基于CBR的机械产品模块化设计方法的研究

提出在模块化设计中模块对设计需求的适应性问题 ;论述了 CBR技术同模块化设计的关系 ,研究了基于实例的模板创建方法以及专家知识集成于设计模板的策略 ;分析液压机典型结构 ,创建模板实例及参数化的机身CAD模型 ;构造参数化设计的专家知识系统 ,实现知识的查询、提取、更新和基于知识的参数驱动机制     

微波等离子体炬(MPT)光谱仪的研制

MPT（Microware Plasma Torch）光谱仪是以吉林大学首创的微波等离子体炬（即MPT）为激发光源而研制出来的具有完全自主知识产权，世界领先的新式光谱分析仪器。研制中开发了大量光、机、电、计算机和化学方面的专有技术，保证了仪器的优良性能，与同类仪器相比具有很高的性能价格比，可广泛适用于临床医学，商品检验，技术侦察，冶金，岩矿分析、石油化工、自来水、饮料、食品、环境保护等部门，特别是在高等学校的原子发射光谱的实验教学中更能反映原子光谱法发展的现状和当代科技水平。     

Statistical modeling of fiber optic current transducer

Fiber optic current transducer (FOCT) is a new type of current measurement instrument with high accuracy, large dynamic range, wide frequency range and excellent insulating property. Thus, it has been highly concerned for using in high voltage grid. Although these FOCTs have such crucial advantages, the stochastic errors, mainly including random walk, bias instability, quantization noise, may suddenly degrade the system performance in a short period of time. To evaluate the stochastic error characteristics of fiber optic current transducer (FOCT), traditional solution adopts Allan variance method to plot the curve on the log–log scale. Considering the deficiency of severe shock in Allan variance, total variance methodology is proposed to analyze the stochastic error characteristics of FOCT with the aim of overcoming the deficiency. Total variance method transforms time-offset data to frequency-offset data and extends the data to be a virtual sequence nearly as three times long as the initial data by mapping. Additionally, the least square fitting (LSF) is applied to identify and obtain each coefficient of FOCT stochastic error, which helps to evaluate the stochastic error characteristics of FOCT excellently. Simulation results demonstrate that both Allan variance result and total variance result are nearly equivalent, when correlation time is short. However, when correlation time exceeds half total duration of the measurement, Allan variance estimation values tend to shock increasingly severely and total variance curve always extends steadily. Besides, every stochastic error coefficient is decreased effectively by total variance method, comparing with Allan variance method. Total variance methodology overcomes the deficiency of severe shock in Allan variance successfully, and then optimizes evaluation method of FOCT stochastic error characteristics.     

A Web-based CBR knowledge management system for PC troubleshooting

Using case-based reasoning (CBR), the authors integrate the techniques of cognitive task analysis (CTA), hierarchical clustering and ontology and propose a Web-based CBR knowledge management (KM) system for investigating the construction of a KM system with multiple information techniques to support KM activity in industry. The maintenance service centres of a computer company are used as an example to illustrate extracting the maintenance knowledge necessary to construct a PC troubleshooting KM system. The effectiveness of applying a Web-based CBR KM system to support KM activities in the KM life cycle is subjected to practical verification.     

基于实例的学习矢量量化神经网络诊断方法及其应用

基于CBR良好的可扩充性、可移植性和神经网络极强的分类能力,提出了基于实例的学习矢量量化神经网络诊断方法。该方法应用于机械故障诊断系统中,可以减小实例搜索空间,提高实例检索效率。论述了系统的设计方法和应用步骤。     

New algorithm for CBR-RBR fusion with robust thresholds

The case-based reasoning(CBR) and rule-based reasoning(RBR) fusion systems include a diverse range of fusion methods and their tasks are characterized by interleaving combination of the reasoning procedures. Existing approaches cannot clarify the complex relationships between data from the knowledge sources nor uniformly represent the heterogeneous case and rule knowledge in one fusion space. As a result, existing approaches fail to solve system fragility due to knowledge uncertainty and reasoning unreliability. For the purpose of addressing the difficulties, a novel algorithm for CBR-RBR fusion with robust thresholds(CRFRT) is proposed. Heterogeneous case and rule knowledge are uniformly represented in one defined fusion unitary space. The robust thresholds have been achieved to distinguish the complex relationships between meta-knowledge in the fusion space and to enhance system capacity of knowledge identification. Furthermore, fusion reasoning strategies are constructed for CRFRT and its procedure based on which robust solution of the fusion reasoning problem is obtained. Finally, CRFRT is validated by benchmark problems in machine learning. Compared with other CBR and RBR approaches, the reasoning efficiency and accuracy are increased by 5% and 2.2% respectively. The variations of system accuracy are decreased by 2% to 3.8%. The above results show that the CRFRT algorithm boosts the system’s effectiveness and robustness. The proposed CRFRT can solve the fragility of complex intelligence decision system and give quality performance for fault diagnosis.     

