A Study Of Dimension-division Based Team Progress Algorithm (DTPA) And Its Applications In The Optimization And Design Of Microstrip Antennas

Optimization issues have long been the difficulty of scientific development,which is related to industrial engineering and other fields.Intelligent optimization algorithm has been widely used by the virtue of its simple,stable,high efficiency.In this paper,the principles and steps of Team Progress Algorithm is stated.Besides,Dimension-Division based Team Progress Algorithm(DTPA)is proposed and applied to the optimization of microstrip patch antenna.The main work is as follows:Firstly,the principle,advantages and disadvantages of several common intelligent algorithms are briefly introduced.To solve the shortcomings of their easy to premature,slow convergence and complex calculation,TPA uses dual-group mechanism,the members study or explore according to the appropriate model.Then,according to the reasonable update mechanism,the population can quickly approximate request.In order to verify the efficiency of TPA,ten functions with different characteristics are used to test TPA and differential evolution algorithm(DE).The results highlight the advantages of TPA: simple parameter setting,high efficiency,good global and local optimization,and so on.Secondly,the dimension-based team progress algorithm(DTPA)is proposed on the basic of all the existing TPA letters.The engineering problems is divided into high-dimension and low-dimension according to their own difficulty.For low dimension,Gaussian distribution is used in both learning and exploration.For high-dimensional,Beta distribution is applied in learning,and Gaussian distribution is used in exploration.In addition,the coefficient of Gaussian distribution is discussed,and the better one is achieved.The procedures of DTPA are firstly presented,together with ten test functions tested.Then the performances,including accuracy and efficiency,of DTPA are compared with the available best TPA and DPDE.The results show that DTPA has improved global success rate and computational efficiency.At last,DTPA is combine with the external optimizer of HFSS to optimize two microstrip antenna in the existing literature.For the gap-coupled microstrip antenna,the patch is fragmented by DTPA to achieve 400% increasement in working bandwidth.And for the microstrip patch antenna,DTPA conducts the research of the ground plate for all metal,fragmented metal and miniaturization plate,to achieve 200%,107.6% and 46.2% increasement in working bandwidth,respectively.Through these two optimization examples,the feasibility and efficiency of DTPA combining with the external optimizer of HFSS are proved.