Research On Differential Evolution Algorithm And Its Application In Electromagnetics

The Differential Evolution(DE)algorithm was proposed by two scholars called Storn and Price at Berkeley College in 1995.This algorithm was originally used to solve the Chebyshev polynomial problem.The algorithm has the characteristics of information sharing and preservation of individual optimal solution within the population.The principle of the algorithm is simple.The algorithm is easy to implement and has strong generality.It is also easy to use with other algorithms.It has been applied to many fields.However,this algorithm also has some shortcomings.When solving some complex optimization problems,the local search ability of the algorithm is not strong enough.In the limited period,this algorithm is difficult to get the global optimal value.The search efficiency of the algorithm is low,and it is easy to fall into the local optimal.The accuracy of the algorithm is also not high enough.Therefore,this paper proposes some improvement of DE algorithm,and the improved algorithm is applied to some complex optimization problems in the field of electromagnetic field.The main research work of this paper is as follows:(1)This paper studies the basic profile of the DE algorithm.This paper mainly introduces the basic principle of the DE algorithm,the mathematical description of the algorithm and the algorithm flow.(2)Due to the poor local search ability and low search efficiency of DE algorithm,an adaptive differential evolution algorithm based on reverse learning is proposed,and several classical test functions are applied to simulate different algorithms.Several classical test functions are used to carry out simulation experiments on different algorithms.The results show that the LMCODE algorithm proposed in this paper is better than the other algorithms.The improved differential evolution algorithm has faster convergence speed,which can effectively avoid the algorithm falling into local optimum and produce premature.Meanwhile,it enhances the global convergence ability of the algorithm and improves the search efficiency of the algorithm.(3)In order to make the convergence speed of the DE algorithm faster and the accuracy of the solution higher,the CCDE algorithm is proposed.Catfish effect and cloud model is introduced into the algorithm.Through the experimental simulation of the test function,the results show that the performance of CCDE algorithm is better.The local search ability of the algorithm is enhanced.Meanwhile,the convergence speed and the solving precision of the algorithm are also greatly improved.The algorithm is used to design the FIR bandpass digital filter,and the reliability and superiority of CCDE algorithm is verified by comparison.(4)The CCDE algorithm is applied to the pattern synthesis of linear array.Three aspects are considered from optimizing the continuous current amplitude,phase and the location of the array elements.By comparison,it is found that CCDE algorithm can achieve the index requirements and is superior to other algorithms.(5)The application of the electromagnetic optimization method based on the combination of CCDE algorithm and HFSS software is proposed.The electromagnetic optimization is carried out by combining MATLAB and HFSS software.E-shaped patch microstrip antenna and compaet low pass filter with regular hexagon shape DGS are optimized respectively.The results show that the CCDE algorithm can achieve the design index well,and it also proves that the combination of CCDE algorithm and HFSS can be applied to the electromagnetic optimization problem effectively.