Efficient model reduction in non-linear dynamics using the Karhunen-Loève expansion and dual-weighted-residual methods

 We look at the task of computing the time-evolution of a non-linear system for a long time, in our case under random external influences. Our specific example is the fatigue evaluation of a wind turbine. To facilitate such a computation, we look at a reduction of the computational effort by projecting everything on a low-dimensional basis. In this case we take the Karhunen-Loève basis generated from running the model a little while under the random loading. It is important that the error which is caused by this reduction process can be controlled. We estimate the error by dual or adjoint methods. This in turn allows the process of model reduction to be performed adaptively. Dedicated to the memory of Prof. Mike Crisfield, for his cheerfulness and cooperation as a colleague and friend over many years.     

Genetically optimized fuzzy polynomial neural networks with fuzzy set-based polynomial neurons

In this paper, we propose and investigate a new category of neurofuzzy networks—fuzzy polynomial neural networks (FPNN) endowed with fuzzy set-based polynomial neurons (FSPNs) We develop a comprehensive design methodology involving mechanisms of genetic optimization, and genetic algorithms (GAs) in particular. The conventional FPNNs developed so far are based on the mechanisms of self-organization, fuzzy neurocomputing, and evolutionary optimization. The design of the network exploits the FSPNs as well as the extended group method of data handling (GMDH). Let us stress that in the previous development strategies some essential parameters of the networks (such as the number of input variables, the order of the polynomial, the number of membership functions, and a collection of the specific subset of input variables) being available within the network are provided by the designer in advance and kept fixed throughout the overall development process. This restriction may hamper a possibility of developing an optimal architecture of the model. The design proposed in this study addresses this issue. The augmented and genetically developed FPNN (gFPNN) results in a structurally optimized structure and comes with a higher level of flexibility in comparison to the one we encounter in the conventional FPNNs. The GA-based design procedure being applied at each layer of the FPNN leads to the selection of the most suitable nodes (or FSPNs) available within the FPNN. In the sequel, two general optimization mechanisms are explored. First, the structural optimization is realized via GAs whereas the ensuing detailed parametric optimization is carried out in the setting of a standard least square method-based learning. The performance of the gFPNN is quantified through experimentation in which we use a number of modeling benchmarks—synthetic and experimental data being commonly used in fuzzy or neurofuzzy modeling. The obtained results demonstrate the superiority of the proposed networks over the models existing in the references.     

Transient aspects of inner wall temperatures at different spatial positions in fluidized bed evaporators

In this work, the instantaneous data of inner wall temperatures obtained at given axial positions but for different heated tubes or at various axial positions but at the same heated tube in a tube-bank fluidized bed evaporator with a vapor–liquid–solid external natural circulation flow were analyzed by employing linear and nonlinear time series analysis methods, including correlation dimension and Kolmogorov entropy analyses besides time domain, power spectrum and autocorrelation analyses. It is found that nonlinear characteristics of the wall temperature time series vary with spatial position in the evaporator at given operation conditions, and thus the multi-phase flow boiling behavior. However, the signals obtained at given axial position but different heated tubes in the evaporator exhibit more similar nonlinear evolution behavior, and those measured in the same heated tube but at different axial heights in the evaporator undergo more distinct nonlinear dynamic behavior. Multi-scale phenomena in different spatial positions are also found and discussed. These transient aspects of inner wall temperatures are closely related to the average holdup of solid particles and motion of vapor bubbles in vapor–liquid–solid flow. These results can provide some valuable information on the system modeling and controlling from the point of view of nonlinearity.     

奇异值分解压制随机噪声的方法及应用

针对随机噪声压制问题，通过对多项式确定的同相轴形状重新选取数据矩阵，再对新选取的数据矩阵进行SVD滤波，从而压制随机噪声。通过对矩阵奇异值分解的研究，揭示了奇异值分解的原理和SVD滤波原理，将一个多道地震记录看作一个图，它可分解为由一系列正交的子图，通过对子图适当选择并结合其它手段，从而达到提高信噪比的目的，为奇异值分解在地球物理中的进一步应用提供了理论依据。实际数据应用表明，该方法去噪手段灵活、保真性好、分辨率高，对一些突然的脉冲干扰、侧反射、以及其他的反向干扰有明显效果。     

基于小波除噪和神经网络理论的气固循环流化床颗粒浓度预测

提出了一种将混沌的相空间重构、小波包分析和神经网络相结合的新方法用于预测气-固循环流化床的颗粒浓度。首先利用小波包进行数据的消噪,然后用混沌方法重构相空间吸引子,用径向基神经网络拟合吸引子上的全局整体映射,构成混合预测模型。实验结果表明,将此混合模型用于预测气-固循环流化床的颗粒浓度,能达到较好的预测效果,预测精度比奇异值分解和傅里叶变换除噪高。     

日产1000t水泥熟料生产线的提产措施

通过对五级旋风筒下料管、回转窑转速、配料方案、物料稳定等四方面进行改进,达到提产目的。     

浅谈SSFC桥梁伸缩装置的原理及施工

从结构组成及工作原理、技术特点及优势等方面简述了SSFC伸缩装置，并对其安装工艺及关键质量控制点和维修及保养进行了阐述，以推广SSFC伸缩装置的应用。     

基于Legendre多项式的随机连续系统的Markov参数估计

本文在讨论连续Ｗｉｓｅｎｒ过程的Ｌｅｇｅｎｄｒｅ多项式逼近值的相关性和Ｗｉｅｎｅｒ过程扰动下连续线性系统基于该正交多项的最小二乘估计有偏性后，提出了无偏一致的且数估计误差方差最小的Ｍａｒｋｏｖ估计（最小方差估计）算法，并给出本文方法的仿真结果。     

粒界缓冲型展宽效应对PTC材料温度系数的影响

本文利用相变扩张理论分析了粒界缓冲型展宽效应对BaTiO_3系PTC陶瓷的介电系数及温度系数a_r的影响,指出了晶粒尺寸和温度系数的关系。