矿用干式变压器局部放电模式识别方法

为提高矿用干式变压器局部放电模式识别准确率,提出了一种矿用干式变压器局部放电模式识别方法。首先,采用正交匹配追踪算法对原始局部放电信号进行去噪,最大程度保留原始局部放电信号的有用信息;然后,通过自回归模型提取去噪后局部放电信号的自回归系数特征;最后,将自回归系数特征输入随机森林集成分类器对局部放电模式进行识别。实验结果表明,该方法平均识别准确率达98%。     

Engineering methods for analytical calculation of error controllability in structured codes

Engineering methods are developed for analytical calculation of the total number of errors corrected by structured codes at different error multiplicities. Regular procedures are suggested for the determination of error controllability of codes using correctable error matrices.Translated from Kibernetika i Sistemnyi Analiz, No. 3, pp. 30–38, May–June, 1992.     

谐波对矿用干式变压器损耗及温升的影响

为了更准确地分析谐波对变压器损耗和温升的影响,首先建立了干式变压器二维结构模型,并对干式变压器的漏磁场与谐波损耗进行了仿真计算,得到干式变压器轴向和径向漏磁场分布规律及谐波损耗与谐波次数、谐波含有率、谐波电流畸变率的关系;然后利用Fluent对干式变压器二维温度场与流体场进行了流固耦合分析,得到干式变压器温升与谐波次数及谐波电流畸变率的关系。仿真结果表明,干式变压器的谐波铜损耗与谐波电流含有率的平方成正比,且当谐波电流含有率一定时,近似与谐波电流次数的平方成正比;而谐波铁损耗与谐波电压含有率的平方成正比,但当谐波电压含有率一定时,谐波铁损耗随谐波次数的增大而减小,并逐渐趋于平缓。     

矿用干式变压器绝缘在线监测及故障诊断预警系统研制

针对传统的矿用干式变压器监测方法采用热电偶检测干式变压器内部温度而存在监测量单一,不能有效反映干式变压器绝缘实时状态的问题,提出了一种矿用干式变压器绝缘在线监测及故障诊断预警系统的设计方案,介绍了系统整体结构和故障诊断方法,建立了干式变压器匝间绝缘故障类型库。现场调试结果表明,该系统运行稳定,可实现对矿用干式变压器的远程实时监测和故障诊断预警功能。     

矿用隔爆型干式变压器绕组局部放电对温升的影响

为了分析矿用隔爆型干式变压器绕组绝缘层中的气隙或杂质等缺陷对绕组温升的影响,以一台连续式绕组1 000kVA/10kV矿用干式变压器作为研究对象,介绍了矿用干式变压器绕组匝间及层间绝缘结构;借助红外摄像仪对不同放电类型和不同老化温度下试样的温升特性曲线进行类比分析,得出以Nomex绝缘纸为绝缘层的干式变压器匝间放电及层间放电与温度之间的关系;运用传热学原理并结合绝缘材料特性进行数值计算及分析,得出矿用干式变压器匝间放电对绕组发热过程的影响,为变压器内部热点温度的定位及局部放电点的准确定位提供了理论依据。     

BA-PNN-based methods for power transformer fault diagnosis

This paper presents a machine learning-based approach to power transformer fault diagnosis based on dissolved gas analysis (DGA), a bat algorithm (BA), optimizing the probabilistic neural network (PNN). PNN is a radial basis function feedforward neural network based on Bayesian decision theory, which has a strong fault tolerance and significant advantages in pattern classification. However, one challenge still remains: the performance of PNN is greatly affected by its hidden layer element smooth factor which impacts the classification performance. The proposed approach addresses this challenge by deploying the BA algorithm, a kind of bio-inspired algorithm to optimize PNN. Using the real data collected from a transformer system, we conducted the experiments for validating the performance of the developed method. The experimental results demonstrated that BA is an effective algorithm for optimizing PNN smooth factor and BA-PNN can improve the fault diagnosis performance; in turn, and the machine learning-based model (BA-PNN) can significantly enhance the accuracies of power transformer fault diagnosis.     

变压器故障诊断的可视化模型

为了提高变压器故障的诊断效率,基于可视化技术提出了分层流水线建模方法.该模型分为数据层、逻辑层和显示层三层.数据层对参数值进行规约,逻辑层提供检测算法实体、实体一参数关系和图形生成算法,显示层给出图形化的诊断结果.三层结构具有层间依赖性低,复用效率高和扩展性好等优点.实际变电站的应用表明,可视化模型可以直观展现故障,提高故障诊断效率.     

