System design by constraint adaptation and differential evolution

Abstract

A simple optimization procedure for constraint based problems which works without an objective function is described. The absence of an objective function makes the problem formulation particularly simple. The new method lends itself to parallel computation and is well suited for tasks where a family of solutions is required, trade-off situations have to be dealt with or the design center has to be found. ________________________________________

Voltage Profile Improvement by Capacitor Placement and Control in Unbalanced Distribution Systems Using Differential Evolution Algorithm

“…However, it has not been regarded in most of the previous works [4][5][6][7][8][9][10][11][12][13][14][15][16][17][18]. On the other hand, in the methods presented in [6][7][8][9][10][11][12][13][14][15]17] the loss reduction has been accomplished only by using the fixed capacitors. Moreover, in many previous methods, the medium voltage capacitors which are more expensive than low voltage ones are used [2,3,[5][6][7][8][9][10][11][12][13][14][15][16][17][18].One of the important issues in the placement of capacitors is considering the load variations of the network.…”

“…Therefore, it is useful to investigate the capacitor placement for unbalanced distribution networks [1][2][3]. However, it has not been regarded in most of the previous works [4][5][6][7][8][9][10][11][12][13][14][15][16][17][18]. On the other hand, in the methods presented in [6][7][8][9][10][11][12][13][14][15]17] the loss reduction has been accomplished only by using the fixed capacitors.…”

Simple Feasibility Rules and Differential Evolution for Constrained Optimization

“…This pseudofeasible region is shrunk each generation until it matches the real feasible region. Also, Storn [15] proposed to use an aging concept in order to avoid that a solution remains in the population too many generations. Furthermore, he modified the original DE algorithm because when a child is created and it is not better than the parent subjectto-replace, another child is created.…”

“…Storn [15] proposed an adaptive mechanism that relaxes the constraints of the problem in order to make all the initial solutions feasible. This pseudofeasible region is shrunk each generation until it matches the real feasible region.…”

“…Both, the aging parameter and N T are defined by the user. Storn [15] used a modified "DE/rand/1/bin" version. The approach showed a good performance in problems with only inequality constraints but presented problems when dealing with equality constraints.…”