Soft Computing Techniques for Optimal Capacitor Placement - Complex System Modelling and Control Through Intelligent Soft Computations - page 606

flowchart of the algorithm is given in Fig. 9 .
Initially, a population of size 3(N + 1) is generated and evaluated. Particles are
ranked based on their

For an N-dimensional problem, the fl

fitness values. The top N elite values are saved and NM
method is applied to the top N + 1 particles. PSO operator is then applied to the
remaining 2 N particles with worst fitness. Then, the stopping criterion is checked

for all the 3(N + 1). If the criterion is achieved, then the process stops and best
solution is achieved, otherwise, the fitness values are again repeated and the

complete process repeats.

5 Solution Methodology

Based on this load flow and proposed objective function, the capacitors are placed

at respective buses using the algorithm presented in Fig. 10 . According to it, the
system parameters are entered in the program. Based on the system parameters and
sensitivity the candidate buses are selected. On the selected buses, the capacitors are
placed using the optimization algorithm and objective function evaluation. After the
capacitor placement, the constraints are checked.

fl

Fig. 10 Flowchart for
Nelder-Mead particle swarm
optimization (NM-PSO)
algorithm