第一，本文研究了故障特征的提取技术。
First, in this paper, we study the technique of extracting fault features.
分析了高压电动机的故障特征和现有保护的不足。
The fault character and insufficiency of the existing protection for the HV motor are analyzed.
电机故障特征相量的确立是进行故障诊断的前提。
The establishment of fault feature vector is a precondition to fault diagnosis.
事实上，这软件是装载故障特征，以防止数据丢失。
In fact this software is loaded with failsafe features to prevent loss of data.
介绍了离心式压缩机旋转失速的故障机理及故障特征。
The fault mechanism and its characteristics of the rotating stall of a centrifugal compressor were introduced in detail.
在建模过程中，利用特征函数强化故障特征作为网络输入。
As network input, the fault feature is enhanced by eigen-function in modeling.
该故障特征在现场诊断时得到验证，并成功应用于现场诊断。
This fault characteristics of rota ting stall have been successfully valida ted practical diagnosis.
应用特征评估的原理，选择最适合于进行故障诊断的故障特征。
Using the principle of feature evaluation, choose the most suitable for fault diagnosis fault features.
在机械故障诊断中，对机器状态信号进行处理可得到故障特征集。
Fault features set can be obtained by processing machine condition signal in machine fault diagnosis.
试验结果表明，这种方法能有效地提取滚动轴承的故障特征，诊断其故障。
The experiment shows that the method can extract effectively the fault characteristics of rolling bearing and detect its failure.
最后，利用概率神经网络技术进一步从关联故障特征中辨识出初始故障源。
Finally, PNN method is used to identify the primary fault sources from the features of correlative faults.
本文的主要内容包括以下几个方面：1论述了大型旋转机械典型故障特征。
The main parts of this dissertation are as follows: 1 Have expounded the main characteristics of typical troubles about the large-scale revolving machinery.
本文研究了支承在挤压油膜阻尼器上的转子系统碰磨故障特征的提取方法。
A method for extracting the features of rub and impact in rotor systems supported on squeezed film dampers is studied in this paper.
各单一选线判据从原始故障数据中提取出各自需要的故障特征量并离散化。
The fault characteristic quantity with each selection method is extracted from the original data and discretized.
与传统方法不同，该方法利用两个节点电压之间的关系函数作为故障特征。
Unlike other diagnosis methods, the reported approach uses the voltage relation function between two nodes as fault characters for predefined faults in the fault set.
采用有限元方法对发生松动故障的转子系统进行模拟，得到系统的故障特征。
With FEM used to simulate the rotor system on which the looseness fault is occurring, the fault characteristics are obtained as follows.
给出了该方法的理论分析，故障特征量的选取，神经网络设置和训练的具体步骤。
Theoretical analysis, choice of fault characteristics and practical procedure of neural network setting and training are given out.
分析了转子断条故障特征在电机启动电磁转矩信号中的表现形式及频率变化规律。
The expression and frequency variety law of the rotor broken bar fault feature component in the startup electromagnetic torque signal are analyzed in detail.
利用双正交小波基将齿轮的故障振动信号分解到时频域，并提取出齿轮的故障特征。
This paper mainly introduces that the fault vibrating signal of gears was decomposed into time-frequency domains by double-orthogonal wavelet analysis and the fault feature of gears was picked up.
在分析空间电源系统电路结构和故障特征基础上，建立起基于模型在线故障诊断系统。
After analyzing the circuit structure and fault characteristics of the aircraft power system, the authors built an on-line model-based fault diagnosis system.
随后计算出代表故障特征的振动信号的峰值指标、峭度指标、脉冲指标和裕度指标等参数。
Then peak index, kurtosis index, impulsion index and tolerance index of the vibration signal are calculated. These parameters represent the characteristic of failures.
滚动轴承的振动信号是非平稳信号，用常规方法难以从振动信号中提取有效故障特征信息。
The fault feature information can not be obtained availably by regular methods to extract the features of nonstationary vibration signals produced by fault rolling bearings.
由于轧机振动信号受到强烈的噪声干扰，给故障特征的有效识别和准确提取带来很大困难。
Due to strong noise in vibration signals of rolling mills, it is very difficult to identify their fault location and to extract their feature.
本文讨论了小波变换及小波变换的奇异性检测理论，并将其应用于暂态行波故障特征的分析。
The paper discuss the singularity detection theory of wavelet transform, this theory is applied to analysis of fault characteristics of the transient traveling waves.
结果表明：利用故障信号的特征向量与故障特征表，通过相似性原理进行故障诊断是有效的。
The actual diagnosis result shows that the fault diagnosis is effective by using the fault signal feature vector and fault table and similarity theory.
对滚动轴承的内圈、外圈故障信号的分析结果表明本文方法可以有效地提取滚动轴承故障特征。
The analysis results from roller bearing signals with inner-race or out-race faults show that the diagnosis approach could extract fault characteristics effectively.
同时利用贝叶斯网络实现概率推理，便于描述故障特征的变化及对变压器故障原因的快速分析。
At the same time, probability reasoning can be realized by BN, which can be used to describe changes of fault symptoms and analyze fault reasons of transformer.
将提取的暂态零序电流故障特征作为条件属性，信号增强倍数作为决策属性，构成一个决策系统。
The extracted fault characteristics of transient zero sequence current is looked as condition attributes and the multiple of signal enhancement is.
在数值仿真和轴承故障特征选择中，采用新方法在保证诊断精度的同时，可以节省大量选择时间。
The composite method saves more computing time than the wrapper method with holding the classification accuracy in data simulation and experiment on bearing fault feature selection.
在数值仿真和轴承故障特征选择中，采用新方法在保证诊断精度的同时，可以节省大量选择时间。
The composite method saves more computing time than the wrapper method with holding the classification accuracy in data simulation and experiment on bearing fault feature selection.
应用推荐