A complete VARMA modelling methodology based on scalar components

Abstract. This article proposes an extension to scalar component methodology for the identification and estimation of VARMA models. The complete methodology determines the exact positions of all free parameters in any VARMA model with a predetermined embedded scalar component structure. This leads to an exactly identified system of equations that is estimated using full information maximum likelihood.     

Nonsmooth identification of mechanical systems with backlash-like hysteresis

Backlash-like hysteresis is one of the nonsmooth and multi-valued nonlinearities usually existing in mechanical systems. The traditional identification method is quite difficult to be used to model the systems involved with such complex nonlinearities. In this paper, a nonsmooth recursive identification algorithm for the systems with backlash-like hysteresis is proposed. In this method, the concept of Clarke subgradient is introduced to approximate the gradients at nonsmooth points and the so-called bundle method is used to obtain the optimization search direction in nonsmooth cases. Then, a recursive algorithm based on the idea of bundle method is developed for parameter estimation. After that, the convergence analysis of the algorithm is investigated. Finally, simulation results to validate the proposed method on a simulated mechanical transmission system are presented.     

新型台式电脑CPU供电电路分析

本文概括了CPU供电电路所采用的新技术,阐述了实现这些技术的VRM电路结构和工作原理,最后详细分析了一种新型X58M主板CPU供电电路的工作原理。     

基于模糊RBF神经网络的换热器建模

针对换热器的复杂非线性特征,利用一种模糊RBF神经网络结构,对其网络学习算法进行部分改进,并用于解决换热器的建模问题。采用模糊RBF神经网络不仅符合人的思维推理方式,也提高了神经网络的学习泛化能力,在改进的学习算法中通过学习率的值的不断变化和添加动量项,可以使学习速度加快,提高了辨识换热器模型的准确性。通过与传统的学习算法的仿真比较,验证了所提出的改进学习算法在辨识精度和稳定性方面具有更好的效果。     

一种分布式网络指纹识别系统

文章介绍了一种分布式网络指纹识别系统NBFIS(Network-BasedFingerprintIdentifyingSystem)。首先提出模型框架,并对其关键技术特别是指纹图像处理技术及面向对象的数据库技术进行讨论。实现了NBFIS系统的原型并给出实验和性能测试结果。     

Identification and control of open-loop unstable processes by relay methods

This paper describes a reliable automatic PID tuning method for open-loop unstable processes. Identification with low order models is performed by means of two relay tests, one with an additional delay, which does not require a priori knowledge about the process, with the only necessary condition being that the process be gain stabilisable. This paper provides an overview of the method, states conditions that need to be satisfied for its successful implementation, and demonstrates its application on a number of examples.     

Ignitable Liquid Classification and Identification Using the Summed-Ion Mass Spectrum

Abstract

The mass spectra calculated by summing the intensities of each nominal mass over all chromatographic times, the summed ion spectra, were calculated for a set of 440 commercially available ignitable liquids that had been analyzed by gas chromatography–mass spectrometry. Each ignitable liquid is generally comprised of a mixture of chemicals, and the mixtures may be very complex in some cases, e.g., petroleum distillates, gasoline, etc. The summed ion spectra were evaluated to determine if they contained sufficient information content to allow their use for rapid and accurate identification of the ignitable liquid in a database or library. The summed ion nominal mass spectra were encoded in one bit per channel at 1% transition intensity and a set of 96,580 unique pairwise comparisons were made between spectra, resulting in an average of approximately 50 differing channels per comparison. A subset of 62 summed ion spectra were further compared by a similarity metric and found by cluster analysis to group closely along the ASTM ignitable liquid classification scheme. Receiver operator characteristic analysis of the use of the similarity index in combination with the summed ion spectra was shown to provide 99% probability of correct liquid identification and 95% probability of correct ASTM classification by primary and sub-class. Interlaboratory tests found 95% probability of correct liquid identification and 85% probability of correct ASTM classification at the primary class level. These results demonstrate the use of the summed ion spectrum as a first step in the rapid identification of an ignitable liquid by database or library searching.     

Least squares support vector machines with tuning based on chaotic differential evolution approach applied to the identification of a thermal process

In the past decade, support vector machines (SVMs) have gained the attention of many researchers. SVMs are non-parametric supervised learning schemes that rely on statistical learning theory which enables learning machines to generalize well to unseen data. SVMs refer to kernel-based methods that have been introduced as a robust approach to classification and regression problems, lately has handled nonlinear identification problems, the so called support vector regression. In SVMs designs for nonlinear identification, a nonlinear model is represented by an expansion in terms of nonlinear mappings of the model input. The nonlinear mappings define a feature space, which may have infinite dimension. In this context, a relevant identification approach is the least squares support vector machines (LS-SVMs). Compared to the other identification method, LS-SVMs possess prominent advantages: its generalization performance (i.e. error rates on test sets) either matches or is significantly better than that of the competing methods, and more importantly, the performance does not depend on the dimensionality of the input data. Consider a constrained optimization problem of quadratic programing with a regularized cost function, the training process of LS-SVM involves the selection of kernel parameters and the regularization parameter of the objective function. A good choice of these parameters is crucial for the performance of the estimator. In this paper, the LS-SVMs design proposed is the combination of LS-SVM and a new chaotic differential evolution optimization approach based on Ikeda map (CDEK). The CDEK is adopted in tuning of regularization parameter and the radial basis function bandwith. Simulations using LS-SVMs on NARX (Nonlinear AutoRegressive with eXogenous inputs) for the identification of a thermal process show the effectiveness and practicality of the proposed CDEK algorithm when compared with the classical DE approach.     

一种新型ICA算法在RFID系统中的应用

针对无线射频识别系统在通信过程中的防碰撞问题,从信号处理的角度,提出一种基于盲源信号并行分离技术的防碰撞处理算法,其中包括信号白化技术,以及一种改进的独立元分析(ICA)信号并行分离方法.功能仿真结果证明,该算法的吞吐率最高达到69％,与吞吐率只有18％的传统Aloha算法相比,该算法能提高防碰撞能力.     

RFID系统防碰撞算法比较分析及其改进算法

防碰撞算法是RFID系统中的一项关键技术。在对基本二进制搜索算法及其各种改进算法进行详细的定量分析的基础上提出了一种改进算法。该算法能有效地降低命令发送的总次数和减少每次命令所附带的参数长度。仿真结果表明了该算法的有效性。