Dissertation > Excellent graduate degree dissertation topics show

Research and Application of flora optimization algorithm

Author: CheWeiWei
Tutor: XiangZuoZuo
School: Nanjing University of Technology and Engineering
Course: Control Theory and Control Engineering
Keywords: Flora optimization algorithm Chaotic search Inertia weight Fuzzy preference Dynamic weighting Economic Dispatch Node Localization
CLC: TP301.6
Type: Master's thesis
Year: 2012
Downloads: 64
Quote: 0
Read: Download Dissertation

Abstract

Flora optimization algorithm is based on swarm intelligence of biological heuristic optimization algorithm, as the new members of the family of evolutionary algorithms to get the attention of many researchers at home and abroad, because of its easy to understand the natural mechanism and potential applications. Focused flora to optimize the algorithm steps improved algorithm, the major work done by the paper is as follows: First, for standard the flora optimization algorithm is easy to fall into local convergence and easy to fall into the local optimal solution near shocks state, at the same time in the optimization process, the lack of population diversity, the introduction of adaptive tendency step algorithm to improve search performance to introduce chaos search mechanism to improve the diversity of the population of the algorithm; Benchmark function and then use the proposed new algorithm respectively, from the dimension of the optimization function (the function dimension were 2,5,10,15,20) and the number of iterations of the algorithm (the number of iterations were 100,150,200,250) two aspects for examination, and the same standard bacteria swarm optimization algorithm, based on the inertia weight particle swarm optimization, genetic algorithms and binary comparison. Optimization results can be seen, the improved algorithm optimization performance both from the convergence precision and convergence rate are significantly better than the other three algorithms optimized results significantly close to the optimal value, the simulation results show that the improvements bacteria presented in this chapter swarm optimization algorithm is feasible and effective. Second, the optimal solution set for multi-objective optimization problem, the lack of the pros and cons of evaluation criteria, weighted sum method to multi-objective optimization problem into the single objective optimization problem, but the determination of the weight coefficient is no uniform standard, therefore proposed kinds of interactive multi-target flora introduced fuzzy preference information optimization algorithm, the use of the adaptive tendency operation and distance-based sorting method calculated non-dominated solutions set and classic multi-objective optimization algorithm NSGA-Ⅱ and NSPSO optimization solution set performance is verified by comparing the chapter presents the new algorithm is feasible and effective. Finally, improvement flora optimization algorithm engineering: First, the optimization algorithm is applied to the economic dispatch of power system dynamic weighted fuzzy preference-based multi-target flora, considering the fuel consumption, pollutant emissions and network transmission loss, simulation results show that the use of the Pareto set obtained by this algorithm is more evenly distributed, and the non-dominant solutions in diversity better. Secondly, based on the traditional DV-Hop localization algorithm positioning accuracy is not high, the error of the larger problem, the unknown node to the beacon node distance as constraints to improve its positioning accuracy, and propose an optimization algorithm based on improved flora node localization method, the positioning process is the third step in the DV-Hop algorithm use its own coordinate position improved based on chaos search the inertia weight flora optimization algorithm to achieve simulation results show that the optimization algorithm based on improved flora DV-Hop node positioning method is feasible and effective.

Related Dissertations

CLC: > Industrial Technology > Automation technology,computer technology > Computing technology,computer technology > General issues > Theories, methods > Algorithm Theory
© 2012 www.DissertationTopic.Net Mobile