Title :
The Choice of regularization parameter by fuzzy inference for magnetic field synthesis
Author :
Ceclan, A. ; Barbulescu, Ct ; Micu, D.D. ; Micu, D. ; Czumbil, L.
Author_Institution :
Tech. Univ. of Cluj Napoca, Cluj-Napoca
Abstract :
The paper illustrates a robust regularization method for numerically synthesize a magnetic field, because the problem proves to be ill-posed and ill-conditioned. Truncated singular value decomposition has to be applied, so as to. reach stable and physically meaning solutions. Fuzzy inference approach is adopted to choose the regularization parameter.
Keywords :
fuzzy reasoning; magnetic fields; singular value decomposition; fuzzy inference; magnetic field synthesis; regularization parameter; robust regularization method; truncated singular value decomposition; Coils; Inference algorithms; Integral equations; Inverse problems; Magnetic fields; Magnetic sensors; Mathematical model; Sensor phenomena and characterization; Shape; Singular value decomposition; field synthesis; fuzzy rules; ill-posed problem; regularization parameter; truncated singular value decomposition;
Conference_Titel :
Optimization of Electrical and Electronic Equipment, 2008. OPTIM 2008. 11th International Conference on
Conference_Location :
Brasov
Print_ISBN :
978-1-4244-1544-1
Electronic_ISBN :
978-1-4244-1545-8
DOI :
10.1109/OPTIM.2008.4602346
