Research on Abnormal Detection Algorithm of Aerospace TT&C Equipment-Based on Temporal Association Rules

Abstract

Aiming at the problems in traditional time series data association mining, such as huge candidate set space, low mining efficiency, and mining results not meeting the time series requirements, we obtain the key parameters of the operation state of the aerospace TT&C equipment firstly, and uses the temporal association rule mining (TSAR) algorithm based on temporal constraints, establish the candidate set of association rule mining, trim the candidate set space according to temporal constraints, and mine frequent patterns that conform to temporal constraints. Then, the asymmetric J-measure method is used to extract the multi-dimensional association rules with the maximum amount of information. Finally, abnormal detection of aerospace TT&C equipment operation state is realized by mining temporal association rules of key parameters. The experimental results show that, compared with the traditional association rule mining algorithm, TSAR model can effectively mine temporal association rules containing more information, and can also efficiently analyze the abnormal time points of the operation state of aerospace TT&C equipment, providing support for knowledge discovery and anomaly detection of aerospace TT&C equipment.