Abnormality monitoring method and system applied to material purification system

The invention relates to the technical field of abnormity monitoring of material purification systems, and relates to an abnormity monitoring method and system applied to a material purification system. According to the method, the system operation event and the anomaly diagnosis data are integrated into the sample learning sequence, and the sample learning sequence is provided for the target candidate neural network to obtain the second anomaly diagnosis data, so that learning can be performed from diversified data sources, and the identification accuracy of the abnormal condition is improved; the network weight information of the target candidate neural network is optimized to generate an offline anomaly diagnosis network and corresponding network training information, and a machine learning algorithm is utilized to perform self-optimization, so that the intelligent level of an anomaly monitoring system is improved; when the offline anomaly diagnosis network meets the network convergence requirement, the ne