Particle Swarm Optimization Algorithm And Its Applications To Power Systems Economic Operation

Particle Swarm Optimization (PSO) is a kind of heuristic algorithm based on swarm intelligence. It is a general framework algorithm, which has fast convergence performance and is easy to implement. PSO can be conveniently applied to solve complicated nonlinear, noncontinuous, multi-dimensioned optimization problem with discrete variables and multiple restrictions. There is a large amount of such problem in modern electric power systems. Generally speaking, traditional optimization method can hardly deal with them. In past decades, some global optimization techniques such as Genetic Algorithm and Simulated Anneal Algorithm have been applied to solve optimization problem in power systems successfully. This paper specially focuses on PSO and its applications to power systems economic operation. A modified PSO is presented in this paper, in which a third extreme is inserted to the iteration formula to guide the search direction of individuals and a "fly back"strategy is introduced. Simulation results on several famous benchmark functions demonstrate that the method can obtain good solutions with higher quality in shorter calculation time, that is, the search efficiency of PSO is improved. The modified PSO has been used to solve reactive power optimization (RPO) problem in power systems. Implement process on how to apply PSO algorithm for RPO control is given in detail and experiment results on standard IEEE-14 bus & IEEE-30 bus systems are exhibit in this paper. Compared to GA and conventional PSO, modified PSO has advantages in solving RPO problem to some extent. At the end of this paper, an evaluation-selected multi-objective PSO (MOPSO) is presented to solve multi-objective optimization problems. Tests on two multi-objective functions have been made and the results show that the MOPSO approach can obtain satisfactory distributed Pareto Front, which reveal the validity of the method. A rough study is carried out on multi-objective RPO problem that minimize total power loss and total voltage offset taking into account simultaneously using the proposed MOPSO. Simulation results on standard IEEE-30 power system demonstrate that the MOPSO method can balance the two objectives and more suitable result has been obtained. Along with the deeper research on PSO and its theoretics, PSO will play more important role in wider fields.