Immune-based fault diagnosis for gas valve

Numerous special working mechanisms provide directions for developing new fault diagnosis methods. In this paper, a negative selection algorithm combining the clonal selection principle with the mutation theory is studied. By comparing the frequency charts of healthy valve and faulty valve, the characteristic frequency interval of gas valve corresponding to a particular fault is extracted. Detector set is trained and produced using the negative selection algorithm proposed, then data from three different faults is tested on three different detector sets. The results prove the efficiency of the method.