Coupled Structural-electromagnetic Modeling And Robust Design Of Active Phased Array Antennas Considering Parameter Uncertainty

Active phased array antenna is an important electronic equipment in modern national defense and commercial communication.It involves various advanced information operation platforms,general command and control system,5G communication base station and intelligent Internet of things.At present,it is developing rapidly in the direction of high frequency band,high anti-jamming ability,high reliability and lightweight.At the same time,the mechanical and electromagnetical coupling relationship between the structure parameters,excitation parameters and electromagnetical performances of active phased array antenna is becoming closer,and the electromagnetic performance deterioration of active phased array antenna is getting more and more serious due to the uncertain parameters caused by device performance deviation,manufacturing assembly accuracy and timevarying service load..How to ensure the high performance of active phased array antenna in the design phase,avoid unclear target identification,shortened operating distance and deterioration of communication quality,so as to improve the security of national defense and commercial communication stability,has become a key technology to be solved.However,due to the complex high-dimensional nonlinear coupling relationship between the electromagnetical performance of the active phased array antenna and the structure and excitation parameters,the structure-electromagnetic coupling model describing the relationship of the interval uncertain parameters and the electromagnetical performance of the active phased array antenna,the corresponding influencing mechanism and optimization design method have been difficult to solve accurately and efficiently for a long time.Therefore,from the perspective of structure-electromagnetic coupling,the structuralelectromagnetic modeling theory and robust design method are studied in this paper.The main contents are as follows:1.The interval errors of element position,excitation amplitude and phase of active phased array antenna will seriously deteriorate the electromagnetical performance.Limited by the overestimation effect of the interval algorithm,the traditional interval algorithm overestimates the influence of interval error on the electromagnetical performance of active phased array antenna.In this paper,based on the first-order Taylor expansion of integral form,each interval variable is separated from the interval function of power pattern of the active phased array antenna.Finally,a concise integral Taylor surrogate model between the electromagnetical performance interval of active phased array antenna and the array element position/excitation phase interval error is established,and the method to evaluate the power bounds of active phased array antenna based on integral Taylor surrogate model has been proposed;2.The Minkowski sum algorithm and the improved dimension-wise method are the mature means to determine the electromagnetical performance range of active phased array antenna when there are interval errors of antenna position,excitation amplitude and phase at the same time.However,the calculation efficiency will decrease rapidly with the expansion of the array size,which brings huge challenges to the efficient analysis and robust design of active phased array antennas.To this end,this paper derives the prior knowledge of the distribution of each complex interval variable when the modulus of the complex interval variable sum is at the upper and lower bounds,and based on the prior knowledge,and establishes the Priori knowledge algorithm to determine the electromagnetical performance bounds of array antenna with interval position,excitation amplitude and phase errors;3.The robust design of the active phased array antenna is a typical constrained optimization problem of high-dimensional complex functions.Its computational efficiency and computational cost are the bottlenecks for achieving efficient and robust design.Therefore,the analysis of influence mechanism and selection of key parameters before robust design become important steps to improve efficiency.To this end,this article first fixes the weighting condition and beam direction of the active phased array antenna,analyzes the effect of interval nature and index of the interval parameter error on the active phased array antenna’s electromagnetical performance parameters which including the maximum power interval and the maximum sidelobe level,the 3d B power beamwidth interval;Then,through the fixed interval index,the variation law of the electromagnetical performance of the active phased array antenna is studied when the interval parameter error and the design parameter of the active phased array antenna are changed at the same time;Finally,combined with the Design of Experiment method,the sensitivity of the electromagnetical performance of the active phased array antenna to the interval parameter error is analyzed when the interval parameter error,weighting condition and beam direction change simultaneously;4.The uniform distribution tolerance scheme of active phased array antenna is a robust design method commonly used in engineering.At the same time,the efficient global optimization algorithm based on the surrogate model has been widely used in the field of robust design due to its compromise search ability for the global and local optimal points in the design space.However,in the entire optimization process,the algorithm has a low speed to improve the accuracy of the complex function surrogate model,and it cannot automatically adjust the global and local optimal search weights,making the algorithm more likely to fall into local search and stagnant convergence.For this reason,this paper proposes an improved Efficient Global Optimization algorithm,which adaptively adjusts the search weights of the global and local optimal points in the algorithm,and adds sample points with large errors between the predicted value of the surrogate model and the true value in the optimization process.While rapidly improving the accuracy of the surrogate model of complex function,it can make the algorithm jump out of local search and stagnation in time,and finally obtain more accurate optimization results.Based on the proposed algorithm,the robust design of the electromagnetical performance of the active phased array antenna under the uniform distribution tolerance scheme is realized;5.In order to further improve the robust design effect of the active phased array antenna,the non-uniform distribution tolerance scheme is one of the effective ways,and how to reduce the more computing resources and longer computing time required to implement this approach is the key difficulty.To this end,this paper proposes a hybrid genetic algorithm based on global sensitivity.First,the Elementary Effect method is used to calculate the global sensitivity of the active phased array antenna’s electromagnetical performance interval to the the width of interval position,the excitation amplitude,and the phase error of the array antenna.The global sensitivity is integrated into the entire optimization process of the Genetic Algorithm to improve its ability to search for the local optimum.Finally,using the proposed method,the robust electromagnetical performance design of the active phased array antenna under the non-uniform distribution tolerance scheme is realized;6.The uncertainties in the structure and excitation parameters of active phased array antennas in the manufacturing and service processes lead to complex and time-consuming design processes,requiring multiple departments to perform reciprocating designs.At the same time,similar processes need to be repeated again when changing product models,which greatly reduces the development speed and iteration efficiency of active phased array antenna products.To this end,this paper builds an active phased array antenna electromagnetical coupling flexible design platform considering parameter uncertainty.It solidifies the main design process of the active phased array antenna into the platform.At the same time,the automatic identification function of parameterized CAD model and CAE model parameters is designed to realize the versatility of platform product models;and based on the “template pattern” design mode,the plug-in module of the software platform can be dynamically loaded and modified,which improves the versatility of the platform analysis process.In addition,the analysis and robust design methods that consider parameter uncertainty proposed in this paper are integrated into the platform.Finally,the platform is used to realize the robust design of the search radar equipped with the missile-artillery integrated tank.