Array Antenna Synthesis Based On Improved Intelligent Optimization Algorithm

With the increasing application of machine learning in various fields,the comprehensive research of arrays using different intelligent optimization algorithms,especially beamforming,has become an important research topic.In this kind of subject,the beamforming problem is generally transformed into an optimization problem.Through optimizing the arrangement of antenna array,the feed amplitude and phase of array elements,the intelligent optimization algorithms are used to achieve beamforming.Based on genetic algorithms,two types of beamforming problems,low sidelobe and null control,are studied deeply.Two improved genetic algorithms are proposed to address the issues of slow convergence speed,high computational complexity,and easy access to local optimal solutions in traditional genetic algorithms.The main work and innovation of this thesis are as follows:1.An improved cross genetic algorithm based on the standard genetic algorithm is proposed and applied to the design of circular array antennas.First,a circular antenna array is established,and then the array radius and element spacing are optimized.The array aperture,element number and element spacing are set to meet certain constraints,and the improved crossover genetic algorithm is used to achieve low sidelobe.The crossover method is improved by increasing the metric of the initial population and replacing individuals who do not meet the set convergence requirements.According to the comparison results of individual correlation and the value that you set,different crossovers are carried out to reduce the repetition rate.The simulation results show that this algorithm compared with traditional genetic algorithms achieves lower sidelobe levels and faster convergence speed.2.A chaotic genetic algorithm combining chaotic optimization and genetic algorithm is proposed,and this algorithm is used to optimize the layout of two-dimensional planar array antennas.By utilizing the characteristics of chaotic sequences to improve population initialization and adding chaotic perturbations to mutation operations,local optima are avoided.At the same time,compared to traditional genetic algorithms,the optimization range is wider and the calculation speed is accelerated.The simulation results show that the convergence speed of chaotic genetic algorithm is further improved compared to traditional genetic algorithm,and the sidelobe level of the array antenna is further reduced.Thus this method has certain feasibility.3.The amplitude and phase errors of phased array antennas are studied,and its impact on the beamforming of the array antennas are analysed too.In order to achieve antenna patterns with broad nulls from the perspectives of single objective optimization and multi-objective optimization,two improved genetic algorithms were employed to optimize the feed amplitude and phase of the array antenna elements.The simulation results of the two improved genetic algorithms were compared with the optimization results of the traditional genetic algorithms and the particle swarm optimization algorithms.The effectiveness of different intelligent algorithms in beamforming problems has been verified,and adding optimization objectives can achieve better optimization results.The improved genetic algorithms have better optimization performance than the traditional algorithms.