An abnormality detection method of injection molding machine blocking based on ensemble learning

The invention relates to an injection molding machine blocking abnormal detection method based on ensemble learning, including raw data acquisition, data preprocessing, eigenvalue extraction, materialblocking abnormality classifier construction. At the eigenvalue extraction stage, at first, the time domain characteristics of the working voltage of each module at the position are extracted from the sample library, at last, the two-level anomaly classify is trained to input these eigenvalues, firstly, three basic classifiers are constructed, and the results of the three basic classifiers are voted to select the most frequent results as new feature columns combined with the original feature dataset and transferred to the second-level strong classifier; finally, the result of the strong classifier is the detection value of injection molding machine blocking abnormity. The invention can accurately and quickly find the abnormal condition of the blocking material of the injection molding machine, overcomes the low a