基于多信息融合的航空线路串联故障电弧识别方法

Series Arc Fault Identification Method in Aviation Lines Based on Multi-Information Fusion

航空电气线路工作环境复杂,故障电弧判别可靠性要求较高,而单特征的识别方法诊断效果相对较差。针对这一问题,提出运用多特征信息融合算法对航空低压配电线路故障电弧检测的方案。采用脉宽百分比、变异系数、间谐波均值和小波奇异熵四个特征量对线路电流时域、频域和时频域特征进行提取;依据突变理论建立故障电弧评价模型,对多特征量信息融合求出故障电弧评价指标;根据评价指标,对线路中是否出现故障电弧进行判断。结果表明,正常情况和故障电弧下的评价指标区分明显,易于设定阈值,且该方法适用于多种负载类型和电流等级。

The working environment of aeronautical circuit is complicated and the reliability of arc fault discrimination is relatively high. However, the single feature identification method has relatively poor diagnostic efficiency. To solve this problem, a multi-feature information fusion algorithm for the detection of arc faults in aviation low voltage distribution lines is proposed. The characteristics of the line current time domain, frequency domain and time-frequency domain were extracted by using the four feature quantities of pulse width percentage, coefficient of variation, inter-harmonic mean and wavelet singular entropy. They were taken for information fusion by establishing the evaluation model of arc fault based on mutation principle to obtain arc fault evaluation index, and the arc fault is judged according to the evaluation index. The results show that the evaluation criteria under normal conditions and arc faults are distinct and easy to set thresholds, and this method is applicable to multiple load types and current levels.