We found a match
Your institution may have rights to this item. Sign in to continue.
- Title
Research on high‐efficiency optimization algorithm applied to near‐field effect error correction.
- Authors
Zhang, Jia; Yu, Mengxia; He, Ke
- Abstract
In order to achieve a more efficient and accurate correction of near‐field error in the semiphysical radio frequency simulation system, the precise control parameters of the three antenna elements need to be obtained. This article is based on the method of moments electromagnetic simulation, and propose corresponding improvement ideas for the problems of limited optimization accuracy and low calculation efficiency in the near‐field error correction process. From the aspects of high‐precision intelligent inversion algorithm and high‐efficiency electromagnetic forward modeling, systematic optimization design and verification were carried out. The results prove that the control parameter filtering scheme based on PSO‐GA hybrid method has better optimization efficiency and accuracy than single genetic algorithm or differential evolution algorithm, which can provide more ideal initial amplitude and phase parameters for the subsequent selection of electromagnetic simulation and forward verification. In order to solve the problem of time‐consuming in the electromagnetic simulation, the multivariate vector forward model based on GA‐BP network and PSO‐SVM network are established, which can achieve high‐precision positioning of synthetic vector target points. The neural network method has been proved to be feasible on the basis of the current sample size. The paper selects hybrid algorithms to improve the shortcomings of single algorithm and uses algorithms to optimize neural networks, thereby obtaining better optimization results and reducing the time‐consuming of electromagnetic simulations, which can realize efficient correction of near‐field error.
- Subjects
ERROR correction (Information theory); MATHEMATICAL optimization; DIFFERENTIAL evolution; COMPUTATIONAL electromagnetics; RADIO frequency; GENETIC algorithms
- Publication
International Journal of RF & Microwave Computer-Aided Engineering, 2022, Vol 32, Issue 12, p1
- ISSN
1096-4290
- Publication type
Article
- DOI
10.1002/mmce.23530