柴油机气阀间隙识别的研究

Study on the identification of diesel engines valve clearance

通过模拟柴油机气阀间隙的异常进行试验,采集缸盖表面的振动信号.利用时间序列分析方法对振动信号进行分析,并建立相应的AR(AutoRegressive)模型,从中提取不同气阀间隙下的敏感参数.最后,基于这些敏感参数,利用神经网络进行气阀间隙的识别.分析结果表明,将两种方法的融合,能够实现优势互补,达到更好的识别效果.

By simulating the abnormal clearance of the diesel engine valves, some vibration signals are collected, and the lime series analysis is used to analyze these signals, some sensitive parameters of the different valve clearance are selected At last, based on these parameters, neural networks are used to identify the fault model of the valve clearance. The analysis conclusion indicated that better identification effect can be reached by integrating the two methods.