Using a principal components analysis for developing a robust design of electron beam welding

This paper presents the application of principal components analysis for Taguchi orthogonal experiments to develop a robust electron beam welding treatment (EBWT) process with high efficiency multiple performance characteristics (MPCs). In this study, the principal components analysis (PCA) design incorporating the correlation matrix of tested trials is employed. In the first step, the MPCs are reduced to two independent components using PCA. Both components accounted for 98.8% of the total variance. The first principal component (PC), which refers to the integrated hardening capability index of the EBWT process, accounts for 70.7% of the total variation. The remaining 28.1% were contributed by the second PC, which can be interpreted as the penetration capability index. In the second step, we identify the most important PC loading vectors using PCA, and estimate the importance of the PCs. By using PCA, relationships between different MPCs can be investigated and the most important factors for the variance of the EBWT process can be identified. The experimental results show that redundant information could be eliminated by using principal components in conjunction with Taguchi’s orthogonal array experiments. This proposed approach is simple, effective, and efficient for developing a robust and high-efficiency EBWT process of high quality. In this study, the MPCs in the EBWT process are successfully optimized .     

基于事例推理的装配顺序规划研究

就现有的计算机辅助装置顺序规划研究中存在的不足，提出基于事例推理（ＣＢＲ）的装配顺序规划思路，研究基于ＣＢＲ的装配顺序规划中的关键技术，开发出产品装配分析原型系统ＰＡＡＳ。系统包含装配数据建模，装配顺序规划和综合ＤＦＡ分析等功能。结合原型系统实例讲解基于ＣＢＲ的装配顺序规划的实现方法。     

Characterization of electrical sags and swells using higher-order statistical estimators

This paper deals with the detection of power quality anomalies which preserve the frequency of the power line, in particular sags and swells. Three statistical estimators have been used (variance, skewness and kurtosis) to enhance characterization of these anomalies. The proposed measurement strategy is funded in the tuning of the signal under test via a sliding window over which computation is performed. Then, the calculation of the statistical features reveals the inherent properties of the signal: amplitude, frequency and symmetry. The work primarily examines a number of synthetics in order to extract the theoretical statistical features. Then the algorithm is corroborated using real-life signals, obtaining an accuracy of 83%. This experience is part of the design of an instrument for the measurement of the power quality.     

注塑模CBR设计决策系统中利用神经网络方法进行事例检索的研究

注塑模 CBR(基于事例的推理 )系统是借助于历史经验来指导设计决策的。在系统中 ,求解事物的相似性是最重要的一环。本文探讨了利用神经网络方法进行事例检索的途径 ,通过对 CBR系统中输入条件的编码 ,经过对神经网络的样本训练 ,最近邻法调整和人类专家调整 ,得出求解事例相似性的神经网络结构。利用此方法 ,可以对后续设计步骤提供了很好的指导     

Parametric analysis on electrochemical discharge machining of silicon nitride ceramics

The electrochemical discharge machining (ECDM) process has a potential in the machining of silicon nitride ceramics. This paper describes the development of a second order, non-linear mathematical model for establishing the relationship among machining parameters, such as applied voltage, electrolyte concentration and inter-electrode gap, with the dominant machining process criteria, namely material removal rate (MRR), radial overcut (ROC) and thickness of heat affected zone (HAZ), during an ECDM operation on silicon nitride. The model is developed based on response surface methodology (RSM) using the relevant experimental data, which are obtained during an ECDM micro-drilling operation on silicon nitride ceramics. We also offer an analysis of variance (ANOVA) and a confirmation test to verify the fit and adequacy of the developed mathematical models. From the parametric analyses based on mathematical modelling, it can be recommended that applied voltage has more significant effects on MRR, ROC and HAZ thickness during ECDM micro-drilling operation as compared to other machining parameters such as electrolyte concentration and inter-electrode gap.     

视频转码压缩技术的研究

介绍一种基于MPEG-2的变码率到定码率的视频转码压缩系统。采用软件实现，极大地降低了成本。     

基于ELM和近似熵的脑电信号检测方法

脑电癫痫波的自动检测与分类对癫痫病情的诊断具有重要意义。提出了一种基于极端学习机(extreme learning ma-chine,ELM)和近似熵的脑电信号检测方法。首先,计算脑电信号的近似熵作为非线性特征,并与利用小波变换技术提取的线性特征波动指数相结合,组成特征向量,然后将特征向量送入单隐层前馈神经网络,采用ELM学习算法训练网络。实验表明,与BP(backpropagation)和SVM(support vector machine)算法相比,ELM在训练时间和识别精度两方面性能最佳,对用于实验的脑电数据检测识别率达到98%以上。     

Research on image reconstruction algorithms based on autoencoder neural network of Restricted Boltzmann Machine (RBM)

