This paper presents a novel model of intrusion detection based on fuzzy cluster and immune principle,the original RPCL algorithm is modified in order to address the problem of different variability of variables and correlation between variables,the sensitivity to initial number of clusters is also solved in this paper.Especially,this paper uses the extended RPCL algorithm to determine the initial number of clusters in the fuzzy cluster algorithm.The genetic algorithm is used to optimize the radius deviation for the determination of characteristic function of abnormal subspace.Some results are finally reported with some concluding remarks.