An electronic nose system based on flow modulation in the case of variable flow rates was designed herein to improve the recognition accuracy and shorten the detection time. The gas response range of sensors to different components and concentrations was maximized by changing the intake flow rate. The age of Chinese yellow wine was classified through adaptive principal component analysis(AD-PCA)to verify this system. The results from AD-PCA were compared with those from support vector machine(SVM)and back-propagation neural network(BPNN). The experimental results showed that among these 5 different ages of Chinese yellow wine, the average correct classification rate of AD-PCA was 93.6%, and that of SVM and BPNN was 92% and 100%, respectively. It was proved that the system could quickly classify the wine age on the basis of a higher accuracy rate, and can shorten the detection time compared with the fixed flow rate.