Research Of Power System Reactive Power Optimization Based On Immune Ant Colony Algorithm

With the rapid development of power industry, how to ensure power system security, stability, and economic operation has become an important issue. The proper distribution of reactive power in grid has an important impact on improving power quality and energy saving, so, the study on reactive power optimization has very important practical significance. This article considering the status of reactive power optimization studying, after carefully studying the principle of ant colony algorithm, did some researching on how we can use immune mechanism in the ant colony algorithm for reactive power optimization.Reactive power optimization problem is a multi-variable, multi-constrained mixed nonlinear planning issues, the optimization processing is very complicated. This paper introduces the field of reactive power optimization and status of today'study to ensure using the level of voltage quality and minimum active power losses as the objective function to establish the mathematical model, and let load node voltage constraints and generator reactive power constraints as the penalty function. The Ant colony algorithm has the ability of distributed parallel global search elements, but its lack of initial pheromone, often trapped in local optimal solution and low efficiency; The immune algorithm has a good system autonomy and speediness, the system has strong self-balance ability when met with strong interference, but can not take full advantage of the feedback information, often do a lot of inaction redundant iteration, solution efficiency is low. According to the basic principle and the advantages also disadvantages of the two algorithms, use immune mechanism in the ant colony algorithm to form immune ant colony ant algorithm, use immune algorithm to optimize the initial pheromone of ant colony algorithm, use IEEE 30 bus system example for analysis, and make a Comparison with Genetic algorithm and the basic ant colony algorithm. The results show that this algorithm some kind of overcoming the disadvantages about the initial lack of pheromone, easy early shortcomings in basic ant colony algorithm, and its convergence speed and accuracy are greatly improved.