Improvement And Application Of Social Group Algorithms

In recent years,intelligent optimization algorithm has gradually played an important role in many fields with its unique and efficient operation mechanism,such as health care,engineering and so on.Social group optimization algorithm is a new optimization algorithm based on social group learning.This algorithm is easy to operate and implement with fewer parameters.At present,the theoretical research of SGO algorithm is still in its infancy.Compared with some mature optimization algorithms,its theory is not perfect enough,and its engineering application needs to be further expanded.Based on the primitive social group optimization algorithm,this thesis proposes a method to improve the overall performance of the algorithm and the improved algorithm is extended to related application fields.The main research contents of this thesis are as follows.(1)A multi-group social group learning algorithm(MPSGO)is proposed in this thesis to solve the problem that the social group optimization algorithm was easy convergent to local optima.This algorithm adopt the multi-subgroup learning method,and make improvements of the individual learning method in the two stages of the algorithm.The diversity of the population was improved with maintaining the convergence of the population.At the same time,it introduced quantum-behaved learning method into MPSGO for parts of individuals to enhance the useful information of individual learning.In addition,the population was randomly regrouped to generate new subgroups after a certain generation.The diversity of subgroups was maintained and the individuals in every subgroup were fully evolved.Based on the designing of algorithm,this thesis analyzed the algorithm of convergence and diversity.Compared with the other 4 algorithms,the effectiveness of the improved algorithm is verified.(2)Design Discrete SGO Algorithms.SGO algorithm was originally used to optimize continuous domain functions,but in practical engineering applications,there are combinatorial optimization problems belong to discrete domain,so the SGO algorithm needs to be discretized.The continuous SGO algorithm is discretized in thisthesis,and some discrete operation rules are designed according to the task requirements,which are successfully applied to TSP problem solving.In the improvement and acquisition stage of SGO algorithm,crossover and mutation operations are introduced respectively,which increases the diversity of population and reduces the probability of algorithm falling into local optimum.The experimental results are carried out under the standard traveling salesman problem test data,the results show that the social group optimization algorithm has good results in solving the traveling salesman problem.(3)Expanding the application field of SGO algorithm.In order to improve the convergence speed of the population on the basis of the original social group optimization algorithm,some individuals are guided to learn from the excellent individuals in the historical population.The improved algorithm is combined with the CECA(Coverage and Energy Aware Clustering Algorithm)protocol to enable the cluster heads in WSN to transmit data to the base station in a multi-hop way and prolong the network lifetime.Compared with other algorithms,the experimental results show that the improved algorithm performs well in WSN problem.In summary,this thesis conducts a comprehensive and in-depth study and analysis of the SGO algorithm,not only two effective improvement methods are proposed,but also the application field of the algorithm is expanded.It provides reference and reference for the future research of SGO algorithm.