Load and Boundary Condition Calibration Using Full-field Strain Measurement

For critical load bearing structures, it is often necessary to experimentally determine the load distribution on the structure so that accurate finite element models can be developed for stress and fatigue life predictions. An inverse problem approach is presented here for computing or calibrating the loads and boundary conditions acting on a structure. This enables the creation of more accurate finite element models, especially for structures that have complicated load distribution and compliant boundary conditions. The method presented here involves minimizing the least square error between the strains computed using the finite element model and the strains and displacements obtained experimentally. The nodal loads and the compliance at fixed boundaries are treated as the variables in the optimization problem. The compliance is modeled as springs attached at the nodes that are on the boundary where the structure is restrained. The method is verified by computing the loads and boundary conditions when displacements, maximum shear strain or both are available at large number of points on the surface of the structure. The experimental data set was generated using the luminescent photoelastic coating (LPC) technique.     

Nondestructive Evaluation of Crack Depth in Concrete Using PCA-compressed Wave Transmission Function and Neural Networks

Cracks in concrete are common defects that may enable rapid deterioration and failure of structures. Determination of a crack’s depth using surface wave transmission measurement and the cut-off frequency in the transmission function (TRF) is difficult, in part due to variability of the measurement data. In this study, use of complete TRF data as features for crack depth assessment is proposed. A principal component analysis (PCA) is employed to generate a basis for the measured TRFs for various simulated crack (notch) cases in concrete. The measured TRFs are represented by their projections onto the most significant PCs. Then neural networks (NN), using the PCA-compressed TRFs, are applied to estimate the crack depth. An experimental study is carried out for five different artificial crack (notch) cases to investigate the effectiveness of the proposed method. Results reveal that the proposed method can effectively estimate the artificial crack depth in concrete structures, even with incomplete NN training.     

Earthquake Forecasting Using Neural Networks: Results and Future Work

In 1994, a new earthquake forecasting method was developed, that integrated in a neural network several forecasting tools that had been originally developed for financial analysis. This method was tested with the seismicity of the Azores, predicting the July, 1998, and the January, 2004, earthquakes, albeit within very wide time and location windows. Work is beginning to integrate physical precursors in the neural network, in order to narrow the forecasting windows     

Engine Control using Neural Networks: A New Method in Engine Management Systems

To be able to meet the demands of low emissions and fuel consumption ofmodern combustion engines, new ways have to be found to control thecombustion. We use new sensors to measure the pressure in the combustionchamber and analyze this signal with a neural network in order to receiveseveral form factors which can be used to control the ignition timing. Theneural network is trained off line with measured data and used on line toderive the form factors. The proposed algorithm can be computed in real timeon conventional digital signal processors and adapted to new engines withvery little effort.     

Using a Lumped Parameter Dynamic Model of a Rock Bolt to Produce Training Data for a Neural Network for Diagnosis of Real Data

Ground anchorage systems are used extensively throughout the world as supporting devices for civil engineering structures such as bridges and tunnels. The condition monitoring of ground anchorages is a new area of research, with the long term objective being a wholly automated or semi-automated condition monitoring system capable of repeatable and accurate diagnosis of faults and anchorage post-tension levels. The ground anchorage integrity testing (GRANIT) system operates by applying an impulse of known force by means of an impact device that is attached to the tendon of the anchorage. The vibration signals that arise from this impulse are complex in nature and require analysis to be undertaken in order to extract information from the vibrational response signatures that is relevant to the condition of the anchorage. Novel artificial intelligence techniques are used in order to learn the complicated relationship that exists between an anchorage and its response to an impulse. The system has a worldwide patent and is currently licensed commercially.A lumped parameter dynamic model has been developed which is capable of describing the general frequency relationship with increasing post-tension level as exhibited by the signals captured from real anchorages. The normal procedure with the system is to train a neural network on data that has been taken from an anchorage over a range of post-tension levels. Further data is needed in order to test the neural network. This process can be time consuming, and the lumped parameter dynamic model has the potential of producing data that could be used for training purposes, thereby reducing the amount of time needed on site, and reducing the overall cost of the system's operation.This paper presents data that has been produced by the lumped parameter dynamic model and compares it with data from a real anchorage. Noise is added to the results produced by the lumped parameter dynamic model in order to match more closely the experimental data. A neural network is trained on the data produced by the model, and the results of diagnosis of real data are presented. Problems are encountered with the diagnosis of the neural network with experimental data, and a new method for the training of the neural network is explored. The improved results of the neural network trained on data produced by the lumped parameter dynamic model to experimental data are shown. It is shown how the results from the lumped parameter dynamic model correspond well to the experimental results.     

Neural networks for tip correction of spherical indentation curves from bulk metals and thin metal films

Spherical indentation is widely used to determine a variety of important mechanical properties from small volumes. However, the available nanoindenter tips mostly deviate from the perfect spherical shape making the application of analysis methods developed for perfect spheres uncertain. In this paper, neural network-based methods are presented that are used to correct force-depth curves measured with such indenter tips. Finite element simulations for imperfect and perfect spherical tips with varying material behaviour are used to train the neural networks, which solve the inverse problem of mapping the true tip shape and the measured force-depth curve to one that corresponds to a perfect spherical indenter. Solutions are provided for bulk materials and thin films. The method has been verified experimentally on nanocrystalline nickel and a copper film on a titanium substrate for different spherical tips.     

Extraction of cohesive-zone laws from elastic far-fields of a cohesive crack tip: a field projection method

A solution method of an inverse problem is developed to extract cohesive-zone laws from elastic far-fields surrounding a crack-tip cohesive zone. The solution method is named the “field projection method (FPM).” In the process of developing the method a general form of cohesive-crack-tip fields is obtained and used for eigenfunction expansions of the plane elastic field in a complex variable representation. The closing tractions and the separation-gradients at the cohesive zone are expressed in terms of orthogonal polynomial series expansions of the general-form complex functions. The series expansion forms a set of cohesive-crack-tip eigenfunctions, which is complete and orthogonal in the sense of the interaction J-integral in the far field as well as at the cohesive-zone faces. The coefficients of the eigenfunctions in the J-orthogonal representation are extracted directly, using interaction J-integrals in the far field between the physical field of interest and auxiliary probing fields. The path-independence of the interaction J-integral enables us to identify the cohesive-zone variables, i.e. tractions and separations, and thus the cohesive-zone constitutive laws uniquely from the far-field data. A set of numerical algorithms is developed for the inversion method and the results from numerical experiments suggest that the proposed algorithms are well suited for extracting cohesive-zone laws from the far-field data. The set includes methods to find the position and size of a cohesive zone. Further included are discussions on error analysis and stability of the inversion scheme.