In the past two decades, a wide variety of optimization
techniques have been applied in solving the optimal power flow
problem, such as linear programming [1-2], non-linear
programming [3-4], Newton-based techniques [5-6], and
interior point methods [7-8]. The optimal power flow problem is
a highly non-linear and a multi-modal optimization problem
which exists more than one local optimum. Hence, most of these
techniques are not suitable for such a problem. Recently, genetic
algorithm and particle swarm optimization (PSO) have been
proposed for solving the optimal power flow problem [9-11].
Particle swarm optimization, first introduced by Eberhart and
Kennedy [12], has been used extensively in many fields,
including function optimization, neutral network and system
control, etc. Unfortunately, the premature convergence of
particle swarm optimization algorithm degrades its
performance.