Formal methods for protocol testing: a detailed study

The authors present a detailed study of four formal methods (T-, U-, D-, and W-methods) for generating test sequences for protocols. Applications of these methods to the NBS Class 4 Transport Protocol are discussed. An estimation of fault coverage of four protocol-test-sequence generation techniques using Monte Carlo simulation is also presented. The ability of a test sequence to decide whether a protocol implementation conforms to its specification heavily relies on the range of faults that it can capture. Conformance is defined at two levels, namely, weak and strong conformance. This study shows that a test sequence produced by T-method has a poor fault detection capability, whereas test sequences produced by U-, D-, and W-methods have comparable (superior to that for T-method) fault coverage on several classes of randomly generated machines used in this study. Also, some problems with a straightforward application of the four protocol-test-sequence generation methods to real-world communication protocols are pointed out     

Multisim变压器参数分析与应用

针对软件Multisim中变压器参数含义不清所造成的困惑问题,结合变压器一般工作原理及方程,通过虚拟仿真实验,研究了Multisim中变压器模型及其部分参数的含义,给出了该软件变压器参数与变压器绕组自感互感的关系。通过一个电路实例,进一步说明了Multisim变压器参数的设置与应用。     

精细神经网络仿真方法研究进展

树突对大脑神经元实现不同的信息处理功能有着重要作用。精细神经元模型是一种对神经元树突以及离子通道的信息处理过程进行精细建模的模型,可以帮助科学家在实验条件的限制之外探索树突信息处理的特性。由精细神经元组成的精细神经网络模型可通过仿真对大脑的信息处理过程进行模拟,对于理解树突的信息处理机制、大脑神经网络功能背后的计算机理具有重要作用。然而,精细神经网络仿真需要进行大量计算,如何对精细神经网络进行高效仿真是一个具有挑战的研究问题。本文对精细神经网络仿真方法进行梳理,介绍了现有主流仿真平台与核心仿真算法,以及可进一步提升仿真效率的高效仿真方法。将具有代表性的高效仿真方法按照发展历程以及核心思路分为网络尺度并行方法、神经元尺度并行方法以及基于GPU(graphics processing unit)的并行仿真方法3类。对各类方法的核心思路进行总结,并对各类方法中代表性工作的细节进行分析介绍。随后对各类方法所具有的优劣势进行分析对比,对一些经典方法进行总结。最后根据高效仿真方法的发展趋势,对未来研究工作进行展望。     

Software for calculation of electric power systems parameters

The calculation of power transmission lines parameters of electric energy has a fundamental importance in the design, construction and simulation of electric power systems. Generally, the results obtained are used in the database for the studies of power flow, short circuit and stability.This paper presents a software with graphical interface which allows the estimation of parameters of power transmission lines of electric energy. By means of a tutorial, the user is oriented step by step about the input of data and other features of the line, passing to the calculation stage when the data are correctly and completely entered. The output of the results is shown on the screen on a Pi circuit and through graphics that simulate the voltage drop for a determined load.     

Studies on performance parameters of a practical transformer for various utilizations

This work has been conducted at the electrical machines lab in European University of Bangladesh (EUB) for both the experimental and analytical determination of the performance parameters of the transformer for practical utilization. No-load and short-circuit tests were done for calculating the various performances parameters, such as core loss(Hysteresis loss and Eddy current loss), winding copper loss (I²R loss), no load current (I_o),active component of no load current (I_w),magnetizing component of no load current (I_µ), no load resistance (R_o), no load reactance (X_o), equivalent resistance referred to primary (R₀₁), reactance referred to primary (X₀₁) and impedance referred to primary (Z₀₁) and efficiency (η) of the practical transformer. Insulation resistance test was done to check the insulation level, which can identify the presence of major faults. The vector group tests also performed to identify the group numbers for various connection facilities of practical transformer. The subject under consideration was a 1.5 kVA small size, dry type practical transformer. This type of research work like 1.5 kVA practical transformer is not yet done before anywhere. This study will be the guide line for practical utilization of transformer for different purposes.

     

保持细节特征的带纹理模型的高质量简化算法

针对大多算法在简化率较大的情况下易丢失模型细节特征的问题,引入顶点尖锐度的概念,并基于QEM（quadric error metric）折叠代价给出一种改进的折叠代价,能更多地保留模型的细节特征;同时针对大多简化算法不包含纹理处理的问题,引入纹理变化因子,更多保留纹理的细节部分。在简化的基础上,还提出一种网格局部优化算法,解决简化后网格质量不高的问题。实验结果表明,该算法不仅可以保持模型的细节特征和纹理的完整性,同时得到的模型网格质量高。     

Mapping of industry building product model for detailed thermal simulation and analysis

Industry Foundation Classes (IFC) is an emerging data standard that promotes the interoperability of digital building product object models, or building product models (BPMs). Implementing IFC-based applications allows reducing data duplication, especially in the aspect of building geometry. With appropriate data mapping engines, an existing BPM may be readily used for performance-based simulation which requires significantly less effort compared with the traditional means. As a result, it is possible to make use of many available proprietary tools to provide additional services such as early-stage building design support. SEMPER-II (S2), which is an active, multi-domain, space-based, object-oriented design support tool for integrated building performance computing, was developed before the IFC standard was established. In order to perform building performance simulations based on the IFC compliant data models expressed through the eXtensible Markup Language (XML) based formats, an add-on data mapping engine has been developed as a front-end application in S2. The data mapping engine extracts all necessary information embedded in the XML data files, which is obtained directly from IFC compliant CAD drawings, and converts the data into the internal data format of S2.This process demonstrates a seamless link between an IFC compliant industry building product model (e.g. CAD) and the Shared Object Model (SOM) of S2 to facilitate building performance simulations.     

