Multi-objective Artificial Immune Algorithm And Its Applications In Reactive Power Optimization

With the development of national economy, the demands of power supply quality from all kinds of industries are increased. Secure and economical power system operation and continuous, high-quality supply of power have become urgent needs of the modern social economic. Reactive power optimization of power system is not only an effective means to keep power system operation securely and steadily, but also one of the most important methods to improve the voltage quality and reduce the transmission loss.Reactive power optimization in power systems is a typical non-linear programming problem with characteristics of multi-objective, multi-variable, multi-restriction discreteness and uncertainty. Conventional mathematical programming techniques are inadequate and insufficient to the optimal operation of power systems due to the inherent complexity. According to the characteristics of reactive power optimization, meanwhile, artificial immune algorithm has many merits, such as high searching efficiency, avoiding immature convergence, colony optimization, keeping individual diversity and so on. So a novel multi-objective optimization artificial immune algorithm is proposed to be used in multi-objective Reactive power optimization. In the proposed algorithm, it adopts an new non-dominant sorting method, an innovating sorting mechanism based on its Pareto ratio is used to sort individuals in antibody population. In addition, it improves selection and cloning scheme by using a fitness assessment based on crowding-distance density and introducing adaptive clone selection mechanism to the preservation of the anibodys' diversity, a new hybrid mutation operater using chaos random series for globally optimization solution has been presented to maintain anibody population diversity. Crowding-distance density operator is also adopted to distribute non-dominated individuals along the discovered Pareto front uniformly for muti-objective optimization. In addition, the effectiveness of the improved algorithm is validated upon the simulation testing of benchmark problems characterized by different difficulties in non-convexity, discontinuity, high-dimensionality and constraints. The comparative study shows the effectiveness of the improved algorithm, which produces solution sets that have highly superiority in terms of global convergency, diversity and distribution.Based on the traditional model of reactive power optimization, index of static voltage stability is introduced and a model of multi-objective reactive power optimization considering security and economy of power system is established, where the active power loss minimization, high voltage quality and static voltage stability margin maximization are taken as objectives,and improved multi-objective artificial immune optimization algorithm is chosen to solve multiobjective reactive power optimization problem for IEEE-30 bus system Then, technique for order preference by included angles is used to get the Pareto optimal solutions for multi-attribute decision making, we get a group of optimal, solution The simulated results under the parameters of the optimal solution show that the model is able to heighten power system voltage stability during the economical operation, and validity of the established model and effectiveness of the proposed algorithm are verified.