Electromechanical Integrated Design Of Skin Antenna Based On Parallel Bayesian Optimization

Skin antenna is a new type of structural integrated antenna which integrates RF components,antenna arrays and sensing elements and its design is usually based on the simulation analysis of mechanical and electromagnetic software.This paper studies two problems existing in the design of skin antennas: First,the design of the skin antenna lacks a rapid electromechanical co-design optimization method.At present,the optimal design of skin antennas is to separate the electrical and mechanical properties of the antenna from each other.Designers use mechanical simulation software and electromagnetic simulation software to independently simulate the design,which often leads to more simulation software calls and time-consuming,and the skin antenna often fail to simultaneously meet the mechanical and electrical performance indicators;Second,the design of the skin antenna lacks a robust optimization design method that takes into account the influence of uncertain factors.There will inevitably be uncertain factors in the manufacturing process of skin antennas,such as manufacturing errors,material property deviations,etc.However,the current skin antenna design methods often ignore the influence of these uncertain factors,and the optimal design of the skin antenna is solved as a deterministic optimization problem so that the antenna electrical performance is sensitive to changes in uncertain factors.Small disturbances of these uncertainties may cause the electrical performance of the antenna to deteriorate or fail to meet the work requirements.In order to solve the above two problems,two parallel Bayesian optimization algorithms are proposed in this paper,and then applied to the optimal design of the skin antenna.The electromechanical co-design and robust optimization design of the skin antenna based on parallel Bayesian optimization are realized.The specific research work of this paper is as follows:(1)Two parallel Bayesian optimization algorithms are proposed: The first one is a parallel Bayesian optimization algorithm that integrates the pseudo-expectation improvement criterion,which improves the single-point iterative optimization of traditional Bayesian optimization to a multi-point parallel iteration.In this way,the optimization efficiency of the algorithm can be greatly improved.The second one is the pseudo-expectation parallel Bayesian optimization algorithm integrating the trust region,which introduces the trust region based on the first parallel Bayesian algorithm.The search space of the design variables is automatically updated,which can further improve the optimization efficiency of the algorithm.The theory and implementation process of the two parallel Bayesian optimization algorithms proposed in this paper are described in detail.Numerical examples and antenna element design examples are used to compare the proposed algorithm with the existing algorithms.The results show that the proposed algorithm has higher optimization efficiency and accuracy than existing algorithms.(2)The mathematical model of the electromechanical co-design optimization of the skin antenna is established and the design idea of electromechanical co-design optimization of skin antenna protection layer and antenna layer is proposed.The algorithm is applied to the electromechanical co-design optimization of two skin antennas,and compared with the existing algorithms.The results show that the proposed algorithms can not only obtain better optimization results,but also greatly improve the optimization efficiency of the skin antenna.(3)Considering the uncertainty of material properties and the manufacturing error,a robust optimization design model of skin antenna based on interval order relationship is established.Then the algorithm proposed in this paper is applied to the robust design of the skin antenna,and a robust optimization design algorithm for the skin antenna based on parallel Bayesian optimization is formed.Finally,the effectiveness of the proposed method is verified by the robust optimization design of the four-element skin antenna array.(4)Developed a set of antenna virtual prototype rapid optimization software.In this paper,Code::Blocks is used as the development environment and C++ is the development language.A set of antenna virtual prototype rapid optimization software based on parallel Bayesian optimization algorithm is developed so that the proposed algorithm in this paper can be applied to engineering designs better.