Data anomaly detection method and device

• Main Authors: ZHU RONGSONG, GU SHAOFENG, WANG YUXIANG, SHI FENG
• Format: Patent
• Language: Chinese; English
• Published: 23.02.2024
• Subjects: CALCULATING; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS; COMPUTING; COUNTING; ELECTRIC DIGITAL DATA PROCESSING; PHYSICS

The invention provides a data anomaly detection method and device, and the method comprises the steps: obtaining the detection data of a to-be-detected object, the detection data being the data information of a plurality of variables at each collection moment obtained through collection according to the operation condition of the to-be-detected object in a first time range; dividing the detection data into a plurality of first fragment data; determining a plurality of first feature codes corresponding to the plurality of first fragment data through a feature extraction model for the plurality of first fragment data; wherein the first feature code is used for indicating a time feature and a space feature corresponding to the first fragment data; and determining whether an abnormal detection result exists in the detection data or not according to the plurality of first feature codes through an unsupervised clustering model. According to the scheme, the abnormal data existing in the detection data can be accurat