Classification of barely visible impact damage in composite laminates using deep learning and pulsed thermographic inspection

نویسندگان

چکیده

Abstract With the increasingly comprehensive utilisation of Carbon Fibre-Reinforced Polymers (CFRP) in modern industry, defects detection and characterisation these materials have become very important draw significant research attention. During past 10 years, Artificial Intelligence (AI) technologies been attractive this area due to their outstanding ability complex data analysis tasks. Most current AI-based studies on damage field focus segmentation depth measurement, which also faces bottleneck lacking adequate experimental for model training. This paper proposes a new framework understand relationship between Barely Visible Impact Damage features occurring typical CFRP laminates corresponding controlled drop-test impact energy using Deep Learning approach. A parametric study consisting one hundred with known material specification identical geometric dimensions were subjected drop-impact tests five different levels. Then Pulsed Thermography was adopted reveal subsurface specimens recorded patterns temporal sequences thermal images. convolutional neural network then employed train models that aim classify captured photos into groups according Testing results trained from time windows lengths evaluated, best classification accuracy 99.75% achieved. Finally, increase transparency proposed solution, salience map is introduced learning source produced models.

برای دانلود باید عضویت طلایی داشته باشید

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

منابع مشابه

Thermographic Inspection of Composite Materials

Thermography is a non-contact NDT technique for inspection of materials in wide application areas, including corrosion detection in metals, and delamination, porosity and moisture detection in composite materials. Composites often are highly anisotropic in nature. This anisotropy coupled with low thermal diffusivity in thickness direction, severely restricts detection of deeper defects in compo...

متن کامل

NASA Thermographic Inspection of Advanced Composite Materials

As the use of advanced composite materials continues to increase in the aerospace community, the need for a quantitative, rapid, in situ inspection technology has become a critical concern throughout the industry. In many applications it is necessary to monitor changes in these materials over an extended period of time to determine the effects of various load conditions. Additionally, the detec...

متن کامل

Ultrasonic NDE Techniques for Impact Damage Inspection on CFRP Laminates

This study investigates the different types of ultrasonic inspection, when applied to assess the impact damage in Carbon Fiber Reinforced Plastic CFRP laminates. The study surveys the two most used ultrasonic testing approaches first; namely the pulse-echo and the through-transmission modes. Then, the manuscript discusses and analyzes the enhanced ultrasonic testing methods that use the polar s...

متن کامل

Modelling low velocity impact induced damage in composite laminates

The paper presents recent progress on modelling low velocity impact induced damage in fibre reinforced composite laminates. It is important to understand the mechanisms of barely visible impact damage (BVID) and how it affects structural performance. To reduce labour intensive testing, the development of finite element (FE) techniques for simulating impact damage becomes essential and recent ef...

متن کامل

ذخیره در منابع من


  با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید

ژورنال

عنوان ژورنال: Neural Computing and Applications

سال: 2023

ISSN: ['0941-0643', '1433-3058']

DOI: https://doi.org/10.1007/s00521-023-08293-7