Decision tree algorithm-based transformer chromatographic peak qualitative method and system

The invention discloses a decision tree algorithm-based transformer chromatographic peak qualitative method and system. The method comprises the following steps of: S10, data preparation and selection: a training sample set, a test sample set and a feature attribute set are extracted from an on-site transformer oil chromatography online monitoring device; S20, decision tree model construction: thetraining sample set is trained with a decision tree algorithm, and a decision tree is constructed, a construction process comprises the feature attribute threshold adaptation of the training sample set, the selection of decision tree node feature attributes and the formation of the decision tree on the basis of feature attribute thresholds and node feature attribute recursion; and S30, result testing: the decision tree is tested with the test sample set, whether a result meets a preset standard or not is checked, and if the result meets the preset standard, and the decision tree is used for transformer chromatographi