Data mining on a transformer partial discharge data using the self-organizing map

Although experts all over the world have investigated methods for partial discharge (PD) detection and classification for over 50 years, until now there is still no universal method available for this purpose. Even in the future, it would not be possible to define a universal method for unambiguous classification and localization of PD sources in complex insulating systems (e.g. transformers or generators) due to the unlimited variations of PD source type and its location. This paper deals with PD signals obtained by remote radiometric measurements performed on a power transformer. Extensive PD measurements were performed in the substation. Since fingerprints of the remote radiometric measurements are not available, the formation of clusters with similar features obtained from captured PD data is crucial. A data mining technique known as self-organizing map (SOM) for the analysis and interpretation of captured PD data is used. Since each signal contains a large number of samples, before applying the SOM, the dimensionality reduction based on principal component analysis is performed. As a result of a data mining process a clear separation of PDs emanating from a transformer and discharges emanating from its surrounding is achieved     

Integrated power transformer diagnosis using hybrid fuzzy dissolved gas analysis

For diagnosing power transformer faults, this study uses hybrid fuzzy technology to integrate five dissolved gas analysis (DGA) approaches: IEEE Standard Doernenburg and Roger ratio methods, IEC60599 Standard ratio method, Electric Technology Research Association of Japan (ETRAJ) Standard ratio method, and Duval triangle method. The IEC TC 10 database and substation equipment maintenance and management web system (SEMMWS) of Taiwan Power Company (TPC) are then used for simulation and verification. Practical cases demonstrate that the integrated approach is more accurate than conventional methods. © 2015 Institute of Electrical Engineers of Japan. Published by John Wiley & Sons, Inc.     

Condition assessment of power transformer onload tap changers using wavelet analysis and self-organizing map: field evaluation

An onload tap changer (OLTC) is the most maintenance intensive subassembly on a power transformer. Vibration monitoring is an effective technique that can be used to assess the condition of an OLTC nonintrusively. The authors have developed a condition monitoring system for common types of OLTCs that enables the condition of tap changer contacts and associated drive system to be inferred from vibration signals. A number of prototype systems have been installed onto OLTCs in distribution zone substations for field trials. Particular emphasis has been given to the detection of faults in a particular type of older tap changer that had been prone to a range of faults associated with the switching contacts and drive mechanism. For this type of tap changer, it has been shown to be possible to determine not only that the tap changer is aging but also to identify the particular part that is degrading.     

面向电力变压器状态评价的油中溶解气体监测数据补全方法

基于马尔可夫理论充分考虑相邻时间点系统在所有状态间的转移特性,提出了一种基于马尔可夫模型的变压器油中溶解气体数据补全方法,将油中溶解气体数据时间序列转化为在不同状态间转移的马尔可夫链,利用正、反向的状态转移矩阵计算得到油中溶解气体数据的补全值。从数据挖掘的角度建立了油中溶解气体数据质量的综合评估体系,从多个角度对数据补全的效果进行评估,并基于D-S证据融合理论融合各个角度的评估结果,得到综合评估结果。利用所提方法对某变压器100组油中溶解气体数据中25组随机缺失值进行补全,结果表明补全后的数据与实际值相似度可以达到99.999%。进一步地,验证其中15组极值点、跃变点处缺失数据补全效果,经过综合评估,补全后的数据与实际值相似度可以达到98.956%。经过验证表明所提方法能够在不改变数据特征的前提下对变压器油中溶解气体的缺失值进行准确的补全,有利于提高变压器状态评估方法的准确性。     

基于溶解气体分析的电力变压器在线监测与诊断技术

油中溶解气体分析是评估变压器绝缘状态和分析变压器绝缘故障的重要依据。在介绍基于油中溶解气体分析的变压器在线监测与诊断技术原理的基础上,进一步总结了当前在线监测与诊断系统的研究现状及存在的问题:系统主要是在线监测油中气体成分及超阈值报警,而对故障性质、种类、定位及发展趋势预测等诊断尚不具备或很不完备,故障诊断需靠离线分析;因缺乏相应的理论指导,在线监测数据在正常情况下应满足的规律还很模糊,装置自身的精确度不能保证。对今后的研究提出了展望,指出如何通过对在线数据的挖掘找到变压器故障诊断的相应理论依据是进一步研究的核心。     

