基于故障树和现场测试的EMC故障诊断方法

发布时间：2018-10-20 08:03

【摘要】：随着电子设备的广泛使用,电磁兼容问题日趋严重,但是对于电磁兼容故障诊断却没有一套简便有效的理论方法。因此,本文提出了基于故障树理论和现场测试技术的电磁兼容故障诊断方法。使用故障树分析电子设备的电磁兼容薄弱环节,然后根据分析结果对故障设备进行现场测试以定位故障源。本文首先对电磁兼容故障进行了分析,并结合通用故障树的建树理论,研究了电磁兼容故障树的建立方法和一般形态。同时文中针对故障树分析中,底事件故障概率难以确定的问题,提出了基于层次分析法的故障概率分析方法:首先对影响底事件发生概率的因素进行递阶层次构造,对同一层次的因素进行成对比较,建立判断矩阵并对其进行一致性检验,并根据判断矩阵确定各指标的权重,最后对各指标进行量化,将量化后的指标和权重相乘,即可计算出底事件故障概率。文中还研究了故障树的定性分析和定量分析的方法,其中定性分析包括故障树最小割集的计算,定量分析包括最小割集重要度和底事件重要度的计算。对于电磁兼容故障树的建立和分析,以及底事件故障概率的计算本文都给出了具体的算例。故障树的最小割集表明了电磁兼容故障发生的最小故障模式,可用来指导测试的进行。使用现场测试对各集合按照重要度从高到低进行测试,可快速的定位电磁兼容故障的具体原因。现场测试是对故障设备在工作状态下进行测试,相比于标准测试有无可比拟的优越性。现场测试硬件技术有两个问题:一个是抗饱和技术,一个是微弱信号检测。本文分别给出了问题分析和解决办法。得到测试数据后,需要先对测试数据进行预处理以使测试数据易于分析。预处理包括消噪和环境剔除两步,文中对于消噪给出了小波消噪和奇异值消噪两种方法。而环境剔除给出了基于阈值的剔除方法。最后提取数据的有用信息,包括峰值信息的提取和包络信息的提取。根据提取的数据特征即可完成对电磁兼容故障的分析与诊断。
[Abstract]:With the wide use of electronic equipment, the problem of EMC is becoming more and more serious, but there is no simple and effective theoretical method for EMC fault diagnosis. Therefore, an EMC fault diagnosis method based on fault tree theory and field test technology is proposed. The fault tree is used to analyze the EMC weakness of the electronic equipment, and then the fault equipment is tested on the spot according to the analysis results to locate the fault source. In this paper, the EMC fault is analyzed, and the general configuration of EMC fault tree is studied based on the general fault tree theory. At the same time, aiming at the problem that the fault probability of bottom event is difficult to determine in fault tree analysis, a fault probability analysis method based on analytic hierarchy process (AHP) is proposed. The factors of the same level are compared in pairs, the judgment matrix is established and its consistency is checked, and the weight of each index is determined according to the judgment matrix. Finally, each index is quantified, and the quantized index and weight are multiplied. The fault probability of bottom event can be calculated. The qualitative analysis includes the calculation of the minimum cut set of the fault tree, and the quantitative analysis includes the calculation of the minimum cut set importance degree and the bottom event importance degree. Examples are given for the establishment and analysis of EMC fault tree and the calculation of fault probability of bottom event. The minimum cut set of the fault tree shows the minimum fault mode of EMC and can be used to guide the test. By using field test to test each set from high to low in importance, the causes of EMC faults can be quickly located. The field test is to test the fault equipment under the working condition, which has unparalleled superiority compared with the standard test. There are two problems in field testing hardware technology: one is anti-saturation technology, the other is weak signal detection. The problem analysis and solution are given in this paper. After getting the test data, it is necessary to preprocess the test data to make the test data easy to analyze. The pretreatment includes two steps: noise elimination and environment elimination. Wavelet and singular value denoising methods are presented in this paper. Environmental culling gives a threshold-based culling method. Finally, useful information is extracted, including peak value information and envelope information. The analysis and diagnosis of EMC faults can be completed according to the extracted data features.
【学位授予单位】：西安电子科技大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：TN03

【相似文献】