Grade discrimination of Chinese liquor was benefit for controlling liquor quality and safeguarding the interests of consumers.In the study,the abundance of ions mass-to-charge ratio(m/z) ranging from 55 to 191 of 120 Yanghe Daqu liquor samples were easily collected by the headspace(HS)-solid phase microextraction(SPME)-mass spectrometry(MS) technique,without pre-treatment or chromatographic separation.By combination of partial least squares discriminant analysis(PLS-DA) and stepwise linear discriminant analysis(SLDA) method,14 characteristic ions were finally picked out and the prediction accuracy of cross validation was up to 100%.And then an artificial neural network(ANN) recognition model using the 14 ions as inputs and different grades as outputs was built,whose prediction accuracy was up to 100% within the error range of ±0.3.