An efficient non linear algorithm predictive model of a robust optimal portfolio

2019 ◽  
Vol 13 (17) ◽  
pp. 805-813
Author(s):  
I. Irakoze ◽  
F. Nahayo
Keyword(s):  
Predictive Model ◽  
Optimal Portfolio ◽  
Linear Algorithm ◽  
Non Linear

scholarly journals Flicker spreading in a transmission network

2020 ◽  
Vol 69 (1) ◽  
pp. 18-22
Author(s):  
Tonko Garma ◽  
Denisa Galzina
Keyword(s):  
Linear Regression ◽  
Predictive Model ◽  
Cost Effective ◽  
Transmission Network ◽  
Transmission Grid ◽  
Key Concepts ◽  
Regression Approach ◽  
Non Linear

This paper reports the flicker spreading in the transmission network. Chapter 1 presents introduction containing brief background and key concepts, followed by description of the corresponding instrumentation in Chapter 2.  Key contribution of the paper is elaborated in Chapters 3 and 4.  Chapter 3 reports measurements of the flicker magnitude along the 400 kV, 220 kV and 110 kV transmission grid for various distances from flicker origin on 400 kV grid, and Chapter 4 gives cost-effective predictive model, enabling estimation of the flicker magnitude for arbitrary selected origin-to-spot distance base on non-linear regression approach. Paper is extension of the work presented at Smagrimet 2019 conference.

scholarly journals A linear algorithm for solving non-linear isothermal ice-shelf equations

2014 ◽  
Vol 7 (2) ◽  
pp. 1829-1864
Author(s):  
A. Sargent ◽  
J. L. Fastook
Keyword(s):  
Differential Equations ◽  
Iterative Algorithm ◽  
Linear Equations ◽  
Direct Methods ◽  
Two Dimensional ◽  
Ice Shelf ◽  
Linear Algorithm ◽  
Manufactured Solutions ◽  
Non Linear ◽  
Non Linear Equations

Abstract. A linear non-iterative algorithm is suggested for solving nonlinear isothermal steady-state Morland–MacAyeal ice shelf equations. The idea of the algorithm is in replacing the problem of solving the non-linear second order differential equations for velocities with a system of linear first order differential equations for stresses. The resulting system of linear equations can be solved numerically with direct methods which are faster than iterative methods for solving corresponding non-linear equations. The suggested algorithm is applicable if the boundary conditions for stresses can be specified. The efficiency of the linear algorithm is demonstrated for one-dimensional and two-dimensional ice shelf equations by comparing the linear algorithm and the traditional iterative algorithm on derived manufactured solutions. The linear algorithm is shown to be as accurate as the traditional iterative algorithm but significantly faster. The method may be valuable as the way to increase the efficiency of complex ice sheet models a part of which requires solving the ice shelf model as well as to solve efficiently two-dimensional ice-shelf equations.

Sign in / Sign up

Export Citation Format

Share Document