Diagnosis of aerospace structure defects by a HPC implemented soft computing algorithm
📝 Abstract
This study concerns with the diagnosis of aerospace structure defects by applying a HPC parallel implementation of a novel learning algorithm, named U-BRAIN. The Soft Computing approach allows advanced multi-parameter data processing in composite materials testing. The HPC parallel implementation overcomes the limits due to the great amount of data and the complexity of data processing. Our experimental results illustrate the effectiveness of the U-BRAIN parallel implementation as defect classifier in aerospace structures. The resulting system is implemented on a Linux-based cluster with multi-core architecture.
💡 Analysis
This study concerns with the diagnosis of aerospace structure defects by applying a HPC parallel implementation of a novel learning algorithm, named U-BRAIN. The Soft Computing approach allows advanced multi-parameter data processing in composite materials testing. The HPC parallel implementation overcomes the limits due to the great amount of data and the complexity of data processing. Our experimental results illustrate the effectiveness of the U-BRAIN parallel implementation as defect classifier in aerospace structures. The resulting system is implemented on a Linux-based cluster with multi-core architecture.
📄 Content
Diagnosis of aerospace structure defects by a HPC implemented soft computing algorithm
Gianni D’Angelo, Salvatore Rampone University of Sannio Dept. of Science and Technology Benevento, Italy {dangelo, rampone}@unisannio.it
Abstract— This study concerns with the diagnosis of
aerospace structure defects by applying a HPC parallel
implementation of a novel learning algorithm, named U-BRAIN.
The Soft Computing approach allows advanced multi-parameter
data processing in composite materials testing. The HPC parallel
implementation overcomes the limits due to the great amount of
data and the complexity of data processing. Our experimental
results illustrate the effectiveness of the U-BRAIN parallel
implementation as defect classifier in aerospace structures. The
resulting system is implemented on a Linux-based cluster with
multi-core architecture.
Keywords— Non-destructive testing; learning algorithm;
parallel computing; HPC; signature-based classifier; eddy current.
I.
INTRODUCTION
The use of composite materials in the aerospace industry is
growing rapidly, especially in the production of the
components the use of which sees them subjected to heavy
loads and efforts. Due to their unique mechanical properties,
namely, high strength-to-weight ratio, high fracture toughness,
and excellent corrosion resistance properties, they are used at
critical points in the construction of an aircraft [1,2]. They are
widely used in the outer covering of the aircraft, such as flaps,
hatches, sides of the engine, floors, rudders, elevators, ailerons
etc. The composite material design and manufacturing
technologies have matured to a level that Boeing Company is
using composite material for 50% of the primary structure in its
787 program. Unfortunately, there is a great variety of possible
manufacturing defects that regards those materials [3]. The
most widespread types of defects are the following:
•
Delamination between plies of outer skin, parallel
to surface;
•
Matrix crack;
•
Disbanding between the outer skin and the
honeycomb core;
•
Fiber fracture;
•
Cracked honeycomb core parallel to the inspection
surface;
•
Crushed honeycomb core in parallel to the area;
•
Disbonding between inner skin and honeycomb
core;
•
Fluid ingress in honeycomb core.
•
Damages induced by the stress, environment
influences and others.
•
Wear, scratch, indentation and cleft
•
Creep deformation.
These defects are difficult to diagnose and analysis is
strongly influenced by many factors that may also arise from
the complexity of manufacturing processes. In addition, some
techniques of inspection and/or some detection equipment may
have systematic errors or accidental ones. The presence of
defects and damages pose a significant threat to the safety of
composite structures. Composite materials are mostly used in
aerospace structures, and their structural reliability and safety is
particularly critical.
Non-Destructive Testing (NDT) allows one to implement a
control over the material at different stages of its evolution and
permits to safeguard the integrity of the structure during the
analysis. Visual and strike method, optical holography, X-ray,
ultrasonic wave, eddy current testing and infrared detection, X-
ray and ultrasonic C-scan are the most methods used. Due to
the heterogeneity of the composite structure, the Non-
Destructive Testing of composites are very complex and
sometimes several methods will take to test the same
component [4]. For this reason, the accuracy of diagnosis of
composite materials is determined not only by physical
methods to obtain experimental data but also with
mathematical models and advanced methods of data processing
[5]. The analysis of the data, generally obtained from tests
based on multi-parameter control, is one of the possible ways
to increase the effectiveness and reliability of non-destructive
testing of composites. The methods of spectrum analysis and
pattern recognition are often used in multi-parametric control
for data processing [6]. However, the application of these
methods requires sophisticated techniques for processing
signals that lead to the solution of nonlinear equations complex
with a high number of variables [7]. The difficult and
sometimes impossible solution of these equations lead to a
reduction in the efficiency of the system of non-destructive
testing. These difficulties also do not allow the automation of
the test and deprive their of the same dynamism typical of a
system able to adapt to changes in the parameters of the testing
system at run-time. Non-destructive testing of composites
should be performed with methods able to collect the most
comprehensive information about new defects, expand existed
base of defects and increase diagnostics system precision in
runtime. Furthermore data processing in defects diagnosis has
to deal with great amount of data and numerous elements are
pro
This content is AI-processed based on ArXiv data.