A new ACE inhibitory peptides database was self-established.After the structures of peptide samples with different lengths were characterized using amino acid descriptors SVHEHS,the data obtained were treated for standardization by auto cross covariances（ACC）.Then three modeling methods,namely multiple linear regression（MLR）,partial least squares（PLS） and artificial neural network（ANN） were used to establish the models of the QSAR of ACE inhibitory peptides,respectively.The results showed that R2（correlation coefficient） of MLR,PLS and ANN models were 0.744,0.862 and 0.958,Q2LOO（leave-one-out cross-validated correlation coefficient） were 0.532,0.829 and 0.948,and Q2ext（external validated correlation coefficient） were 0.567,0.632 and 0.634,respectively.Hence,the combinations of SVHEHS and the above three modeling approaches were all useful for the QSAR of ACE inhibitory peptides,in which ANN modeling approach is the best.