1,720,975 research outputs found
A NEURAL-NETWORK APPROACH FOR THE SOLUTION OF ELECTRIC AND MAGNETIC INVERSE PROBLEMS
No full textMultilayer neural networks, trained via the back-propagation rule, are proved to provide an efficient means for solving electric and/or magnetic inverse problems. The underlying model of the system is learned by the network by means of a dataset defining the relationship between input and output parameters. The merits of the method are illustrated at the light of three example cases. The first two samples deal with inverse electrostatic problems which are relevant for nondestructive testing applications. In a first problem, a boss on an earthed plane is identified on the basis of the map of potential produced by a point charge. In the second problem, the geometric parameters of an ellipsoid carrying an electric charge are identified. In both cases, database of simulated measurements has been generated thanks to the available analytical solutions. As a sample magnetic inverse problem, the identification of a circular plasma in a tokamak device from external flux measurements is carried out. The results achieved show that the method here proposed is promising for technically meaningful applications
Wavelet tools for improving the accuracy of neural network solution of electromagnetic inverse problems
No full textA neural network model is proposed to treat inverse problems in electromagnetics, which includes wavelet functions to improve local approximation capabilities, This processor couples the advantages of an interpretation of the problem based on "features" to the accuracy derived from using wavelets where local corrections are needed. The combined model allows to cope with singularities of the mapping and to slightly modify the mapping in real time. The detection and characterization of a circular defect in a conducting plate by using eddy current testing is shown to take advantage from the proposed approach in a test case, when unforeseen disturbances are present
A genetic design technique for field correction systems in NMR devices
No full textThe optimal design of a set of actual shimming coils for NMR magnets is addressed. The problem is firstly described in the framework of electromagnetic inverse problems and then regularized on the basis of feasibility considerations, in the limit of negligible cross sections. The effects of finite cross section are then introduced, and an assessment of the results is done. Finally, by employing a genetic, global optimisation technique, a broad range search in the parameters space is performed, leading to solutions not foreseen by the studies based on the thin coil assumption
- …
