基于FNN-GA的车用传感器在线诊断

基于模糊神经网络与遗传算法,提出了柴油机传感器故障的模糊融合诊断策略;运用模糊推理算法,依据不同故障的传感器波形信号,对传感器故障模式进行了判别,验证了故障诊断网络的可靠性。利用柴油机电控平台,进行了柴油机MAP、RPS和TPS的硬故障和软故障等性能试验。结果表明:所设计的传感器故障诊断模型合理,诊断策略具有较好的识别率,可用于柴油机传感器故障在线诊断。

On-Line Diagnosis on Automotive Sensor Based on FNN -GA

The sensor fault diagnostic strategy of the diesel engine was put forward based on fuzzy neural network and the genetic arithmetic.The fault types of the sensors were diagnosed on fuzzy reasoning logic algorithm,the reliability of the fault diagnostic network on the sensors was verified according to the different malfunction signal of the sensors.Based on the simulative electrical control equipments of the diesel engine,the malfunction tests on the hard malfunction and soft malfunction of the sensors,such as MAP,RPS and TPS,were made by fuzzy neural network syncretic strategy.The test result shows that the diesel engine sensors fault diagnosis model is reasonable;the diagnostic strategy has the good resolving power and could be much fitted for the on-line diagnosis of the sensors malfunction.