Font Size: a A A

The Improvement And Application Of Shuffled Frog Leaping Algorithm

Posted on:2012-07-23Degree:MasterType:Thesis
Country:ChinaCandidate:Y Q DaiFull Text:PDF
GTID:2218330362950004Subject:Agricultural Electrification and Automation
Abstract/Summary:PDF Full Text Request
The Shuffed Frog Leaping Algorithm(SFLA) is a biological evolutionary algorithm based on swarm intelligence, was proposed by Eusuff and Lansey in 2003. The SFLA has simple concept, few parameters, quick computation speed,a strong global optimization ability, and the algorithm is easy to realize. At present,this algorithm was already improved in many other applications, including water resources network distribution, function optimization and oil pipe network optimization, combinatorial optimization, image processing, multi-user detection, index weight determination and the power system optimization scheduling, and achieved good effect.This paper focuses on the principle, theory and applications of SFLA, the convergence property and the deficiency of SFLA were analysed, and the significance of research was discussed, then several improved algorithm were proposed, which were used to solve the problems such as the optimization of projection pursuit model and economic dispatch of the power system.1. The Shuffled Frog Leaping Algorithm with Memory Function(MSFLA)is proposed. Through introducing the adaptive learning operator, make the algorithm convergence with faster speed and expand search area in early iterations, and search precisely in the global optimal neighborhood in later iterations, which kept the balance between exploration and development and improved the convergence precision. By using randomized grouping strategy, the optimization ability of memeplexes were balanced, and the population diversity was kept. Through testing six benchmark functions, and comparing with basic SFLA and the improved SFLA in related references, the results showed that MSFLA had better performance.2. The Shuffled Frog Leaping Algorithm with a Contraction Factor(SFLACF) is proposed. By importing acceleration factor, the algorithm improved the capacity of the worst individual learn from the best individual of memeplexes or the entire population,and speeded up the convergence rate. The contraction factor ensured the convergence of the algorithm. The new algorithm improved the self-learning ability of the individuals and optimization precision by full using of the useful information of the worst individual and executing small-scale self-learning algorithm.The results shows that the improved algorithm has better optimization performence than basic SFLA by testing the six standard functions.3. The Hybrid Algorithm of Particle Swarm Optimization and Shuffled Frog Leaping Algorithm(PSO-SFLA) is proposed by combining the rapid convergence properties of PSO algorithm and the outstanding global cooperative search ability of SFLA. Hybrid algorithm divides the swarm into two sub-groups. In each iteration, one sub-group evolves using PSO algorithm, the other one evolves using SFLA, and two algorithms shared the information of the two sub-groups extremum. Through comparing PSO-SFLA hybrid algorithm with basic PSO algorithm and the improved SFLA of related literature in volving solution to three standard functions, results showed that PSO-SFLA hybrid algorithm outperforms PSO algorithm.4. Improved SFLA was used to optimize projection pursuit model, then using the model to evaluate the national economic comprehensive index and construction of ecological agriculture comprehensive index, and overall assessment ranking of samples and the influence weight of each index on overall evaluation results were obtained.5. Improved SFLA was used to optimize unit power allocation of economic dispatch in the power system, considering single target or multiple target cases, the optimal unit configuration parameters or optimal unit configuration non-inferior parameters set were obtained.
Keywords/Search Tags:SFLA, memory function, contraction factor, projection pursuit, power system, economic dispatch
PDF Full Text Request
Related items