Complex Variable Methods for Shape Sensitivity of Finite Element Models (PREPRINT)
Abstract
Complex variable methods have some potential advantages over classical finite differencing methods for sensitivity analysis. Two methods, complex Taylor series expansion and Fourier differentiation, are applied and compared to central differencing for shape sensitivity analysis. A two dimensional finite element model with an analytical solution is chosen for use in comparing the accuracy of the methods. It is found that for the accuracy of the model chosen, the error in the sensitivities is primarily defined by the error in the solution, not the error in the sensitivity method.
Document Details
- Document Type
- Technical Report
- Publication Date
- Jan 01, 2010
- Accession Number
- ADA520913
Entities
People
- Andrew Voorhees
- Harry Millwater
- Ronald Bagley
Organizations
- University of Texas at Austin