Zhengjun Li

Dynamic multi-objective differential evolution for solving constrained optimization problem

Dynamic constrained multi-objective differential evolution(DCMODE) is designed for solving constrained optimization problem(COP). Main feature presented in this paper is to construct dynamic multi-objective optimization problem(DMOP) from COP. The two evolved objectives are original function objective and violation objective. Constraints are controlled by dynamic environments, where the relaxed constraints boundaries are gradually tightened to original boundaries. After this dynamic process, DMOP solutions are close to COP solution. This new algorithm is tested on benchmark problems of special session at CEC2006 with 100% success rates of all problems. Compared with several state-of-the-art DE variants referred in this paper, our algorithm outperforms or performs similarly to them. The satisfactory results suggest that it is efficient and generic when handling inequality/equality constraints.

Dynamic constrained multi-objective model for solving constrained optimization problem

Constrained optimization problem (COP) is skillfully converted into dynamic constrained multi-objective optimization problem (DCMOP) in this paper. Then dynamic constrained multi-objective evolutionary algorithms (DCMOEAs) can be used to solve the COP problem by solving the DCMOP problem. Seemingly, a complex DCMOEA algorithm is used to solve a relatively simple COP problem. However, the DCMOEA algorithm can adopt Pareto domination to achieve a good tradeoff between fast converging and global searching, and therefore a DCMOEA algorithm can effectively solve a COP problem by solving the DCMOP problem. An instance of DCMOEA was used to to solve 13 widely used constraint benchmark problems, The experimental results suggest it outperforms or performs similarly to other state-of-the-art algorithms referred to in this paper. The efficient performance of the DCMOEA algorithm shows, to some extend, the DCMOP model works well.

Design of wide-beam antenna using dynamic multi-objective BBO/DE

Because the requirements of antennas get more and more complicated, an effective evolutionary algorithm for solving complex antenna problems is necessary. The hybrid biogeography-based optimisation with differential evolution BBO/DE and dynamic multi-objective technique are combined into dynamic multi-objective BBO/DE, which is aimed at improving the global search ability in solving antenna optimisation problem. We then used the effective algorithm to solve a challenging wide-beam antenna problem with beam width of 140 degrees. A small antenna with only seven wires, however, with 20 variables and 4,322 constraints, has been designed, and an acceptable one, which is satisfied in real application, has been obtained.

Research on simulation consistency between NEC and HFSS on strange antenna for evolving antenna

Evolutionary algorithm (EA) is a promising technology in automatic antenna design and it can invent new and strange antennas which are more effective than would otherwise be developed. However, a challenge in evolving antenna is that evaluating antenna fitness in the EA algorithm is usually time-consuming due to the time-consuming electromagnetic simulations. Numerical electromagnetics code (NEC) is a moment method solver with high computing speed and the code is publicly available for general use. It is then usually used to evaluate antenna fitness in evolving antenna. However, it does not go through a full reliability test yet, then it needs reliability test in simulating an antenna, especially a strange antenna before it is adopted to evolve antenna. This is the focus of this paper. And the reliability test is achieved by the simulation consistency between NEC and HFSS since HFSS is a commercial finite element method solver with high reliability. We find that NEC is reliable in simulating strange wire antennas, however, unreliable in simulating surface (patch) antennas.

Twisted Helical Antenna for Satellite-Mobile Handset Using Dynamic Multi-objective Self-adapting Differential Evolution Algorithm

This paper presents a twisted helical antenna, which operates at 2.4GHz and is applied in satellite mobile communications. It is designed using dynamic multi-objective self-adaptive differential evolution (SaDE) algorithm which combines SaDE and dynamically controlling constraint technique. The main reason for adopting dynamic multi-objective SaDE in this paper is that few literatures have used dynamic multi-objective differential evolution (DE) for solving antenna optimization problem so far. A small twisted helical antenna optimized on a perfect ground plane. Results show that while our optimized antenna met the design requirements, its performance is not inferior to the quadriflilar helical antenna as seen in reference. The design shows the capability of evolution algorithm (EA) as an efficient optimization tool for searching globally optimal solutions for antenna design.

An Effective Combination of Genetic Operators in Evolutionary Algorithm

An evolutionary algorithm (EA) is designed and then is used to solve constrained optimization problems in this paper. The difference of the proposed algorithm from other EAs stays in combination of two crossover operators: one is affine crossover which inherits characteristics of the parents by using function continuity, one is uniform crossover which preserves some discrete genes of the parents by using Darwins principle. Since both crossovers are independent to some extent, population diversity could be well maintained, then the new EA (denoted FUXEA) could enhance capacity in global search. The FUXEA algorithm is compared with some state-of-the-art algorithms which were published in a best journal in evolutionary computation area, and 13 widely used constraint benchmark problems to test the algorithm. The experimental results suggest it outperforms to or not worse than others, especially for the problems with many local optima, it performs much better.

Automated Antenna Design Using Self-adaptive Differential Evolution Algorithm

As we knew, an evolved X-band antenna for NASAs Space Technology 5 (ST5) spacecraft had been deployed on schedule in 2006. This paper uses Self-adaptive Differential Evolution Algorithm (SaDE) to solve ST5 antenna design problem. The performance of SaDE is reported on the set of 24 benchmark functions provided by CEC2006 special session on real parameter optimization. The experimental result shows SaDE can solve ST5 antenna design problem effectively. The simulation result indicates our evolved antennas voltage standing wave ratio performance is better than NASA evolved antenna and gain performance is competitive. Moreover, our evolved antenna was evolved on finite ground plane, the specified size for the ST5 antenna, while NASAs on infinite ground.

The study in fast calculation of Numerical Green Matrix of a surface antenna in a plane

In this paper, it is discovered that the correlation matrices of antennas whose scattering surface are within the same plane are diagonal matrices. By applying this conclusion to the Numerical Green Matrix Module in the NEC (Numerical Electromagnetic Code), the memory space used by simulation program is significantly reduced, and the computational speed is improved, when the initial structures are face structures within the same plane. In a simulated experiment of NASA ST5 antenna designing, the speed of computation with the method introduced in this paper is 25.3 times faster than that of original NEC, and uses as 1/858 memory as original NEC. Applying the approach to evolutionary antenna research can help with solving one of its great challenges: expensive and time-consuming computation.