic fracture model

As a kind of multiphase composite material, the basic mechanical behaviors of concrete are randomness and nonlinearity. The mesoscopic stochastic fracture model (MSFM) which can reflect the coupling effect of randomness and nonlinearity, has been widely used for the nonlinear analysis of concrete structures. In this paper, we proposed a new stochastic modeling principle to identify the probabilistic distribution parameters of MSFM. In order to reduce the modeling works, a dimension-reduced algorithm is proposed as well. In this paper, an overview of MSFM is firstly presented to introduce the background of the research. Then the stochastic harmonic function (SHF) representation is introduced to express the random field mentioned in the MSFM, and the probability density evolution method (PDEM) is applied to obtain the theoretical probability density function (PDF) of the stress–strain relationships. Furthermore, a stochastic modeling principle is proposed, in which minimizing the Kullback–Leibler divergence (KLD) is taken as the optimization object. Based on the framework of genetic algorithm, a dimension-reduced algorithm is proposed to identify the parameters with reference to the data from tested complete curves of uniaxial compressive and uniaxial tensile stress–strain relationship of concrete. The results indicate that the proposed principle and algorithm can be used to identify the parameters of MSFM accurately and efficiently.     

Identification of material parameters of the Gurson–Tvergaard–Needleman model by combined experimental and numerical techniques

To identify material parameters from suitable experiments it is prevalent to use global informations like force–displacement or force–necking curves. The quality of accordance between measured and calculated forces at given displacements can be expressed by a least-squares functional. In this contribution a non-linear optimization method will be presented, which minimizes the least-squares functional by use of a gradient based method. The gradient of this functional is calculated in a semi-analytical sensitivity analysis. To determine the derivatives of the force with respect to the material parameters, the local sensitivities on an intersection will be added together. On this intersection, the total nodal force and the external force have to be equal and the normal displacements have to be independent on the material parameters. The parameter identification is embedded in the finite element code SPC-PMHP for solving non-linear boundary and initial value problems on parallel computers. The Gurson–Tvergaard–Needleman model is used to describe the plastic deformation and damage behaviour of the ductile structural steel StE 690. The developed algorithm is applied to tensile tests with notched cylindrical bars.     

A goal-oriented field measurement filtering technique for the identification of material model parameters

The post-processing of experiments with nonuniform fields is still a challenge: the information is often much richer, but its interpretation for identification purposes is not straightforward. However, this is a very promising field of development because it would pave the way for the robust identification of multiple material parameters using only a small number of experiments. This paper presents a goal-oriented filtering technique in which data are combined into new output fields which are strongly correlated with specific quantities of interest (the material parameters to be identified). Thus, this combination, which is nonuniform in space, constitutes a filter of the experimental outputs, whose relevance is quantified by a quality function based on global variance analysis. Then, this filter is optimized using genetic algorithms.     

Role of cell polarity dynamics and motility in pattern formation due to contact-dependent signalling

A key challenge in biology is to understand how spatio-temporal patterns and structures arise during the development of an organism. An initial aggregate of spatially uniform cells develops and forms the differentiated structures of a fully developed organism. On the one hand, contact-dependent cell–cell signalling is responsible for generating a large number of complex, self-organized, spatial patterns in the distribution of the signalling molecules. On the other hand, the motility of cells coupled with their polarity can independently lead to collective motion patterns that depend on mechanical parameters influencing tissue deformation, such as cellular elasticity, cell–cell adhesion and active forces generated by actin and myosin dynamics. Although modelling efforts have, thus far, treated cell motility and cell–cell signalling separately, experiments in recent years suggest that these processes could be tightly coupled. Hence, in this paper, we study how the dynamics of cell polarity and migration influence the spatiotemporal patterning of signalling molecules. Such signalling interactions can occur only between cells that are in physical contact, either directly at the junctions of adjacent cells or through cellular protrusional contacts. We present a vertex model which accounts for contact-dependent signalling between adjacent cells and between non-adjacent neighbours through long protrusional contacts that occur along the orientation of cell polarization. We observe a rich variety of spatiotemporal patterns of signalling molecules that is influenced by polarity dynamics of the cells, relative strengths of adjacent and non-adjacent signalling interactions, range of polarized interaction, signalling activation threshold, relative time scales of signalling and polarity orientation, and cell motility. Though our results are developed in the context of Delta–Notch signalling, they are sufficiently general and can be extended to other contact dependent morpho-mechanical dynamics.     

A modified version of a T‐Cell Algorithm for constrained optimization problems

In this paper, we present a heuristic inspired on the T‐Cell model of the immune system (i.e. an artificial immune system). The proposed approach (called T‐Cell) is used for solving constrained (numerical) optimization problems, and is validated using several test functions taken from the specialized literature on evolutionary optimization. Additionally, several engineering optimization problems are also used for assessing the performance of the proposed approach. The results are compared with respect to approaches representative of the state‐of‐the‐art in constrained evolutionary optimization. Copyright © 2010 John Wiley & Sons, Ltd.     

