Open Access
 Issue JNWPU Volume 39, Number 1, February 2021 71 - 76 https://doi.org/10.1051/jnwpu/20213910071 09 April 2021

This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

## 1 DC-DC变换器失效模式分析

 图1基于Sepic结构的DC-DC变换器PSPICE模型

## 2 DC-DC变换器故障诊断框架

 图2DC-DC变换器故障分类技术路线图

## 3 基于多模型融合的DC-DC变换器实时故障诊断

### 3.1 基于神经网络的故障分类

 图3BP神经网络拓扑结构图

### 3.2 基于KNN的故障分类

K最近邻(k-nearest neighbor, KNN)算法是一种有效的分类算法。它以对象间的距离作为衡量指标进行分类, 一般使用欧氏距离或曼哈顿距离:

KNN算法描述可如下:

1) 计算测试数据与各个训练数据之间的距离;

2) 按照距离的递增关系进行排序;

3) 选取距离最小的K个点;

4) 确定前K个点所在类别的出现频率;

5) 返回前K个点中出现频率最高的类别作为测试数据的预测分类。

### 3.3 基于PSO-SVM的DC-DC变换器故障诊断模型

 图4PSO优化SVM结构参数流程图

## 4 仿真验证

MOS管漏极电压信号如图 6所示:

 图5输出电压信号部分统计量变化曲线
 图6MOS漏极电压信号部分统计量变化曲线
 图7二极管p极电压信号部分统计量变化曲线
 图8特征向量主成分分析图(95%)

 图9神经网络模型分类结果
 图10KNN模型分类结果
 图11PSO-SVM模型分类结果

## References

1. Xu Mingxuan. Research on fault analysis method and fault diagnosis technology of aircraft avionics system[J]. China Plant Engineering, 2018, 399(14): 102–103 [Article] (in Chinese) [Google Scholar]
2. Platus D L. Negative-stiffness-mechanism vibration isolation system[C]//Proceedings of the SPIE-the International Society for Optical Engineering, 1999: 98–105 [Google Scholar]
3. Song Fuchao, Hou Wenkui, Shi Long. The information-enhanced fault diagnosis system design of avionics power supply module[C]//2013 International Conference on Quality Reliebility, Rish, Maintenance, and Safety Engineering, 2013 [Google Scholar]
4. Izadian A, Khayyer P. Application of Kalman filters in model-based fault diagnosis of a DC-DC boost converter[C]//36th Annual Conference on IEEE Industrial Electronics Society, Glendale, AZ, 2010: 369–372 [Google Scholar]
5. Park T, Kim T. Novel fault tolerant power conversion system for hybrid electric vehicles[C]//2011 IEEE Vehicle Power and Propulsion Conference, Chicago, 2011: 1–6 [Google Scholar]
6. Kim S Y, Nam K, Song H, et al. Fault diagnosis of a ZVS DC-DC converter based on DC-link current pulse shapes[J]. IEEE Trans on Industrial Electronics, 2008, 55(3): 1491–1494 [Article] [Google Scholar]
7. Ribeiro E, Cardoso A J M, Boccaletti C. Fault-tolerant strategy for a photovoltaic DC-DC converter[J]. IEEE Trans on Power Electronics, 2013, 28(6): 3008–3018 [Article] [Google Scholar]
8. Sheng H, Wang F, Tipton C W. A fault detection and protection scheme for three-level DC-DC converters based on monitoring flying capacitor voltage[J]. IEEE Trans on Power Electronics, 2012, 27(2): 685–697 [Article] [Google Scholar]

## All Figures

 图1基于Sepic结构的DC-DC变换器PSPICE模型 In the text
 图2DC-DC变换器故障分类技术路线图 In the text
 图3BP神经网络拓扑结构图 In the text
 图4PSO优化SVM结构参数流程图 In the text
 图5输出电压信号部分统计量变化曲线 In the text
 图6MOS漏极电压信号部分统计量变化曲线 In the text
 图7二极管p极电压信号部分统计量变化曲线 In the text
 图8特征向量主成分分析图(95%) In the text
 图9神经网络模型分类结果 In the text
 图10KNN模型分类结果 In the text
 图11PSO-SVM模型分类结果 In the text

Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.

Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.