Applications Of Improved Particle Swarm Optimization In Antenna Design

As an important branch of meta-heuristic intelligent optimization algorithm,swarm intelligence algorithm is constantly growing with its unique adaptability.It can be used to solve the optimization problems which cannot be solved by traditional optimization algorithm,such as gradient descent method,etc.,which has been favored by scholars in various fields.Particle swarm optimization is an intelligent optimization algorithm proposed in 1995,which has the advantages of less parameters,simple logic,strong adaptability and high convergence accuracy.But in the face of complex problems,there will be slow convergence speed,easy to fall into local optimum,premature convergence and so on.In view of this,this paper has improved the classical particle swarm optimization algorithm's performance.Then the improved particle swarm optimization algorithm is applied to the array antenna synthesis and then optimize the WLAN antenna structure.The main research results of this thesis include:1.Introduce the origin of basic particle swarm optimization,its mathematical principle,research status,practical application and so on,pointed out the shortcomings of the algorithm,and provided the ideas to improve the algorithm.2.In order to further improve the convergence speed and optimization ability of particle swarm optimization,the main improvements in this paper are as follows:first,using multiple group initialization strategy to ensure the uniform distribution of initial particles in the solution space;then,the elite guidance strategy is introduced to guide the step direction of particle and improve the convergence speed.Finally,we use the hyper spherical Disturbance strategy to perturb the elite particles are trapped in the local optimum,guarantee the global optimization ability of the swarm.finally,the convergence of the whole iteration is balanced by improving the basic parameters.3.Using the standard test function,the improved particle swarm optimization algorithm in this paper is compared with the basic particle swarm algorithm and other literature improvement algorithms to verify the effectiveness of the improved strategy.The standard test function is translated and rotated to investigate the optimization ability of the improved algorithm on complex functions with unknown space.4.The improved particle swarm optimization(PSO)algorithm is applied to the pattern synthesis of array antennas to reduce the side-lobe level and achieve deep null depths in the specified direction.5.The improved algorithm is applied to optimize the structural parameters of WLAN antenna for improving the performance of the antenna.In view of the slow speed of HFSS simulation and the long time optimization of algorithm,the agent model of antenna is built on the basis of neural network,instead of time-consuming HFSS simulation,a lot of time is saved and good results are achieved.