Aiming at the problem of low quality in image reconstruction of traditional image reconstruction algorithm of electromagnetic tomography(EMT), an EMT image reconstruction algorithm based on autoencoder neural network of Restricted Boltzmann Machine (RBM) is proposed. Firstly, the basic principles of EMT system and autoencoder neural network are analyzed. Autoencoder neural network is a deep learning model, which contains two parts: encoder and decoder. The encoding process of the encoder is equivalent to the object field detection process in the EMT system; the decoding process of the decoder is equivalent to the image reconstruction process. On this basis, an autoencoder neural network model is built. In this model, the RBM is used for layer by layer pre-training to obtain the initial weight and offset, and the global weight and offset are adjusted by BP algorithm. The parameter file generated in the trained autoencoder neural network is used to construct a decoder. Finally, the detected voltage value output by the EMT system is input into the decoder network to obtain the reconstructed image of the EMT. Furthermore, data with Gaussian noise and data regarding flow pattern not in training dataset are used to test the generalization ability and practicability of the network, respectively. The experimental results show that the method in this paper is a kind of EMT image reconstruction method with higher accuracy, which also provides a new means for EMT image reconstruction.     

Real-Time Simulator based hybrid control of DFIG-WES

This paper presents the analysis and design of recurrent neural network (RNN) and proportional and integral(PI) controller based hybrid control(HC) of doubly fed induction generator (DFIG). The proposed HC shows the quick dynamic and good transient response during the sudden changes in wind speed as well as generator speed. Further performance of proposed hybrid controller has been analyzed independently with RNN and PI control as its components. The HC along with RNN gives effective performance compared to conventional DTC (CDTC) and PI DTC in terms of flux ripples,torque ripples and settling time. The proposed technique is implemented in real-time simulator (RTS) based OPAL-RT and MATLAB/SIMULINK environment at a rating of 5.5 KW system under steeply and randomly change in wind velocity.     

基于PSD的电振动台特性检测系统研究

详细介绍了基于位置敏感传感器的电振动台特性检测系统的研究。该系统采用自准直光路原理,利用PSD对位置的高分辨能力特性,对电振动台微小角扭动进行精密测量。数据经高速AD采样后送入计算机,通过分析即可获得电振动台在各种振动模式下角速率时域频域特性。对系统存在的各误差分析得出,该系统200H z频率处角速率分辨力为19.8×1-0 3(°/s)。     

Spatially resolved diffractometry with atomic-column resolution

We demonstrate spatially resolved diffractometry in which diffraction patterns are acquired at two-dimensional positions on a specimen using scanning transmission electron microscopy (STEM), resulting in four-dimensional data acquisition. A high spatial resolution of about 0.1 nm is achieved using a stabilized STEM instrument, a spherical aberration corrector and various post-acquisition data processings. We have found a few novel results in the radial and the azimuthal scattering angle dependences of atomic-column contrast in STEM images. Atomic columns are clearly observed in dark field images obtained using the excess Kikuchi band intensity even in small solid-angle detection. We also find that atomic-column contrasts in dark field images are shifted in the order of a few tens of picometers on changing the azimuthal scattering angle. This experimental result is approximately interpretable on the basis of the impact parameter in Rutherford scattering. Spatially resolved diffractometry provides fundamental knowledge related to various STEM techniques, such as annular dark field (ADF) and annular bright field (ABF) imaging, and it is expected to become an analytical platform for advanced STEM imaging.     

虚拟仪器技术在传动轴寿命试验机中的应用

介绍传动轴寿命试验机测控系统的主要结构和特点，讨论基于虚拟仪器技术、多进程和DDE技术进行系统硬件及软件设计的原理及方法，VI技术提高了系统的实时性和可靠性，使系统具备了较强的扩展性与灵活性。     

Construction of pcAFM module to measure photoconductance with a nanoscale spatial resolution

A photoconductive atomic force microscopy (pcAFM) module was designed and the performance was tested. This module consisted of three units: the conductive mirror-plate, the steering mirror and the laser source. The module with a laser irradiation unit was equipped to a conventional conducting probe atomic force microscopy (CP-AFM) instrument to measure photoconductance in a nanoscale resolution. As a proof-of-concept experiment, the photoconductance of aggregated fullerene on indium tin oxide (ITO) substrate was measured with this module. The electrical signals (currents) of aggregated fullerene under the conditions of laser on/off at about -10V sample bias voltage were -100 to -160nA and 0 to -20nA, respectively. Results indicated that the pcAFM with this module allowed one to observe photoinduced changes of electrical properties in nanodevices with nanoscale spatial resolution.     

秦氏模型智能虚拟控件的实现

介绍了智能虚拟控件的模型要素，论述了智能虚拟控件的形成原理，讨论了智能虚拟控件的设计方法与优化准则，给出了智能虚拟控件的计算机表达方法。智能虚拟控件是对虚拟仪器的重要发展，它把仪器的功能、性能、控制关系等全部有机地融合于一个或几个部件之中。