摘要
研究了一种结合克隆和变异原理的反面选择算法,利用傅立叶变换把时域振动信号转换为频域信号,提取出某一故障的特征频段,基于生物免疫系统的反面选择机理,并利用反面选择算法训练和产生适合于这一故障的检测器集。通过对三种气阀故障的检测,实验结果很好地说明了本算法的有效性,为研究新的故障诊断方法提供了可能。
Negative selection principle based on the natural immune system makes it possible to develop a new fault diagnosis technique. This paper investigates a negative selection algorithm combining the clonal selection principle with the mutation theory. In the present paper, via fourier transformation, the time vibration signals are transformed into the frequency ones, and then corresponding to the particular fault, the characteristic frequency interval of the gas valve is extracted. The detector set is trained and produced by using the negative selection algorithm proposed, and then the data taken from three different faults are tested on three different detector sets, and the result proves the efficiency of the method.
出处
《振动工程学报》
EI
CSCD
北大核心
2006年第4期465-468,共4页
Journal of Vibration Engineering
关键词
免疫系统
故障诊断
反面选择算法
气阀
immune system
fault diagnosis
negative selection algorithm
gas valve
