"This research highlights the use of moment closure approximation techniques to provide analytical approximations to nonlinear state dependent systems. Nonlinear systems are often difficult to solve and often approximated through the use of simulation. However, Moment Closure has the capability to derive an approximate solution by making a closed form solution out of open set of differential equations representing nonlinear system behavior in both transient and equilibrium state. It is this capability of approximating the closed form solution to complex systems that draws interest in the use of this method within systems engineering. This research further extends to the stability analysis of the derived differential equations using Jacobian Matrix and Eigen Value method. We demonstrate the effectiveness of the method through various examples. We address the issue of using underlying parametric assumptions for closing the open set of differential equations for large systems, which are typically hard to solve. In this research we seek to show that it is better to have a distributional assumption then just having no assumption and neglecting the cumulants arbitrarily. We also show that how this method can be applied to systems of systems with little or no interdependence between the systems, assuming that each and every system can be represented by a single stochastic variable"--Abstract, page iii.
Dagli, Cihan H., 1949-
Engineering Management and Systems Engineering
M.S. in Engineering Management
Missouri University of Science and Technology
viii, 41 pages
© 2012 Ritesh Arora, All rights reserved.
Thesis - Open Access
Nonlinear systems -- Mathematical models
Stability -- Analysis
Print OCLC #
Electronic OCLC #
Link to Catalog Record
Arora, Ritesh, "Parametric moment closure for non-linear state dependent stochastic systems" (2012). Masters Theses. 5320.