Research Of Optimal Power Flow Problem Based On Artificial Immune Algorithm

Optimal Power Flow(OPF) can be defined as a typical nonlinear programming problem. The computational difficulties in solving the OPF problem have limited its use in power system operations. In the past four decades, various optimization techniques were proposed to solve the problem. But people fail to make the generally acknowledged satisfied achievement so far. In recent years, artificial immune algorithm, because of the unique theoretical basic and virtue, has provided new ways and means to solve the large-scale and non-linear complicated power system problem. First, this dissertation compared and discussed the method of all kinds of optimization based on summarizing the research in recent advances about the OPF algorithm both at home and abroad. It pointed out the potential research direction of OPF problem.Second, this dissertation used artificial immune algorithm to solve OPF problem. The method can convert OPF problem into unconstrained extreme value problem using intellectual optimizing characteristics. The artificial immune algorithm is used in the searching, which has the parallel processing characteristic is easy to realize, and can enhance robustness and effectively improve the global convergence performance and calculation accuracy. This method uses the target function information to guide the search of solution space and overcomes the difficulties in approximate treatment and assumption of classical optimization algorithm.At the same time, the basic principles and calculation steps of artificial immune algorithm is introduced. It also Analysis the reason that artificial immune algorithm can avoid falling into premature convergence. This dissertation combined artificial immune algorithm and the model of OPF and conducted a study on the Optimal Power Flow.Because that most of artificial immune algorithms about OPT are based entropy, this dissertation utilize artificial immune algorithms based on vector distance to calculate the OPF problem. The real number coding is used to code antibody in this dissertation, for the OPF is a large-scale multi-variable optimization problem. The vector distance is utilized to calculate the distance of two antibodies. Thereby, the algorithm maintains good global search capability.At last, the artificial immune algorithm is demonstrated on the standard IEEE30 bus system. The investigations has revealed that the artificial immune algorithm is capable of obtaining higher quality solutions efficiently in OPF problem, and the results show its high efficiency and promising practical applications.