Identificação de danos estruturais utilizando técnicas de otimização.

Sistemas estruturais em suas variadas aplicações incluindo-se veículos espaciais, automóveis e estruturas de engenharia civil tais como prédios, pontes e plataformas off-shore, acumulam dano durante suas vidas úteis. Em muitas situações, tal dano pode não ser visualmente observado. Do ponto de vista da segurança e da performance da estrutura, é desejável monitorar esta possível ocorrência, localizá-la e quantificá-la. Métodos de identificação de sistemas, que em geral, são classificados numa categoria de Técnicas de Avaliação Não-Destrutivas, podem ser utilizados para esta finalidade. Usando dados experimentais tais como frequências naturais, modos de vibração e deslocamentos estáticos, e um modelo analítico estrutural, parâmetros da estrutura podem ser identificados. As propriedades estruturais do modelo analítico são modificadas de modo a minimizar a diferença entre os dados obtidos por aquele modelo e a resposta medida. Isto pode ser definido como um problema inverso onde os parâmetros da estrutura são identificados. O problema inverso, descrito acima, foi resolvido usando métodos globais de otimização devido à provável presença de inúmeros mínimos locais e a não convexidade do espaço de projeto. Neste trabalho o método da Evolução Diferencial (Differential Evolution, DE) foi utilizado como ferramenta principal de otimização. Trata-se de uma meta-heurística inspirada numa população de soluções sucessivamente atualizada por operações aritméticas como mutações, recombinações e critérios de seleção dos melhores indivíduos até que um critério de convergência seja alcançado. O método da Evolução Diferencial foi desenvolvido como uma heurística para minimizar funções não diferenciáveis e foi aplicado a estruturas planas de treliças com diferentes níveis de danos.

Robust filtering and smoothing with gaussian processes

We propose a principled algorithm for robust Bayesian filtering and smoothing in nonlinear stochastic dynamic systems when both the transition function and the measurement function are described by non-parametric Gaussian process (GP) models. GPs are gaining increasing importance in signal processing, machine learning, robotics, and control for representing unknown system functions by posterior probability distributions. This modern way of system identification is more robust than finding point estimates of a parametric function representation. Our principled filtering/smoothing approach for GP dynamic systems is based on analytic moment matching in the context of the forward-backward algorithm. Our numerical evaluations demonstrate the robustness of the proposed approach in situations where other state-of-the-art Gaussian filters and smoothers can fail. © 2011 IEEE.

ℓasso MPC: Smart regulation of over-actuated systems

In this paper, a novel MPC strategy is proposed, and referred to as asso MPC. The new paradigm features an 1-regularised least squares loss function, in which the control error variance competes with the sum of input channels magnitude (or slew rate) over the whole horizon length. This cost choice is motivated by the successful development of LASSO theory in signal processing and machine learning. In the latter fields, sum-of-norms regularisation have shown a strong capability to provide robust and sparse solutions for system identification and feature selection. In this paper, a discrete-time dual-mode asso MPC is formulated, and its stability is proven by application of standard MPC arguments. The controller is then tested for the problem of ship course keeping and roll reduction with rudder and fins, in a directional stochastic sea. Simulations show the asso MPC to inherit positive features from its corresponding regressor: extreme reduction of decision variables' magnitude, namely, actuators' magnitude (or variations), with a finite energy error, being particularly promising for over-actuated systems. © 2012 AACC American Automatic Control Council).

Reconstruction of arbitrary biochemical reaction networks: A compressive sensing approach

Reconstruction of biochemical reaction networks (BRN) and genetic regulatory networks (GRN) in particular is a central topic in systems biology which raises crucial theoretical challenges in system identification. Nonlinear Ordinary Differential Equations (ODEs) that involve polynomial and rational functions are typically used to model biochemical reaction networks. Such nonlinear models make the problem of determining the connectivity of biochemical networks from time-series experimental data quite difficult. In this paper, we present a network reconstruction algorithm that can deal with ODE model descriptions containing polynomial and rational functions. Rather than identifying the parameters of linear or nonlinear ODEs characterised by pre-defined equation structures, our methodology allows us to determine the nonlinear ODEs structure together with their associated parameters. To solve the network reconstruction problem, we cast it as a compressive sensing (CS) problem and use sparse Bayesian learning (SBL) algorithms as a computationally efficient and robust way to obtain its solution. © 2012 IEEE.

CFD investigation of turbulent premixed flame response to transverse forcing

Screech is a high frequency oscillation that is usually characterized by instabilities caused by large-scale coherent flow structures in the wake of bluff-body flameholders and shear layers. Such oscillations can lead to changes in flame surface area which can cause the flame to burn unsteadily, but also couple with the acoustic modes and inherent fluid-mechanical instabilities that are present in the system. In this study, the flame response to hydrodynamic oscillations is analyzed in a controlled manner using high-fidelity Computational Fluid Dynamics (CFD) with an unsteady Reynolds-averaged Navier-Stokes approach. The response of a premixed flame with and without transverse velocity forcing is analyzed. When unforced, the flame is shown to exhibit a self-excitation that is attributed to the anti-symmetric shedding of vortices in the wake of the flameholder. The flame is also forced using two different kinds of low-amplitude out-of-phase inlet velocity forcing signals. The first forcing method is harmonic forcing with a single characteristic frequency, while the second forcing method involves a broadband forcing signal with frequencies in the range of 500 - 1000 Hz. For the harmonic forcing method, the flame is perturbed only lightly about its mean position and exhibits a limit cycle oscillation that is characteristic of the forcing frequency. For the broadband forcing method, larger changes in the flame surface area and detachment of the flame sheet can be seen. Transition to a complicated trajectory in the phase space is observed. When analyzed systematically with system identification methods, the CFD results, expressed in the form of the Flame Transfer Function (FTF) are capable of elucidating the flame response to the imposed perturbation. The FTF also serves to identify, both spatially and temporally, regions where the flame responds linearly and nonlinearly. Locking-in between the flame's natural self-excited frequency and the subharmonic frequencies of the broadband forcing signal is found to alter the dynamical behaviour of the flame. Copyright © 2013 by ASME.

