Detection and classification of impact-induced damage in composite plates using neural networks

Dua, Rohit; Watkins, Steve Eugene; Wunsch, Donald C.; Chandrashekhara, K.; Akhavan, Farhad

doi:10.1109/ijcnn.2001.939106

Cited by 16 publications

(14 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To provide additional confirmation of the simulation validity, XY shear strain profiles at 2.54 cm from the impact point and the associated contact force profiles were simulated for the impact. The simulation results for all types of profiles were consistent with the experimental measurements in structure and amplitude [27]. These X and Y strain profiles were used to train and test a neural network approach to classifying the type and extent of damage.…”

Section: Finite Element Simulation and Damage Classificationsupporting

confidence: 62%

“…They can learn to process data one way, and when conditions change, the processing can adapt to new conditions. ANN applications for delamination detection in composite structures, including damage assessment and fatigue monitoring, have been extensively studied [24][25][26][27][28][29][30][31][32][33][34][35][36][37][38]. Neural networks have been coupled with advanced sensing technologies to predict and generalize unknown parameters in physical systems.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Impact-induced damage characterization of composite plates using neural networks

Watkins

Akhavan

Dua

et al. 2007

Smart Mater. Struct.

View full text Add to dashboard Cite

Impact-induced damage in fiber-reinforced laminated composite plates is characterized. An instrumented impact tower was used to carry out low-velocity impacts on thirteen clamped glass/epoxy composite plates. A range of impact energies was experimentally investigated by progressively varying impactor masses (holding the impact height constant) and varying impact heights (holding the impactor mass constant). The in-plane strain profiles as measured by polyvinylidene fluoride (PVDF) piezoelectric sensors are shown to indicate damage initiation and to correlate to impact energy. Plate damage included matrix cracking, fiber breakage, and delamination. Electronic shearography validated the existence of the impact damage and demonstrated an actual damage area larger than visible indications. The strain profiles that are associated with damage were replicated using an in-house finite element code. Using these simulated strain signatures and the shearography results, a backpropagation artificial neural network (ANN) is shown to detect and classify the type and severity of damage.

show abstract

Section: Finite Element Simulation and Damage Classificationsupporting

confidence: 62%

Section: Introductionmentioning

confidence: 99%

Impact-induced damage characterization of composite plates using neural networks

Watkins

Akhavan

Dua

et al. 2007

Smart Mater. Struct.

View full text Add to dashboard Cite

show abstract

“…Host materials include concrete, 29 metals, and composites, 27,30 and applications include the associated civil, aerospace, and automotive structures. [20][21][22][31][32][33] In particular, EFPI strain sensors have high dynamic and static sensitivity, a low physical profile, [12][13][14] and good compatibility with fiberglass reinforced plastic ͑FRP͒ composites. 30,34 However, EFPI suffers from nonlinearity, requiring extra processing.…”

Section: Efpi Sensor-based Hmssmentioning

confidence: 99%

“…[6][7][8]12 Strain sensing techniques are shown to be capable of characterizing impact-induced damage, warning of impending weakness in structural integrity, and assessing performance of composite structures. [20][21][22] For modal testing of structures, fiber optic sensors and ANNs have been combined to locate and classify damage from changes in resonant frequencies. 23 A Fourier series neural network ͑FSNN͒ has been employed in obtaining modal frequencies from the fiber optic sensor output.…”

Section: Introductionmentioning

confidence: 99%

Demodulation of extrinsic Fabry-Pe´rot interferometric sensors for vibration testing using neural networks

2004

View full text Add to dashboard Cite

Strain level measurement on a periodically actuated and instrumented structure can provide information about the health of that structure. A simple demodulation system employing artificial neural networks (ANNs) is analyzed for an extrinsic Fabry-Pé rot interferometric (EFPI) strain sensor. The harmonic content of the nonlinear sensor output for the sinusoidal strain case is used to predict the maximum strain level. Implementations of the demodulation system are demonstrated for both simulated and experimental data using back-propagation neural networks. The simulation uses the theoretical response of the EFPI sensor and the experiment uses an EFPI-instrumented smart composite beam to obtain training and testing data. Excitation is provided by a piezoelectric actuator operating from 50 Hz to 1 kHz. System performance is compared for two-stage and single-stage networks and for differing types of data preprocessing. The ANN systems successfully extract the signal harmonics and predict the peak strain levels. Data preprocessing using principal component analysis produces the best accuracy. The architecture of an EFPI-based health monitoring system is proposed.

show abstract

“…In some cases, these parameters are determined by trial and error. However, there are current techniques for obtaining automatic estimations, like the expressed by [18] or [32]. This research follows the steps described by [52].…”

Section: Neural Network's Structurementioning

confidence: 99%

A neural network--based methodology for the recreation of high-speed impacts on metal armours

González‐Carrasco

García-Crespo

Ruíz‐Mezcua

et al. 2011

Neural Comput & Applic

View full text Add to dashboard Cite

The prediction of the consequences of a ballistic impact is highly relevant in the advanced material engineering. Traditionally, the solution of this kind of problems was made by means of experimental tests, analytical models or numerical simulations. In this domain, the particularities of the phenomenon at high speed increase the difficulty of the mathematical resolution of the equations associated, and the complexity of the mechanical behaviour of the materials at high strain rates complicates the numerical simulation of the problem. Therefore, this paper describes a neural network-based methodology applied to recreate the ballistic impact phenomenon. The objective of this study is threefold. Firstly, to obtain the most precise prediction possible, minimizing the amount of data used. Secondly, to discover and analyse the influence of each of the variables on the entire neuronal model. Finally, to compare the precision and performance of this methodology with other alternatives of learning machine. The empirical results have shown that the proposed methodology is an interesting approach to reliably solving ballistic impact problems.

show abstract

Detection and classification of impact-induced damage in composite plates using neural networks

Cited by 16 publications

References 14 publications

Impact-induced damage characterization of composite plates using neural networks

Impact-induced damage characterization of composite plates using neural networks

Demodulation of extrinsic Fabry-Pe´rot interferometric sensors for vibration testing using neural networks

A neural network--based methodology for the recreation of high-speed impacts on metal armours

Contact Info

Product

Resources

About