Method for similarity query of abnormal modes of flight data

Abstract

The invention provides a method for similarity query of abnormal modes of flight data, which specifically comprises the step 1 of mode representation process; the step 2 of abnormal mode retrieving process; and the step 3 of similarity query process. By utilizing the mode representation technology based on important points, the method for inquiring the similarity of abnormal modes of flight data not only reduces the dimension of the flight data and reduces the calculation of similarity query, but also effectively removes noise interference in the flight data and improves the accuracy and the reliability of the similarity query. And the method for similarity query of abnormal modes of flight data, provided by the invention, achieves compact partitioning of the abnormal modes of flight databy utilizing competitive clustering technology, resulting in simple and concise index structure and improving the efficiency of the similarity query.

Description

technical field [0001] The invention belongs to the technical field of aircraft fault diagnosis, and in particular relates to a similarity query method for abnormal patterns of flight data. Background technique [0002] In the 21st century, with the continuous injection of high technology, the high integration, high intelligence, and efficiency of analysis and processing of modern aircraft are increasing day by day, and the subsequent safety, reliability, economy and maintenance of aircraft Issues such as security have become increasingly prominent, so fault diagnosis technology has received unprecedented attention. [0003] A large amount of flight data generated during the flight and maintenance of the aircraft (such as data recorded in the flight data recorder and QAR, data collected by the aircraft status monitoring system, data in the take-off report, data in the cruise report, etc.) contains Therefore, it is of great significance to improve the safety and reliability ...

AI Technical Summary

Problems solved by technology

Although there are already some high-dimensional data indexing methods, such as R tree, R ＊ Trees, etc., but these methods are mostly based on the direct division of the data space, and then use the hierarchical structure to index, so when the dimensionality of the data is relatively high, it is likely to cause a large increase in the number of indexes and storage space, which seriously affects the similarity Query efficiency, not suitable for indexes applied to flight data

Although clustering technology can be used to increase the correlation of data in each node in the index structure, thereby reducing the number of layers and storage space of the index, most of the flight data is unmarked data, that is, it is impossible to know which category it belongs to. Therefore, traditional clustering methods cannot be used

Competitive clustering technology was proposed by researchers such as Hichem Frigui in the document "A Robust Competitive Clustering Algorithm With Applications in ComputerVision" in 1999. This method is suitable for dealing with the division of unlabeled high-dimensional data and has been successfully applied. Classification of unlabeled data, but there is no precedent for the establishment of flight data index

Images