Research On Improved Swarm Intelligence Algorithm And Its Application In Practical Problems

With the rapid development of science and technology,more and more people need to find an efficient technology to deal with complex optimization problems.Because of its outstanding advantages in complex optimization problems,swarm intelligence algorithm has attracted the attention of many scholars.In this paper,two improved swarm intelligence algorithms are proposed,namely grey Wolf enhanced comprehensive learning particle swarm optimization algorithm and chaos enhanced bat algorithm.The improved algorithm is applied to practical optimization problems.The main research contents of this paper are as follows:(1)In this paper,a chaotic enhanced bat algorithm is proposed.The bat algorithm is an optimization algorithm proposed by researchers using the predation principle of bat echolocation in nature.It has been applied in many fields,such as engineering design,feature selection and machine learning.However,when solving complex optimization problems,the bat algorithm is prone to fall into a local optimum.In this paper,a chaotic mapping model is added to bat algorithm.The number of chaotic maps is controlled by a threshold and the search speed is controlled by velocity inertia weight.These mechanisms improve the convergence accuracy and speed of bat algorithm.The bat algorithm is compared with 10 original algorithms and 8 latest improved algorithms on 30 benchmark functions to verify the effectiveness of the proposed algorithm.Experimental results show that the proposed algorithm is superior not only to the original algorithm but also to the latest improved algorithm.Chaos enhanced bat algorithm was used to optimize parameters C and ? parameters of the kernel limit learning machine.A hybrid model of nuclear limit learning machine(CEBA-KELM)based on chaos enhanced bat algorithm is proposed.CEBA-KELM model is better than other models in diagnosing Parkinson's data set,which proves that chaos enhanced bat algorithm has better effect in medical diagnosis.Chaos-enhanced bat algorithm has been successfully applied to many constrained engineering problems.The results show that the algorithm can effectively deal with constraint problems.(2)This paper presents a grey Wolf enhanced comprehensive learning particle swarm optimization(CLPSO)algorithm.Comprehensive Learning Particle Swarm Optimization(CLPSO)is an improved particle swarm optimization(PSO)using a comprehensive learning strategy(CLS),which effectively improves the performance of the particle swarm optimization algorithm.However,when dealing with single mode function,the convergence speed of the algorithm is slow,so it cannot converge to the optimal value quickly in the process of function optimization.This article combines another mature algorithm,Gray Wolf Optimization Algorithm(GWO),with the comprehensive learning particle swarm optimization algorithm to enhance the local search capability of the comprehensive learning particle swarm optimization algorithm,and compared experimentally with 7 classic original algorithms and 8 advanced improved algorithms on 30 benchmark functions.Gray Wolf enhanced comprehensive learning particle swarm algorithm is superior to other comparison algorithms on four different types of functions,Effectiveness of the proposed algorithm.The improved algorithm is applied to the fertilizer effect function.The model optimized by the swarm intelligence algorithm is superior to the model optimized by the traditional method.At the same time,the improved algorithm was successfully applied to the three constraint engineering design problems of welded beams,I-beams and pressure vessels.Experimental results show that the algorithm can effectively deal with constraints and solve problems encountered in actual production.