基于自组织映射网络的峰值负荷预测方法

应用扩展自组织映射网络研究了电力系统峰值负荷预测问题。在传统的Kohonen自组织映射(SOM)网络的学习算法的基础上,为了提高电力系统峰值负荷预测的精度,进一步提出了一种扩展的自组织映射算法。在这个SOM网络中,除了权矩阵外,还有一个输入输出对的局部梯度(Jocobian)矩阵也被存储在神经元中。这样,在输出空间中梯度信息围绕输出权值产生了一个一阶扩展,便可得到一个输出的改进估计值。同时,提出了一个Jocobian矩阵的生成算法。最后采用纽约市的电力负荷数据为研究对象,证明了所提出方法的有效性。     

基于离散灰色模型的变压器油中溶解气体浓度预测

油中溶解气体浓度分析被广泛用于监测油浸电力设备的早期故障。将改进灰色模型引入变压器油中溶解气体浓度预测。分析了灰色预测模型GM(1,1)和离散灰色预测模型DGM(1,1)两者之间的关系,讨论了模型预测的准确性和稳定性。分别用GM(1,1)模型和DGM(1,1)模型对变压器油中溶解气体浓度进行建模预测。对比分析结果表明:GM(1,1)模型从离散形式到连续形式的跳跃使得模型不够稳定,随着发展系数的增加,预测精度下降。DGM(1,1)模型是GM(1,1)模型的精确形式,具有更高的预测精度和较好的稳定性,被推荐替代GM(1,1)模型预测变压器油中溶解气体浓度。     

A self-organizing fuzzy logic based protective relay-an applicationto power transformer protection

In the paper the fuzzy logic based multi-criteria protective relay for a three-phase power transformer is presented. Twelve criteria used to stabilize the relay are introduced and integrated by means of multi-objective decision-making methods. The introduced protection scheme contains several internal functions and coefficients which are subject to off-line pre-installation self-setting. Three unique procedures are delivered for self-adjusting the relay, and thus, making it self-organized. The results of testing show significant gains in both sensitivity and selectivity of the self-organized relay compared with traditional approaches     

A new fuzzy logic approach to identify power transformer criticality using dissolved gas-in-oil analysis

Dissolved gas analysis (DGA) of transformer oil is one of the most effective power transformer condition monitoring tools. There are many interpretation techniques for DGA results however all current techniques rely on personnel experience more than analytical formulation. As a result, the current techniques do not necessarily lead to the same conclusion for the same oil sample. A significant number of DGA results fall outside the proposed codes of the ratio-based interpretation techniques and cannot be diagnosed using these methods. Moreover, ratio methods fail to diagnose multiple fault conditions due to the mixing up of produced gases. To overcome these limitations, this paper introduces a new fuzzy logic approach that aids in standardizing DGA interpretation and identifies transformer critical ranking based on DGA data. The approach relies on incorporating all traditional DGA interpretation techniques (Roger, Doerenburg, IEC, key gas and Duval triangle methods) into one expert model. In this context, DGA results of 338 oil samples of pre-known fault conditions that were collected from different transformers of different rating and different life span are used to establish the model. Traditional DGA interpretation techniques are used first to analyze the DGA results to evaluate the consistency and accuracy of each method in identifying various faults. Results of this analysis were then used to develop the proposed fuzzy logic model. The model is validated using another set of DGA data that were collected form previously published papers.     

基于自组织神经网络的变压器故障诊断

变压器故障诊断实质上是属于一种模式识别,基于类内样本与类中心的距离的不同而对类中心的偏移产生不同影响的思想,改善了传统ART2网络存在模式飘移的不足。首先对变压器DGA故障样本的输入矢量进行扩展处理,然后用改进后的ART2网络对样本进行诊断。通过与传统的三比值法和BP神经网络的诊断结果对比,验证了该方法的有效性和可行性。     

基于自组织神经网络的变压器故障诊断

变压器故障诊断实质上是属于一种模式识别,基于类内样本与类中心的距离的不同而对类中心的偏移产生不同影响的思想,改善了传统ART2网络存在模式飘移的不足.首先对变压器DGA故障样本的输入矢量进行扩展处理,然后用改进后的ART2网络对样本进行诊断.通过与传统的三比值法和BP神经网络的诊断结果对比,验证了该方法的有效性和可行性.     

