Study On Uncertain Programming For Reactive Power Optimization Considering Static Voltage Stability

Voltage stable-lost breakdown easily occurs in modern power system approaching its limited operation state. Therefore, taking account of voltage stability in optimal reactive power planning is of great necessity. Moreover, in consideration of the research of VAR planning affected by its own development, environment changes, social economy development, and so on, the uncertainty problem has become outstanding in filed of programming. According to stochastic dependant-chance programming theory of uncertain programming, some research works for reactive power operation planning based on static voltage stability were done in this paper.At first, the background of knowledge relevant to this topic was comprehensively analyzed, including the meaning of voltage stability research and its development process, current status of VAR planning, the studies on VAR planning based on static voltage stability, uncertain programming and its applications in power system, ad so on. Then, in such background above, taking the random perturbation of nodal power injections as the major uncertain factors in power system operation, a stochastic dependant-chance programming model for reactive power operation planning was set up with maximal probabilities of voltage stability on Critical generator buses as the objectives and minimal active power loss as the secondary objective. What's more, this idea of model-building is of real significance for the secure and economic operating of power systems.Aiming at different models of reactive power optimization, diverse algorithms, them may be divided into two types (the methods of mathematical optimization and the heuristic searching technique), were developed. In this paper, particle swarm optimization (PSO) was employed to deal with the problem of VAR planning. Similar to general evolutional computation, such as genetic algorithm, ant colony optimization algorithm and so on, PSO also is an optimization algorithm based on the "flocking behavior", which, because of its simple principle, can be easily implemented in most programming languages and has proven both very effective and quick when applied to a diverse set of optimization problems. The simulation results show the convergence of the algorithm proposed is good.