Dr. Harry Millwater
Professor of Mechanical Engineering
The University of Texas at San Antonio
Will present a talk titled:
High Order Sensitivity And Uncertainty Analysis Using The Hypercomplex Finite Element Method
Abstract: The use of hypercomplex variables and algebra within a finite element code provides a convenient method to obtain highly accurate, high order (1st through 5th) sensitivities of the displacements with respect to arbitrary shape, material and loading parameters. Subsequent post processing leads to sensitivities of the strains, stresses, strain energy, etc. The attraction of the method is that no new equations or finite element formulation is required – the method augments current formulations. The University of Texas at San Antonio (UTSA) has developed the hypercomplex finite element code ZFEM using a user-element within the Abaqus commercial finite element software. ZFEM has been applied successfully in a number of disciplines such as fracture mechanics, plasticity, heat transfer, and structural dynamics, among others. ZFEM has recently been coupled with a moment-based uncertainty quantification method such that estimates of the mean, variance, skewness, and kurtosis can be obtained using only the ZFEM outputs. Applications to a number of disciplines will be presented.
Date: April 30, 2021
Start Time: 4:30 pm
Zoom: https://clarkson.zoom.us/j/98943528473?pwd=Z2VSekJ6THF1YWo2c1ZNWE05SmVJUT09
Meeting ID: 989 4352 8473
Passcode: 293416
Bio: Dr. Harry Millwater is the Samuel G. Dawson Professor of Mechanical Engineering at the University of Texas at San Antonio. His area of interest includes probabilistic mechanics and sensitivity methods as applied to structures subject to fatigue and fracture. He is the originator of the hypercomplex finite element method ZFEM. His grants include funding from AFOSR, AFRL, ARO, DoD, FAA, NIH, NNSA, NRC, ONR, and PNNL. He is the Principal Investigator for the development of the Smart suite of probabilistic fatigue codes for the FAA, and a past developer of the Nessus and Darwin probabilistic software codes.