Common Aircraft Surface Material Recognition Method Based on Neural Network

Abstract

The invention relates to a method for identifying common aircraft surface materials based on neural networks. The method comprises the following steps: step 1, collecting a large number of diffuse reflection infrared spectra of multiple common aircraft surface materials; step 2, pretreating the collected spectra; step 3, detecting the materials according to a traditional material detection method,and calibrating the infrared spectra of the corresponding materials to establish an infrared spectrum data and material matching model; step 4, classifying the infrared spectra of a database; and step 5, performing format conversion on the spectrum data, and performing deep learning and training on the database by utilizing a neural network to establish a neural network model. The neural networkis divided into an input layer and a competition layer, each neuron of the network competition layer acquires a response opportunity to an input mode through competition, and finally, only one neuronbecomes a winner. The method can analyze samples in an external field without sample preparation, and the average identification accuracy reaches 85% or more.

Description

technical field [0001] The invention belongs to the field of material identification, and in particular relates to a neural network-based identification method for the surface material of commonly used aerospace vehicles. Background technique [0002] Aircraft materials have a wide range and are divided into airframe materials (including structural materials and non-structural materials), engine materials and coatings, the most important of which are airframe structural materials and engine materials. Non-structural materials include: transparent materials, cabin facilities and decoration materials, accessories and piping materials for hydraulic, air-conditioning and other systems, radome and electromagnetic materials, tire materials, etc. The amount of non-structural materials is small but there are many varieties, including: glass, plastic, textiles, rubber, aluminum alloy, magnesium alloy, copper alloy and stainless steel. [0003] The first airplanes to fly people into ...

AI Technical Summary

Problems solved by technology

[0007] In order to overcome the difficulties in the classification and identification of commonly used aircraft surface materials, the present invention provides an effective identification method suitable for commonly used aircraft surface materials. Classification

Images