Structures and Materials Commons^™

Neural Network Fatigue Life Prediction In Notched Aluminum Specimens From Acoustic Emission Data, Muhammed Arif Okur

Master's Theses - Daytona Beach

This purpose of this research was to identify fatigue crack growth and predict failure for 7075-T6 aluminum notched bars under uniaxial tensile loading using acoustic emission (AE) data. The experiments performed in this study extend the results obtained by previous researchers who used maximum cyclic loads of 4,000, 3,000, and 2,000 pounds at a stress ratio of R = 0.0 and a frequency of 1 Hz to perform the fatigue tests. For this research the cyclic load remained at 2,000 pounds, but an additional ten specimens were tested in order to increase the amount of AE data available to the …

Compression After Impact Load Prediction In Graphite/Epoxy Laminates Using Acoustic Emission And Artificial Neural Networks, Anthony Michael Gunasekera

Master's Theses - Daytona Beach

The purpose of this research was to investigate the effectiveness of artificial neural networks (ANNs) in predicting the compression after impact (CAI) load of graphite/epoxy laminates from acoustic emission (AE) nondestructive testing (NDT) data. Thirty-four 24-ply bidirectional woven cloth laminate coupons were constructed and impacted at various energy levels ranging from 8 to 20 Joules, generating barely visible impact damage (BVID). Acoustic emission data were acquired as the coupons were compressed to failure. Not having been analyzed by previous experimenters, several noise tests were also performed to determine the impact of external noise on acoustic emission data during testing. Once …

Prediction Of Fatigue Life In 7075-T6 Aluminum From Neural Network Analysis Of Acoustic Emission Data, Nicholas S. Spivey

Master's Theses - Daytona Beach

Through the use of an acoustic emission (AE) data acquisition system, a Kohonen self-organizing map, and a back-propagation neural network, AE data from 7075-T6 aluminum specimens were used to classify failure mechanisms and predict the number of fatigue cycles to failure. AE waveforms were captured from 40 notched tensile specimens during the low-cycle fatiguing process. A Kohonen self-organizing map and initial data filters were used to classify the data into two distinct failure mechanisms, plane strain and plane stress fracture, plus a third less prevalent mechanism. These results were employed to construct a back-propagation neural network to predict the number …

Neural Network Burst Pressure Prediction In Composite Overwrapped Pressure Vessels From Acoustic Emission Data, Seth-Andrew T. Dion

Master's Theses - Daytona Beach

Composites have grown in importance in the aerospace industry where high specific strength is a priority. Weight reduction in space vehicles is critical because of the exorbitant cost associated with placing objects into space. Major weight savings have been obtained by switching from all metal pressure vessels to composite overwrapped pressure vessels (COPVs). Due to the nature of composites, current nondestructive analysis procedures for COPVs are not adequate for assessing structural integrity. As such, new methods must be developed. Presented herein is one such method.

A method for burst pressure prediction using parametric filtering of acoustic emission (AE) data along …

Neural Network Fatigue Life Prediction In 7075-T6 Aluminum From Acoustic Emission Data, Emeka Chigozie Ibekwe

Master's Theses - Daytona Beach

The objective of this research was to classify acoustic emission (AE) -data associated with fatigue cracks in aluminum fatigue specimens and to use early cycle life AE data to predict failure in such members. An AE data acquisition system coupled with a Kohonen self organizing map and a back propagation neural network were used to perform the analysis. AE waveforms were recorded during fatigue cycling of twenty-four notched 7075-T6 aluminum specimens using broad-band piezoelectric transducers. A Kohonen self organizing map was used to classify the AE flaw growth signals. The signals were classified into three categories based on their AE …

Ultimate Strength Prediction In Fiberglass/Epoxy Beams Subjected To Three-Point Bending Using Acoustic Emission And Neural Networks, Michele D. Dorfinan

Master's Theses - Daytona Beach

The research presented herein demonstrates the feasibility of predicting ultimate strengths in composite beams subjected to 3-point bending using a neural network analysis of acoustic emission (AE) amplitude distribution data. Fifteen unidirectional fiberglass/epoxy beams were loaded to failure in a 3-point bend test fixture in an MTS load frame. Acoustic emission data were recorded from the onset of loading until failure. After acquisition, the acoustic emission data were filtered to include only data acquired up to 80 percent of the average ultimate load.

A backpropagation neural network was constructed to predict the ultimate failure load using these AE amplitude distribution …

Torque Limit Of A Mechanical Fastener In A Graphite/Epoxy Joint, Kristian M. Kostreva

Master's Theses - Daytona Beach

Presently there is a lack of confidence amongst engineers when specifying the preload of a mechanical fastener in a composite joint due to a lack of a fundamental knowledge base regarding the behavior of composites under fastener compressive load. As such, a novel experimental procedure was developed herein to determine the through-the-thickness compressive (TTTC) material properties. A total of 206 property tests were performed on four different graphite/epoxy material systems. The results confirmed that TTTC material properties vary with fiber orientation, laminate thickness, fiber volume fraction, and even laminate surface finish. Hence, the 'rule of mixtures' provides a poor estimate …

Modeling Of Acoustic Emission Failure Mechanism Data From A Unidirectional Fiberglass/Epoxy Tensile Test Specimen, Daniel R. Lendzioszek

Master's Theses - Daytona Beach

The purpose of this work was to model the acoustic emission (AE) flaw growth data that resulted from the tensile test of a unidirectional fiberglass/epoxy specimen. The data collected and stored during the test were the six standard AE quantification parameters for each event. A classification neural network was used to sort the data into five failure mechanism clusters. The resulting frequency histograms of the sorted data were then mathematically modeled herein using the three types of Johnson distributions: bounded, lognormal, and unbounded. These provided a reasonably good fit for all six AE parameter distributions for each of the five …

Classification Of In-Flight Fatigue Cracks In Aircraft Structures Using Acoustic Emission And Neural Networks, Christopher Lee Rovik

Master's Theses - Daytona Beach

The research encompassed within this paper deals with the analysis and classification of fatigue cracks in aircraft structures. The particular structure that was examined was the vertical tail section of a Cessna T-303 Crusader aircraft. The analysis was performed using the nondestructive evaluation technique known as acoustic emission (AE), as well as the artificial intelligence of neural networks. Data were taken in a controlled laboratory environment as well as in a flying testbed aboard the aircraft.

The first part of the research involved the analysis of a typical aircraft structure in a controlled laboratory environment. This support structure was fabricated …

Classification Of Acoustic Emission Signals From An Aluminum Pressure Vessel Using A Self-Organizing Map, Weldon Paul Thornton

Master's Theses - Daytona Beach

Acoustic emission nondestructive testing has been used for real-time monitoring of complex structures. All of the structures were made of materials at least 0.070 inch thick. The purpose of this research was to demonstrate the feasibility of using neural networks to classify acoustic emission signals gathered from a pressure vessel made of 2024-T3 aluminum 0.040 inches thick, i.e. thin aluminum sheet. AE waveforms were recorded during fatigue cycling of one pressure vessel using a wide band transducer and a digital oscilloscope connected to a computer. The source for each signal was determined using two narrow band transducers and a LOCAN-AT …