摘要
鉴于目前广泛使用的OBDII标准的车载自诊断系统的诊断精度和准确度都有待于进一步的提高,以电控发动机怠速不稳征兆为例,可提出一种将神经网络和D-S证据理论融合的多传感器故障诊断方法,并用于车载自诊断系统诊断数据的融合处理和分析。实验结果表明,该方法能够充分利用各种故障的冗余和互补信息,从而显著提高故障的识别能力。
The precision and accuracy of on-board diagnosis system with OBDII standard which has been widely used at present need to be further improved. In this context, taking the engine idling instability as example, a multi-sensor diagnosis method integrating neural network and D-S evi-dence theory, which is mainly used in on-board forward. The experimental result shows that this and complementation information sufficiently, and diagnosis system＇s data process and analysis, is put method can make use of various then promote the recognition ability fault＇ s redundant obviously.
出处
《交通标准化》
2009年第11期47-51,共5页
Communications Standardization
关键词
神经网络
证据理论
融合
自诊断
ANN
evidence theory
fusing
on-board diagnosis
