Abstract: The structural condition can be identified by the abnormal behavior detection of the vibration response of the structure. The artificial immune system algorithm is an appropriate approach for anomaly detection. Therefore, artificial immune system is introduced to detect the damage in a structure. First, the positive selection algorithm inspired by immune system is discussed. Then the wavelet packet energy spectrums are extracted as features using wavelet packet decomposition with an arbitrary time-frequency resolution. At last, the positive selection algorithm is employed to detect the abnormality of the features. The improvement on the algorithm is implemented, helpful to the reduction of detector number and the acceleration of detection speed. For demonstration, a numerical study on health monitoring of the ASCE benchmark model is performed. The results show that the health condition of the structure can be accurately monitored by the proposed method and it is compared with the negative selection algorithm.