A real-time detection method for equipment anomalies based on synchronous data stream compression

Abstract

The invention discloses a real-time detection method for equipment abnormality based on synchronous data stream compression. By collecting the characteristics of each equipment, and then grouping them, and constructing a group data set representing the normal operation state of the group of equipment and a data set representing the normal operation state of the equipment. In this way, the records of the two data sets are used for comparison, and the abnormal detection results are obtained comprehensively, which improves the accuracy of detection. At the same time, considering that the operating status of the equipment is different in different environments, the present invention uses a concept drift detection method based on principal component analysis to detect the operating status data to see if it has evolved, and if it has evolved, re-initialize the two data set, which further improves the detection accuracy. In addition, the present invention uses synchronous data stream compression to reduce the amount of calculation in the comparison process, thereby realizing the real-time detection of equipment abnormalities.

Description

technical field [0001] The invention belongs to the technical field of equipment abnormality detection, and more specifically relates to a real-time detection method for equipment abnormality based on synchronous data stream compression. Background technique [0002] With the development of science and technology, various types of equipment used in the production and use of various fields of the national economy have become increasingly complex and refined. How to detect the status of these devices in real time, judge whether there is an abnormality, and prevent equipment failures will play an inestimable role in reducing equipment failure losses and reducing safety hazards. [0003] Traditional equipment anomaly detection methods, such as anomaly detection based on signal processing, not only require a large amount of expert knowledge, but also cannot detect equipment anomalies in real time. However, equipment operating status records are dynamically generated in real time,...

AI Technical Summary

Problems solved by technology

[0003] Traditional equipment anomaly detection methods, such as anomaly detection based on signal processing, not only require a large amount of expert knowledge, but also cannot detect equipment anomalies in real time. However, equipment operating status records are dynamically generated in real time, and the anomaly detection system must be limited. Under the condition of time and memory, the real-time and dynamic running status records, and fast and accurate analysis and prediction of abnormalities are the core problems that need to be solved in the field of equipment abnormality detection.

Images