Aircraft thermal protection system damage diagnosis method based on machine learning algorithm

Abstract

The invention discloses an aircraft thermal protection system damage diagnosis method based on a machine learning algorithm. The method comprises the steps of firstly, acquiring structure strain fielddistribution information by using an embedded optical fiber sensor network; then, positioning an abnormal value position by using a method of combining polynomial fitting with a confidence interval,extracting the amplitude, the signal zero-crossing rate, the signal fineness ratio and other characteristics of an abnormal signal segment, and carrying out self-organizing mapping neural network training; and inputting test data into the self-organizing mapping neural network after training is completed, and specific situations such as the damage position and type are obtained. In the flying process of the aircraft, the neural network can be trained according to real-time strain data reflected by factors such as the service environment, special events and maintenance testing, the damage diagnosis processing speed and recognition precision are improved, and the aircraft heat insulation layer integrity problem is solved.

Description

technical field [0001] The invention belongs to the technical field of aerospace applications, and in particular relates to an online damage identification method for an aircraft thermal protection system based on a machine learning algorithm. Background technique [0002] Thermal protection system is an important part to protect the overall structural safety of space launch vehicles, and its structural integrity has become a key issue in the development of reusable space vehicle equipment. Compared with other structural components, the damage of the thermal protection system of the thermal insulation layer bonded to the main structure is more subtle, resulting in more sudden damage and failure of the thermal protection system. Therefore, for the online identification of bonded thermal insulation layer damage, Positioning and classification are particularly important. [0003] In addition, during the service process of the aircraft, the thermal insulation layer of the therm...

AI Technical Summary

Problems solved by technology

[0004] The traditional non-destructive testing method identifies and locates the damage of the structure when the aircraft is stopped. This method requires the assembly of external equipment, data collection and manual analysis, which takes a long time and increases the cost of testing. At the same time, it cannot distinguish the type of damage.

Images