Cecchini-Brigi, Andres

Profile Picture

Filters

2005 - 200912010 - 20141
Reset filters

Settings

Citations

Publication Search Results

Now showing 1 - 2 of 2
  • Thumbnail Image
    Publication
    Fatigue life assessment of sandwich composite hulls under repeated slamming loads
    (2014) Cecchini-Brigi, Andres; Serrano-Acevedo, David; College of Engineering; Just, Frederick; Suárez, Luis E.; R. López, Ricardo; Department of Civil Engineering; Huérfano, Víctor A.
    Slamming loads induced by waves on ships can cause severe damages on structural members which may compromise integrity and safety. Therefore, better understanding of the slamming problem is a topic of interest in marine applications. In addition, sandwich composite structures are being widely used in the marine industry due to their extremely high flexural stiffness and light weight. This thesis presents a novel predictive computational technique for fatigue life assessment of sandwich composite hulls subject to repeated slamming loads. The slamming modeling was approached in the framework of the two dimensional water entry problem, in which, the fluid is assumed ideal and potential, the angle of incidence between the fluid and the structure is small and gravitational effects are neglected. Numerical models based on explicit finite element analysis (FEA) were developed in LS-DYNA to simulate a single impact event. The multimaterial Arbitrary-Lagrangian-Eulerian (ALE) formulation and the Eulerian-Lagrangian penalty coupling algorithm were used. Initially, the study focused on the impact of rigid hulls with constant velocity. Pressure distribution on the contact surface was investigated and compared with analytical solutions and experimental data. Later, the analysis was extended to include metallic and sandwich composite hulls. As a result, stress time histories for a single impact were obtained at critical locations. To simulate the effect of multiple impacts, these stresses were extrapolated using Peak Over Threshold (POT) analysis assuming a gamma distribution for the exceedances. Then, the Rainflow cycle counting method was used to reduce the complex slamming stresses to a series of simple cyclic stresses. For each stress level, the degree of damage induced in the structure was calculated from the S-N curves and the individual contributions were combined using a damage accumulation model. For metallic hulls, linear Miner’s rule was used. For sandwich composite hulls two damage models were investigated: Miner’s rule, based on number of cycles, and the non-linear stiffness degradation approach, based on reduction of fatigue (shear) modulus. The selection of damage accumulation models was based on the predominant mode of failure of the structure’s material. As a result of this study, it was found that sandwich composite hulls are more susceptible to fatigue failure due to slamming loads than steel hulls. Fatigue life of sandwich hulls was limited by the high shear stresses in the core.
  • Thumbnail Image
    Publication
    Damage detection and identification in sandwich composites using neural networks
    (2005) Cecchini-Brigi, Andres; Serrano-Acevedo, David; College of Engineering; Just-Agosto, Frederick A.; Basir Shafiq, Abdul; Department of Mechanical Engineering; Godoy, Luis A.
    Marine, aerospace, ground and civil structures can receive unexpected loading that may compromise integrity during their life span. Therefore, improvement in detecting damage can save revenue and lives depending upon the application. The prognostic capability is usually a function of the examiner’s experience, background and data collection during the evaluation. Nondestructive evaluation (NDE) methods are varied and specific to a given type of system (material, damage type, loading and environmental scenarios). As a result, one method of damage detection alone cannot examine all possible conditions and may even give false readings. In other words, by using more than one NDE technique, the probability of ensuring a more accurate detection increases. This work examined various existing NDE techniques to assess damage in sandwich composites structures including: vibration modal analysis, transient thermal response, and acoustic emission. Sandwich composites consisting of two carbon fiber/epoxy matrix face sheets laminated onto a urethane foam core were experimentally and analytically characterized using vibration, and thermal response to detect the presence of various types of damages. A neural network (NN) approach that uses vibration and thermal signatures to determine the condition of a composite sandwich structure is purposed. The data used to train a probabilistic neural network (PNN) were provided by numerical simulations. Literature offers substantial evidence of the validity of each of the chosen damage detection schemes separately. However, we will show that these methods can work jointly to complement each other in detecting the state of a sandwich composite structure.