Andris Chate

A method for condition monitoring and localization of defects in mass-produced structural members using supervised learning is presented. An example for the effectiveness of the developed method comprises cantilevered carbon composite plate. In a numerical finite element model, the plate is partitioned into zones and a point mass is put on several locations within each zone. Point mass is treated as a pseudo-defect locally modifying structural properties of the plate. For each act of mass application, strain values are recorded and serve as defect-sensitive feature. Two variables of classification are tested – two different supervised learning algorithms (linear discriminant and non-linear k-nearest neighbours) and a limited number of strain data points per class which is varied in the range of 2 to 9 points. Several query points are simulated and subjected to classification in terms of belonging to particular zones of the partitioned plate. This step can be treated as a defect localization. It is shown that only 2 strain readings per class are sufficient for defect localization. The methodology is experimentally validated on a cantilevered carbon composite prepreg of the same dimensions and properties.

The main purpose of this study was to investigate the micro-indentation test method on the branched epoxy novolac resin (ENR) and to validate it by means of finite element simulation. At the first stage, the micro-indentation test was carried out with the purpose of establishment of the Young's modulus of ENR by the empirical relation to polymer materials. Experiments were performed by using the Vickers-type indenter. At the second stage, the finite element model was used to model the load-displacement curve of the material with the use of the initial values of elastic modulus and yield strength of material. The effectiveness of the use of these properties was investigated by using the planning of experiments and response surface technique. Full factorial design method was used to perform a series of numerical experiments. Then, a comparison of the results of micro-indentation test and finite element simulations was performed in order to compare and validate the values of Young's modulus obtained from two different techniques. Good correlation between the experimental and numerical Young's modulus was observed.

The paper presents numerical results of prestress loss in the prestressed hollow core slabs by using few natural frequencies. Loss of prestress has been investigated by the 3D finite element method, by using ANSYS software. Loss of prestress is modelled by decreasing of prestressing forces in the 7 wire strand that simulated damage or severing of prestress strands. The prestressing force in the strands is applied as an initial stress. The effectiveness of using frequency percentage changes as an indicator of the loss of prestress and effectiveness of the proposed method are investigated using the planning of experiments and response surface technique. After selection of equation of regression, the parametric studies are carried out in order to determine the influence of the slab length, concrete strength and initial stress on the frequency percentage changes and study the dependence of the percentage changes on these parameters. The present study shows that natural frequencies percentage change is dependent on the span of slab and compressive strength of concrete.

Problem of damage localization in a cantilevered composite plate using strain data is tackled. The plate is partitioned into a number of zones and a mass equal to 9.43 % of plate's mass is put in each of these zones subsequently. The composite plate is harmonically excited via piezoelectric actuator with a series of driving frequencies equal to natural frequencies of the plate extracted in modal analysis test. In each of these events, the mechanical strains are recorded with strain gauges. Mean values of strain time series are calculated and used as a feature for building a classification model in which zones of the plate serve as classes. Linear discriminant classifier initially yielding the best classification accuracy (90.7 %) is selected. The classification model is validated by selecting 5 query points on the plate and localizing the unknown query points in terms of classes (zones) of the plate. Overall, the unknown query points were classified successfully with only slight misclassification for the point at the boundary between 2 zones.

In this paper, a damage identification algorithm based on continuous wavelet transform of one-dimensional structures exploiting mode shapes is presented. Numerical models of aluminium and carbon composite beams, containing a mill-cut and impact damage, respectively are considered for this study. Wavelet scalogram is used to obtain the transform coefficients at different wavelet scales and is subsequently normalized in order to emphasize locations with largest coefficients. Variance of normalized wavelet scalogram is computed along the axis of the beam yielding sharp peaks in the zones corresponding to damage in beams. This operation excludes wavelet scale factors as variables for damage localization problems. The universal threshold is applied to filter out lower amplitude peaks that do not indicate damage. These results are summed up for all nodes of beams and all wavelet functions that are analysed in this paper in order to also exclude the number of different wavelet functions as another variable for damage localization. The universal threshold is applied the second time to yield the final result on the locations of damage. Results suggest that the proposed damage localization method is a fast and reliable tool for damage detection in one-dimensional metal and composite beam structures.