Economic Dispatch of Power Systems Based on the Modified Particle Swarm Optimization Algorithm

This paper presents a new versatile optimization algorithm called modified particle swarm optimization algorithm (MPSO) for solving the economic dispatch (ED) problem of power systems. Compared with the classical PSO algorithm, several modified operators, such as mutation operator and neighborhood magnifying operator, are employed by MPSO to ensure the algorithm search globally in theory. Based on the stochastic analysis theorem, it is proved that MPSO has global convergence property, the convergence property is independent of the initialization, and a sufficient condition for convergence is presented. This algorithm is used to solve the complicated, nonconvex, nonlinear economic dispatch problems. Several factors such as, valve-point effects of fuel cost functions, transmission capacity constraints, and system stability constraints are considered in the computation models. Numerical results show that the proposed method is feasible and efficient.