Some of the material in is restricted to members of the community. By logging in, you may be able to gain additional access to certain collections or items. If you have questions about access or logging in, please use the form on the Contact Page.
Georgiadou, A. (2019). Global Optimization in Stellar Evolution Applications. Retrieved from http://purl.flvc.org/fsu/fd/2019_Fall_Georgiadou_fsu_0071E_15471
Computer modeling is extensively used to probe structure and evolution of stars and planets. These computations allow astrophysicists to connect theoretical models of star formation and evolution to astronomical observations. Because stellar evolution is a highly complex process and long evolutionary timescales only allow one to obtain just a glimpse into the stellar life, finding initial stellar model parameters that lead to specific observed objects requires substantial effort and expert knowledge. Typically probing parameter space of this problem is done using a trial-and-error approach, which is inefficient, incomplete and prone to bias. We propose to decipher this scientific challenge by formulating a suitable constrained global optimization problem in which the optimization problem is solved using either the Controlled Random Search method (CRS) or the MIDACO evolutionary ant-colony opti- mization algorithm. The MESA stellar evolution code is used to calculate the cost func- tion, and problem constraints are a combination of simple ranges of input parameters and observations-dependent penalty function. We use the resulting new package, MESA-GO, to identify initial conditions for progenitor stars of young, nearby, and therefore relatively well-observed supernovae. We present the results for our initial target objects, supernovae SN 1987A and SN 1993J. Luminosity and effective temperature are the observed variables and are part of the cost function. The initial progenitor mass, convective α parameter, semi-convection coef- ficient αsc and metallicity Z are the control variables (input variables). The objective is to minimize the difference between an observed state for the star and a model state using control variables (input variables) given the constraint that the model is a reacting system of hyperbolic conservation laws, with bounds on initial conditions.
A Dissertation submitted to the Department of Mathematics in partial fulfillment of the requirements for the degree of Doctor of Philosophy.
Bibliography Note
Includes bibliographical references.
Advisory Committee
Tomasz Plewa, Professor Co-Directing Dissertation; Mark Sussman, Professor Co-Directing Dissertation; Anke Meyer-Baese, University Representative; Giray Okten, Committee Member; Nick Cogan, Committee Member.
Publisher
Florida State University
Identifier
2019_Fall_Georgiadou_fsu_0071E_15471
Georgiadou, A. (2019). Global Optimization in Stellar Evolution Applications. Retrieved from http://purl.flvc.org/fsu/fd/2019_Fall_Georgiadou_fsu_0071E_15471