基于遗传算法的磁流变阻尼器Bouc_Wen模型参数辨识

在除试验数据外无任何先验知识的条件下如何识别Bouc_Wen模型的参数是一个亟待解决的问题。本文在对磁流变阻尼器进行力学特性试验的基础上,采用遗传算法对磁流变阻尼器Bouc_Wen模型进行参数辨识,并采用缩小参数取值范围的方法逐渐提高遗传算法的识别精度。通过分析参数值与电流间的特征曲线,采用曲线拟合的方法确定它们之间的函数关系,再利用遗传算法得到具体的函数表达式。最后用不同幅值和频率的激励试验数据对识别结果和拟合结果进行了验证。结果表明：利用缩小取值范围的方法得到的Bouc_Wen模型参数识别结果在全局最优解的附近,拟合结果和辨识出的模型均能满足要求     

Parameter estimation of a meal glucose–insulin model for TIDM patients from therapy historical data

The effect of meal on blood glucose concentration is a key issue in diabetes mellitus because its estimation could be very useful in therapy decisions. In the case of type 1 diabetes mellitus (T1DM), the therapy based on automatic insulin delivery requires a closed‐loop control system to maintain euglycaemia even in the postprandial state. Thus, the mathematical modelling of glucose metabolism is relevant to predict the metabolic state of a patient. Moreover, the eating habits are characteristic of each person, so it is of interest that the mathematical models of meal intake allow to personalise the glycaemic state of the patient using therapy historical data, that is, daily measurements of glucose and records of carbohydrate intake and insulin supply. Thus, here, a model of glucose metabolism that includes the effects of meal is analysed in order to establish criteria for data‐based personalisation. The analysis includes the sensitivity and identifiability of the parameters, and the parameter estimation problem was resolved via two algorithms: particle swarm optimisation and evonorm. The results show that the mathematical model can be a useful tool to estimate the glycaemic status of a patient and personalise it according to her/his historical data.Inspec keywords: medical control systems, closed loop systems, particle swarm optimisation, parameter estimation, biochemistry, diseases, patient monitoring, patient diagnosis, blood, sugar, patient treatment, medical computingOther keywords: meal intake, metabolic state, mathematical modelling, postprandial state, closed‐loop control system, automatic insulin delivery, T1DM, type 1 diabetes mellitus, therapy decisions, blood glucose concentration, TIDM patients, meal glucose–insulin model, mathematical model, parameter estimation problem, data‐based personalisation, glucose metabolism, insulin supply, carbohydrate intake, glucose records, therapy historical data, glycaemic state     

The model of the residual life time estimation of tribojoint elements by formation criteria of the typical contact fatigue damages

A two-dimensional theoretical model is proposed for investigation of the fracture processes and assessing residual contact durability of solids subjected to cyclic contact. The model is based on the step-by-step calculation of fatigue crack propagation paths in the contact region which includes the criteria of local fracture of materials under complex stress–strain state, characteristics of fatigue crack growth resistance of materials and also presupposes the possible change of fracture mechanisms (transversal shear – normal opening fracture mechanisms). Within the frames of the model the peculiarities of formation of such typical contact fatigue damages like pits, spalls, squat (“dark spot”) and cracking (“checks”) in rolling bodies and edge cracks growth in the elements of fretting couples under conditions of sliding/sticking between them are investigated. Examples of assessing the life time by damages formation (pitting and spalling) in the contact region are presented.     

Simulations for dynamics of granular mixtures in a rotating drum

We present a simple model and carry out simulations to investigate the dynamics of mixtures of granular material within a rotating drum. On the basis of the commonly held belief (supported by considerable experimental evidence) that segregation is due to motion of particles on the active layer, the bulk playing little or no role, we introduce a 2d lattice gas model which takes into account the rotational frequency, frictional forces, and the gravitational field, and represents segregation tendencies via activated effective grain-grain interactions. Our results include the onset of segregation perpendicular to the drum axis, the appearance and subsequent coarsening of bands and peculiarities of the effects of periodic modulation of the drum. Observed effects such as the segregation of rougher (smoother) particles into the bellies (necks) of the modulation are reproduced by our simulation. Received: 30 March 2000     

Bammann-Chiesa-Johnson粘塑性本构模型的参数识别方法与验证

Bammann-Chiesa-Johnson(BCJ)粘塑性本构模型对材料力学响应的再现和预测能力强烈依赖于其模型参数的确定,而模型参数的确定往往是通过反分析方法来进行。由于BCJ粘塑性模型包含了应变、应变率和温度耦合效应以及加载路径和温度历史,其常数多达18个,所以寻找最佳的模型参数识别值十分繁琐。针对BCJ本构模型参数复杂、识别困难的问题,本文基于参数的物理意义,在准静态、蠕变及动态加载试验基础上,通过模型参数解耦分离、粒子群智能优化的方法分6步对18个材料常数进行识别,并用识别结果对1060纯铝动态加载试验力学响应进行模拟,模拟结果与试验结果符合良好。通过定量化误差分析,证明了BCJ粘塑性模型对实验数据的预测具有较高精度,该模型参数识别方法科学可行。     

Dynamics and pattern formation in thermally induced phase separation of polymer–solvent system

In this paper, the characteristic dynamics and pattern formation of polymer–solvent system, undergoing thermal induced phase separation (spinodal decomposition), have been studied using non-linear simulations. The modified Cahn–Hilliard equation with the incorporation of the Flory–Huggins free energy model is used. The development of a wide variety of initial and intermediate stage morphology is feasible either by introducing sharp/continuous spatial temperature variation or by altering the initial mean solvent concentration (C_o) in the vicinity of the minima (C_m) of the spinodal parameter (second derivative of excess Gibbs free energy w.r.t concentration). For C_o < C_m initial stage of phase separation proceeds by the formation of columns/island of the solvent rich phase dispersed in polymer rich continuous phase. On the contrary, C_o > C_m, phase separation leads to formation of holes of polymer rich phase dispersed in the continuous solvent rich phase. For an intermediate mean concentration very close to C_m inter-connecting structures are evolved. A continuous variation of spatial temperature leads to the directional (rather than random) evolution of phase separated pattern. The initial stage of evolution always propagates from the lowest to the highest temperature regime. A spatial step variation of temperature leads to formation of localized novel phase separated patterns. For a flow system, the dynamics and pattern formation can be controlled by tuning the different timescales: timescale of convection, de-mixing and coalescence.