Multilevel methods for discontinuous Galerkin FEM on locally refined meshes

We review shortly the analytical results on multilevel preconditioning for DG methods for elliptic problems. An algorithm is presented which does not loose efficiency under local refinement. It uses subspace smoothing and does not require integration of additional matrices. This algorithm is applied to the interior penalty method and the local discontinuous Galerkin method. A feasible way of implementing the scheme is presented.     

Visualization methods in analysis of detailed chemical kinetics modelling

Sensitivity analysis, principal component analysis of the sensitivity matrix, and rate-of-production analysis are useful tools in interpreting detailed chemical kinetics calculations. This paper deals with the practical use and communication of the sensitivity analysis and the related methods are discussed. Some limitations of sensitivity analysis, originating from the mathematical concept (e.g. first-order or brute force methods) or from the software-specific implementation of the method, are discussed. As supplementary tools to the current methods, three novel visual tools for analysis of detailed chemical kinetics mechanisms are introduced: (a) scaled sensitivity analysis which is especially suited for studying initiation reactions where the span of reaction rates is high; (b) automated generation of reaction pathway plots which provides an immediate graphical illustration of the chemical processes occurring; (c) explorative (or chemometric) analysis of accumulated rate of progress matrices which assist in the identification of reaction subsets. The application of these tools are demonstrated by analysing NOx enhanced oxidation of methane at 700-1200 K.     

PIMS模型中原油保本价的常用测算方法分析与选择

利用PIMS模型中增量法、替换法和影子价格法3种常用原油保本价的测算方法,针对一实际炼厂PIMS模型的同一种原油分别进行了保本价的测算,通过对3种方法测算保本价的对比与分析,认为不同测算方法相对于测算环境优化空间的不同是造成测算保本价出现差异的原因,并基于此归纳了3种测算方法的特点及适用条件。通过对保本价验证条件的分析推导,提出能够确保保本价验证通过的前提条件。在此基础上,提出了根据PIMS模型特点和测算环境的不同情况进行原油保本价测算的流程,并针对测算流程中的不同情况推荐了相应的保本价测算方法。     

压电发电器建模研究

压电发电器的优化设计必须以模型为基础。提出了一种新的压电发电器模型建立方法,运用该方法分别推导出2种结构的压电发电器的性能参数表达式,实验数据与表达式计算结果比较吻合,证明表达式可以用于压电发电器的性能预测。通过对输出参数表达式进行深入讨论,得出压电发电器的一般特性和相关参数与发电器性能参数之间的关系,可指导压电发电器的优化设计。     

An analytical model for multi-epidemic information dissemination

Contemporary distributed systems usually involve the spreading of information by means of ad-hoc dialogs between nodes (peers). This paradigm resembles the spreading of a virus in the biological perspective (epidemics). Such abstraction allows us to design and implement information dissemination schemes with increased efficiency. In addition, elementary information generated at a certain node can be further processed to obtain more specific, higher-level and more valuable information. Such information carries specific semantic value that can be further interpreted and exploited throughout the network. This is also reflected in the epidemical framework through the idea of virus transmutation which is a key component in our model. We establish an analytical framework for the study of a multi-epidemical information dissemination scheme in which diverse ‘transmuted epidemics’ are spread. We validate our analytical model through simulations. Key outcomes of this study include the assessment of the efficiency of the proposed scheme and the prediction of the characteristics of the spreading process (multi-epidemical prevalence and decay).     

How membrane properties shape the discharge of motoneurons: a detailed analytical study

Electrophysiological experiments and modeling studies have shown that afterhyperpolarization regulates the discharge of lumbar motoneurons in anesthetized cats and is an important determinant of their firing properties. However, it is still unclear how firing properties depend on slow afterhyperpolarization, input conductance, and the fast currents responsible for spike generation. We study a single-compartment integrate-and-fire model with a slow potassium conductance that exponentially decays between spikes. We show that this model is analytically solvable, and we investigate how passive and active membrane properties control the discharge. We show that the model exhibits three distinct firing ranges (primary, secondary, and high frequency), and we explain the origin of these three ranges. Explicit expressions are established for the gain of the steady-state current-frequency (I-f) curve in the primary range and for the gain of the I-f curve for the first interspike interval. They show how the gain is controlled by the maximal conductance and the kinetic parameters of the afterhyperpolarization conductance. The gain also depends on the spike voltage threshold, and we compute how it is decreased by threshold accommodation (i.e., the increase of the threshold with the injected current). In contrast, the gain is not very sensitive to the input conductance. This implies that tonic synaptic activity shifts the current-frequency curve in its primary range, in agreement with experiments. Taking into account the absolute refractory period associated with spikes somewhat reduces the gain in the primary range. More importantly, it accounts for the approximately linear and steep secondary range observed in many motoneurons. In the nonphysiological high-frequency range, the behavior of the I-f curve is determined by the fast voltage-dependent currents, via the amplitude of the fast repolarization afterspike, the duration of the refractory period, and voltage threshold accommodation, if present.