Bayesian inference and learning in Gaussian process state-space models with Particle MCMC

State-space models are successfully used in many areas of science, engineering and economics to model time series and dynamical systems. We present a fully Bayesian approach to inference and learning (i.e. state estimation and system identification) in nonlinear nonparametric state-space models. We place a Gaussian process prior over the state transition dynamics, resulting in a flexible model able to capture complex dynamical phenomena. To enable efficient inference, we marginalize over the transition dynamics function and, instead, infer directly the joint smoothing distribution using specially tailored Particle Markov Chain Monte Carlo samplers. Once a sample from the smoothing distribution is computed, the state transition predictive distribution can be formulated analytically. Our approach preserves the full nonparametric expressivity of the model and can make use of sparse Gaussian processes to greatly reduce computational complexity.

SARV光纤收放装置的控制系统设计与实现

根据小型自治遥控水下机器人SARV的运动特性,研制了光纤微缆收放的控制系统。设计使用了嵌入式QNX软件开发技术,系统稳定可靠。采用系统辨识的方法,获得被控对象的等效数学模型。采用单神经元自适应PID控制器对控制参数进行在线自调节,实现了SARV在水中运动时光纤收放的恒张力控制,满足光纤收放装置的设计要求。

AUV系统辨识与广义预测控制研究

本论文的研究内容分为两方面：AUV的建模和控制。 建模方面，主要对当前用于AUV的建模方法进行了分类及对比，给出了水动力机理建模、水动力辨识、面向目标的系统辨识三类方法的优缺点。 根据可辨识性理论，对AUV闭环系统进行分析，给出了AUV闭环系统可辨识的充分条件。为了提高辨识算法的实时性，解决辨识过程中的“数据饱和”问题，给出了改进的变步长增广卡尔曼滤波辨识算法。利用小型AUV湖上试验数据辨识出航向回路、深度回路的系统模型，通过不同的试验数据与模型预测值的相关性验证模型，试验结果表明了该算法应用于AUV闭环系统建模的可行性。 控制方面，在传统PID控制、S面控制方法基础上，借鉴单神经元PID控制思想，将积分环节加入S面控制中来简化S面PID控制算法，并通过仿真验证了算法的可行性。上述方法参数调节依赖工程经验，而广义预测控制具有对模型要求低、算法鲁棒性强、参数调节简单等优点。因此，本文对输入输出约束的广义预测控制快速算法应用于AUV系统进行仿真，通过小型AUV水池试验验证了算法的有效性。

利用计算机仿真与系统辨识求解导弹弹道轨迹

提出了求解某武器系统导弹弹道轨迹的两种方法：计算机仿真和系统辨识方法．详细介绍了导弹系统的计算机仿真模型，并利用控制理论和数值分析的方法对仿真模型求解；根据系统辨识理论，将整个系统看作“黑箱”，建立与输入、输出数据等价的模型，引入折息因子对模型进行辨识．最后分别给出了计算机仿真试验曲线和系统辨识试验曲线，证明了两种求解方法的有效性。

双线性系统辨识及其在海洋机器人中的应用

本文用双线性系统的辨识方法对海洋机器人航向与侧推的控制系统建模,并对所得模型进行了计算机仿真,仿真结果表明了模型的有效性。

Modelling and Prediction of NOx Emission in a Coal-Fired Power Generation Plant

In this paper NOx emissions modelling for real-time operation and control of a 200 MWe coal-fired power generation plant is studied. Three model types are compared. For the first model the fundamentals governing the NOx formation mechanisms and a system identification technique are used to develop a grey-box model. Then a linear AutoRegressive model with eXogenous inputs (ARX) model and a non-linear ARX model (NARX) are built. Operation plant data is used for modelling and validation. Model cross-validation tests show that the developed grey-box model is able to consistently produce better overall long-term prediction performance than the other two models.

Computationally efficient sequential learning algorithms for direct link resource-allocating networks

Computionally efficient sequential learning algorithms are developed for direct-link resource-allocating networks (DRANs). These are achieved by decomposing existing recursive training algorithms on a layer by layer and neuron by neuron basis. This allows network weights to be updated in an efficient parallel manner and facilitates the implementation of minimal update extensions that yield a significant reduction in computation load per iteration compared to existing sequential learning methods employed in resource-allocation network (RAN) and minimal RAN (MRAN) approaches. The new algorithms, which also incorporate a pruning strategy to control network growth, are evaluated on three different system identification benchmark problems and shown to outperform existing methods both in terms of training error convergence and computational efficiency. (c) 2005 Elsevier B.V. All rights reserved.