Abstract:

Since current conventional negative selection algorithm does not pay attention to the occurrence of autoimmune detector, and the presence of autoimmune detector will reduces the accuracy of fault diagnosis and other analogous areas, a method which can discover and eliminate pathogenicity detector was proposed. We specially designed a rule to detecting the presence of pathogenicity detector. With the help of MATLAB software, we simulated irregular dist
ributed selfdata, annulus distributed selfdata and nearly circle distributed selfdata. Results show that our method can effectively eliminate 90% of the pathogenic detector, enhance the robustness and adaptive ability of the immune negative selection algorithm.

Key words: artificial immune system, negative selection, autoimmune detector, algorithm optimization, algorithm simulation