Objective To develop a nanocolorimetric sensor array for identifying different types of edible oils.Methods The colorimetric sensor array was composed of chemically responsive dyes and modified with porous silica nanospheres (PSNs) to improve its sensitivity and stability. Four types of edible oils were classified and identified using the nanocolorimetric sensor. Principal component analysis (PCA) was used to reduce the dimensionality of the feature data from the four oil samples, and the reduced data were then imported into three classification models, i.e., support vector machine (SVM), K-nearest neighbor (KNN), and linear discriminant analysis (LDA).Results The SVM classification model established in the experiment effectively distinguished the four types of edible oils, with a 4% improvement in test set accuracy compared to the other two methods.Conclusion The nanocolorimetric sensor array technology can be applied for the non-destructive detection of edible oil types.