Zhenjiang vinegar is one of the most representative types of vinegar in China. It is famous for certain characteristics， such as being sour but not astringent， fragant and a little sweet， having a thick color and delicious taste， and improvement with age. At present， Zhenjiang aromatic vinegar has a great diversity of types and quality on the market， and identification of processes and ages is a lack of standardization， which could cause the disorderly competition of the vinegar industry and bring some trouble to the consumers and producers. So we urgently need scientific ways to correct and discriminate the processes and ages of vinegar. In this paper， we used SPME-MS technology to obtain the ion abundance of Zhenjiang vinegar using different processes with five vinegar ages （fresh， six month-aged， two year-aged， three year-aged， and four year-aged）. We used three chemometric methods-linear discriminat analysis（LDA）， support vector machine（SVM） and back-propagation neural network（BPANN） to establish the discriminant model. The results showed the recognition rate of train group and test group in the BPANN model reached 100% and 99%， respectively， when it was used to identifying the different processes or five vinegar ages of Zhenjiang aromatic vinegar. To identifying two different processes and five vinegar ages of Zhenjiang aromatic vinegar simultaneously， the performance of the LDA model was 100%. So the SPME-MS technology combined with chemometric methods could quickly identify the different processes and vinegar ages in Zhenjiang aromatic vinegar.