基于自组织映射神经网络的市场清算电价预测

市场清算电价预测是电力市场中交易决策的基础.人工神经网络是电价预测较为理想的方法,但依然存在一些问题,如样本训练有时需要很长时间,存在收敛问题,特别是当样本特征量不明显的时候,这种现象更为突出.针对这一问题,利用自组织映射的聚类特性将历史数据进行特征分类和筛选处理,处理后形成的新数据用于训练三层BP神经网络,仿真结果表明,经过这种数据处理后,网络的收敛速度得到了显著提高,且预测效果良好.     

变压器油中溶解气体实时在线监测系统的研制

设计了一种新型的变压器油中溶解气体实时在线监测系统.系统由三部分组成:油气分离、多组分气体检测和控制系统.在油气分离部分,通过采用特富龙膜和优化的油气回路设计,实现了1 h/次的脱气周期;将固体氧化物高温燃料电池(SOFC)引进到微量气体检测,取得的气体体积分数为0.1μL/L,通过与色谱双柱连用实现了多组分气体的单通...     

基于自组织映射神经网络的市场清算电价预测

市场清算电价预测是电力市场中交易决策的基础。人工神经网络是电价预测较为理想的方法,但依然存在一些问题,如样本训练有时需要很长时间,存在收敛问题,特别是当样本特征量不明显的时候,这种现象更为突出。针对这一问题,利用自组织映射的聚类特性将历史数据进行特征分类和筛选处理,处理后形成的新数据用于训练三层BP神经网络,仿真结果表明,经过这种数据处理后,网络的收敛速度得到了显著提高,且预测效果良好。     

变压器油中溶解气体在线监测方法

介绍了一种在线检测的方法,实时分析油中溶解气体的含量,对现场油气分离、气体检测、数据处理、自动控制、数据远程传输等关键技术的研究方法进行了介绍。     

大型变压器油中溶解气体在线监测技术进展

介绍了近年来国内外大型变压器油中溶解气体在线监测技术的最新进展。结合在线监测原理．描述了其监测过程。重点阐述了油气分离技术和气体分离与检测技术的发展．讨论了各种方法的技术特点和应用前景。通过对多种油气分离方法的性能比较,认为高分子透气膜法最适合变压器油中溶解气体在线监测。概括了多种基于油中溶解气体分析DGA(Dissolved Gas Analysis)技术的变压器故障诊断方法．并指出在线监测数据和离线监测数据获得的条件不同,两种数据存在差异,其处理方式也应该有所不同。     

基于SDS-SSA-LSTM的变压器油中溶解气体浓度预测

油中溶解气体浓度预测对变压器早期故障检测至关重要。为了提高预测精度,本文提出了奇异谱分析(SSA)结合长短期记忆网络(LSTM)的预测模型。首先针对传统序列分解做法中的数据泄露问题,提出一种基于SSA逐步分解的采样策略,然后基于该策略将特征复杂的原始油中溶解气体浓度序列分解为特征相对单一的趋势分量与波动分量,最后利用LSTM网络对各个分量分别进行单步和多步预测。累加各分量的预测值,得到原气体浓度的预测结果。算例表明,相较于单一LSTM,本文所提模型在实验天数内整体的预测精度更高。     

Generalized fuzzy inference neural network using a self-organizing feature map

A new model for generalized fuzzy inference neural networks (GFINN) is proposed in this paper. The networks consist of three layers: an input-output layer, an if layer, and a then layer. In each layer, there are the operational nodes. A GFINN can perform three representative fuzzy inference methods by changing the connectivity and the operational nodes. There are three learning processes in a GFINN: a self-organizing process, a rule-integration process, and a LMS learning process. In the rule-integration process, a GFINN employs two feature maps in order to integrate appropriate rules effectively. Computer simulations were carried out to show the superiority of a GFINN over back-propagation networks. © 1998 Scripta Technica, Electr Eng Jpn, 125(3): 40–49, 1998     

电力变压器的故障和瓦斯保护

针对电力变压器在运行中产生的故障的原因以及排除的方法等做出了简要的分析,笔者认为正确的、合理的故障排除法对于变压器设备能够正常运行起到促进作用。同时结合瓦斯保护的几个方面进行阐述。