The invention relates to a relearning industrial control intrusion detection method based on multi-algorithm fusion parallel. The method comprises steps of acquiring control network data and extracting data packet characteristics; establishing a data feature training set with a time sequence; adopting a support vector machine and an Adaboost algorithm based on a single-layer decision tree CART forparallel detection; reclassifying difference data obtained after the data output by the two methods are compared through an Adaboost algorithm based on a single-layer decision tree CART; and returning the attacked data output after re-classification to a training set of the support vector machine for re-learning, and integrating and outputting the attacked data classified after re-training of thesupport vector machine and normal data obtained after re-classification. Compared with a traditional detection mode based on a white list, the method has the advantage that the detection accuracy ofdetecting the attack behavior in the industrial control network is improved to the maximum extent.