Reactive Power Optimization Of Power Systems Based On An Improved Persudo-gradient Search Particle Swarm Optimization Algorithm

With the rapid development of national economy and with the development of power system and the complexity of system structure,the demands on the quality of power provided by power grid will get more pressing.With the introduction of electricity market,adopting some effective measures to decrease power loss,improve voltage and enhance the voltage stability of the system has become an important topic for electric power companies.Reactive power optimization of power systems a mixed nonlinear optimization with multivariable and multiple constraint.Its control variables include continuous variables and discrete variables,and the optimization process is very complex.With advantages such as easy implementation,few parameters,good adaptability,particle swarm optimization algorithm is widely used in reactive power optimization.Based on comprehensive analysis of domestic and foreign studies on particle swarm optimization algorithm and reactive power optimization,the application of particle swarm optimization algorithm on reactive power optimization has been thoroughly researched by this paper.The main contents are as follows:(1)After adjusting and modifying the formula of PSO for location updating,this paper proposes a persudo-gradient search particle swarm optimization based on opposite-based learning(obl PGPSO).This algorithm is used to determine the search direction for each particle in a population.It can accelerate the particle with good direction,and disturb the particle with bad direction though opposite-based learning,in order to enhance the diversity of algorithm.(2)This paper develop a mathematical model of reactive power optimization with with a target function of minimum power loss,and the rate of cluster focus distance changing and the location weight were introduced in this new strategy and the inertia weight was formulated as a function of these factors according to its impact on the search performance of the swarm.Based on researches above,this paper proposes a persudo-gradient search particle swarm optimization based on opposite-based learning with dynamically changing inertia weight(DCW-obl PGPSO)for solving the power system reactive power optimization problem,and verifies its efficiency and practicability through IEEE-30 system.(3)This paper develop a mathematical model of reactive power optimization with a target function of minimum power loss and voltage excursion,and maximum voltage stability.A persudo-gradient search particle swarm optimization based on opposite-based learning with pareto solution space division(PSSD-obl PGPSO)is proposed to solvemultiobjective optimization problem.The algorithm,though pareto solution space division,explore the convergence and diversity of pareto solution in external archive,based on which it brings out a new strategy of external archive maintenance and optimal particle selection,and enhance the ability of optimization by using obl PGPSO.At last,this paper verifies its efficiency and practicability through IEEE-30 system and a regional power grid.