Hi Bala,
Here’s an introductory MATLAB code template for Multi-Objective Genetic Algorithm (MOGA) applied to Optimal Siting and Sizing of Distributed Generation (DG) in Radial Distribution Systems (RDS).
Following is an example code for the same:
% MOGA for Optimal Siting and Sizing of DG in RDS
% ----------------------------------------------------------
% Requirements: Global Optimization Toolbox (for gamultiobj)
clc; clear; close all;
%% --- Parameters ---
nBuses = 33; % Example: IEEE 33-bus system
DG_min = 0.1; % Minimum DG size (pu or MW)
DG_max = 1.0; % Maximum DG size (pu or MW)
nDG = 1; % Number of DG units to place
% Decision variables: [DG1_location, DG1_size, DG2_location, DG2_size, ...]
nVars = nDG * 2; % Each DG: [location, size]
lb = [ones(1,nDG), DG_min*ones(1,nDG)]; % Lower bounds
ub = [nBuses*ones(1,nDG), DG_max*ones(1,nDG)]; % Upper bounds
IntCon = 1:2:nVars; % Location variables are integers
%% --- MOGA Options ---
options = optimoptions('gamultiobj', ...
'PopulationSize', 50, ...
'MaxGenerations', 50, ...
'Display', 'iter', ...
'PlotFcn', {@gaplotpareto});
%% --- Run MOGA ---
[obj, fval, exitflag, output, population, scores] = ...
gamultiobj(@(x)objectives(x, nBuses), nVars, [], [], [], [], lb, ub, [], IntCon, options);
%% --- Show Results ---
disp('Pareto optimal solutions:');
disp(fval);
%% --- Objective Function ---
function f = objectives(x, nBuses)
% x: [DG1_location, DG1_size, DG2_location, DG2_size, ...]
% 1. Round locations to nearest bus
locs = round(x(1:2:end));
sizes = x(2:2:end);
% 2. Power flow analysis with DGs at locs with sizes
% [You must implement your own power flow and loss calculation]
[Ploss, Vdev] = evaluate_RDS(locs, sizes, nBuses);
% 3. Objectives: [minimize Ploss, minimize voltage deviation]
f = [Ploss, Vdev];
end
%% --- Power Flow and Loss Calculation (Stub) ---
function [Ploss, Vdev] = evaluate_RDS(locs, sizes, nBuses)
% This is a placeholder!
% Replace this with your own load flow code for the RDS,
% and compute total real power loss (Ploss)
% and total voltage deviation (Vdev)
%
% For demonstration, we use random values.
Ploss = rand() + sum(sizes)*0.01; % Replace with actual loss
Vdev = rand() + abs(mean(locs)-nBuses/2)*0.01; % Replace with actual voltage deviation
end
