New results on PWARX model identification based on clustering approach

This paper deals with the problem of piecewise auto regressive systems with exogenous input(PWARX) model identification based on clustering solution. This problem involves both the estimation of the parameters of the affine sub-models and the hyper planes defining the partitions of the state-input regression. The existing identification methods present three main drawbacks which limit its effectiveness. First, most of them may converge to local minima in the case of poor initializations because they are based on the optimization using nonlinear criteria. Second, they use simple and ineffective techniques to remove outliers. Third, most of them assume that the number of sub-models is known a priori. To overcome these drawbacks, we suggest the use of the density-based spatial clustering of applications with noise(DBSCAN) algorithm. The results presented in this paper illustrate the performance of our methods in comparison with the existing approach. An application of the developed approach to an olive oil esterification reactor is also proposed in order to validate the simulation results.     

Closed‐Loop Identification for Control of Linear Parameter Varying Systems

This paper deals with system identification for control of linear parameter varying systems. In practical applications, it is often important to be able to identify small plant changes in an incremental manner without shutting down the system and/or disconnecting the controller; unfortunately, closed‐loop system identification is more difficult than open‐loop identification. In this paper we prove that the so‐called Hansen scheme, a technique known from linear time‐invariant systems theory for transforming closed‐loop system identification problems into open‐loop‐like problems, can be extended to accommodate linear parameter varying systems as well. We investigate the identified subsystem's parameter dependency and observe that, under mild assumptions, the identified subsystem is affine in the parameter vector. Various identification methods are compared in direct and Hansen Scheme setups in simulation studies, and the application of the Hansen Scheme is seen to improve the identification performance.     

软件项目风险管理系统设计研究

本文采用B/S系统架构和Java框架进行设计开发,提出能够满足软件项目风险管理需求的软件系统设计,并将风险识别功能、风险分析功能、风险计划功能以及风险监控功能等集为一体进行设计,从而保证新系统本身应用的有效性与稳定性,达到减少项目开发的风险性、提高项目成功率的目的。     

Non‐Parametric Identification Method of Volterra Kernels for Nonlinear Systems Excited by Multitone Signal

To solve the problem of Volterra frequency‐domain kernels (VFKs) of nonlinear systems, which can be difficult to identify, we propose a novel non‐parametric identification method based on multitone excitation. First, we have studied the output properties of VFKs of nonlinear systems excited by the multitone signal, and derived a formula for identifying VFKs. Second, to improve the efficiency of the non‐parametric identification method, we suggest an increase in the number of tones for multitone excitation to simultaneously identify multi‐point VFKs with one excitation. We also propose an algorithm for searching the frequency base of multitone excitation. Finally, we use the interpolation method to separate every order output of VFK and extract its output frequency components, then use the derived formula to calculate the VFKs. The theoretical analysis and simulation results indicate that the non‐parametric method has a high precision and convenience of operation, improving the conventional methods, which have the defects of being unable to precisely identify VFKs and identification results are limited to three‐order VFK.     

基于半监督距离学习的运动相似性度量算法

人体运动捕获技术的发展使得运动捕获数据不断积累,人体运动的检索技术成为运动数据管理和重用过程中的关键环节。由于逻辑相似的运动在数值上并不一定相似,使用欧式距离度量2个运动间的逻辑相似性难以取得理想的结果。为此,提出一种半监督的距离度量学习算法,利用带标记的运动和未标记运动进行训练以得到运动间的马氏距离度量,从而判断2个运动之间的逻辑相似性,实现运动检索。实验结果表明,与现有的大部分检索算法相比,该算法能够得到更高的查询精度,且没有任何人工干预,可应用于自动检索领域。     

煤部分气化燃烧集成系统最优化模型及其求解

运用已有的基础动力学数据，将流化床气化过程用全混流模型描述，同时认为部分气化煤焦的燃烧速率相当快，其产生的电能只与燃烧的炭量成线性关系，不受其他因素的影响，在此简化条件的基础上建立了一个部分气化、燃烧集成优化的经济评价模型。通过对模型的模拟计算发现：煤部分气化、燃烧集成优化是可行的，采用这一集成技术在提高煤炭利用率的同时，也可提高经济效益；1000℃时，三种煤的最优气化率在60％～80％之间，活性好的神木煤其最优部分气化率最高；提高气化温度，可以提高煤焦最优气化率，从而可提高总体经济效益。     

Fatty acid esterification by reactive distillation: Part 2—kinetics-based design for sulphated zirconia catalysts

In a previous study on the basis of reaction equilibrium, it was demonstrated that the synthesis of fatty acids esters of heavy alcohols by catalytic reactive distillation (RD) is basically feasible. Presently, the analysis is expanded to the kinetics-based design applied to the synthesis of 2-ethylhexyl dodecanoate. Experimental data for a super-acidic sulphated zirconia catalyst are employed. It is shown that this catalyst is highly selective even for the high alcohol/acid ratios as prevail in a RD column. To avoid catalyst deactivation, liquid-liquid segregation into an organic and an aqueous phase must be prevented, and operating temperatures beyond must be chosen. A kinetic model based on liquid activities is proposed, with parameters determined from both kinetic and equilibrium measurements. A typical column design has 13 reactive stages with a catalyst loading of , temperature profile of 400-, and a space velocity of for 99.9% purity product ester. A modified Damköhler number is proposed.     

基于有限状态机的RFID流数据过滤与清理技术

为了实现从大量不可靠、冗余的RFID（radio frequency identification）流数据中提取有效信息,提高RFID系统中数据的质量,提出了一种基于有限状态机的RFID 流数据过滤与清理方法.实验结果表明：该方法能够有效过滤系统外标签数据,清理系统内部冗余标签数据,筛选有效标签数据,并能够降低漏读、误读带来的风险.最后,利用地理信息系统的可视化技术,将过滤与清理结果展示在地图上.     

基于深度学习的作曲家分类问题

在音乐信息检索领域,作曲家分类是一个十分重要的问题,这一问题的目标是通过音频数据来识别相应的作曲家信息.传统的分类算法都是通过提取复杂的特征来进行分类的,而深层神经网络在特征学习上具有比较强的能力,因此提出用深层神经网络来解决这一问题.为了结合不同深层神经网络模型的优点,设计了一种混合模型,该模型基于深度置信网络(deep belief network,DBN)和级联去噪自编码器(stacked denoising autoencoder,SDA),可以较好地解决作曲家分类问题.实验表明,该模型取得了76.26%的正确率,这一结果比单纯用某一种模型搭建的深层神经网络以及支持向量机要好.和图像数据类似,人脑在提取音乐特征也是分层的,每一层对信号的处理不一样,因此混合模型在解决作曲家分类问题上具有一定的优势.     

基于前向预测策略的块正交匹配追踪算法研究

在块正交匹配追踪算法中,候选集匹配原子块的选择策略对信号重建具有重要作用,但是该算法在迭代选择原子块的过程中,每次选择当次最优迭代的原子块,并不能保证最终迭代性能是最优的。为此,提出一个基于前向预测的最优匹配原子块选择策略的块正交匹配追踪算法,通过预测原子块在未来迭代过程中的性能选择最优匹配的原子块。实验结果表明,与块正交匹配追踪算法相比,该算法的重建误差随前向预测参数的增大而减小。