CN103353904A - Data-driven design method integrating active interlayer microstrip antenna structure and electromagnetism and antenna - Google Patents

Data-driven design method integrating active interlayer microstrip antenna structure and electromagnetism and antenna Download PDF

Info

Publication number
CN103353904A
CN103353904A CN2013101273196A CN201310127319A CN103353904A CN 103353904 A CN103353904 A CN 103353904A CN 2013101273196 A CN2013101273196 A CN 2013101273196A CN 201310127319 A CN201310127319 A CN 201310127319A CN 103353904 A CN103353904 A CN 103353904A
Authority
CN
China
Prior art keywords
radiating
data
interlayer
microstrip antenna
antenna
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN2013101273196A
Other languages
Chinese (zh)
Other versions
CN103353904B (en
Inventor
周金柱
黄进
段宝岩
宋立伟
王从思
李鹏
章丹
郭东来
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Xidian Univ
Original Assignee
Xidian Univ
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Xidian Univ filed Critical Xidian Univ
Priority to CN201310127319.6A priority Critical patent/CN103353904B/en
Publication of CN103353904A publication Critical patent/CN103353904A/en
Application granted granted Critical
Publication of CN103353904B publication Critical patent/CN103353904B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • YGENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02EREDUCTION OF GREENHOUSE GAS [GHG] EMISSIONS, RELATED TO ENERGY GENERATION, TRANSMISSION OR DISTRIBUTION
    • Y02E60/00Enabling technologies; Technologies with a potential or indirect contribution to GHG emissions mitigation

Abstract

The invention discloses a data-driven design method integrating an active interlayer microstrip antenna structure and electromagnetism and an antenna, and solves the problem that the prior art can not realize the integrated design of the structure and electromagnetism. The method comprises the following steps: firstly, fixing the material, size, panel and cellular size of a radio frequency function layer according to electrical property indexes, secondly, obtaining front deformation data and stress data of the antenna structure through mechanical analysis, thirdly, preprocessing the front deformation data, so as to obtain the position error of each radiating element, fourthly, calculating a far-field pattern of an interlayer microstrip antenna according to a data-driven coupling model, fifthly, calculating and considering the wave-transparent performance influenced by the panel and cellular, sixthly, establishing an optimization design model integrating the interlayer microstrip antenna structure and electromagnetism, and seventhly, solving the optimization design model, so as to obtain an optimal integrating result. The method can realize the simultaneous optimal design of the active interlayer microstrip antenna structure and electromagnetism, shortens the development period, and improves force-electricity properties of products.

Description

Data-driven method for designing and antenna that active interlayer microstrip antenna structure and electromagnetism are comprehensive
Technical field
The invention belongs to antenna technical field, be specifically related to the comprehensive data-driven method for designing of active interlayer microstrip antenna structure and electromagnetism, that the method can be used for is airborne, the dynamo-electric Integrated design field of the structure-function integration interlayer microstrip antenna in the missile-borne, the platform such as vehicle-mounted and carrier-borne.
Background technology
Active interlayer microstrip antenna refers to the feeding network of integrated T/R assembly, wave beam control module and micro-strip antenna array are embedded in the stressed-skin construction of weapon platform, under the prerequisite that satisfies the platform structure mechanical property, also to realize higher electromagnetic performance, this antenna is a kind of active phase array antenna of Highgrade integration, it namely can as the load-carrying construction functor of weapon platform, also can be used as the radio-frequency enabled part that receives emitting electromagnetic wave.Active interlayer microstrip antenna can be applied to following various aeroamphibious weaponrys such as variant aircraft, unmanned plane, dirigible early warning plane, intelligent battlebus, stealthy battleship etc., is the gordian technique that realizes stealthyization of weapon platform, multifunction, intellectuality and high maneuverability.
At present, some scientific research institutions have recognized the importance of active interlayer microstrip antenna in weaponry of new generation both at home and abroad, and have carried out some correlative studys, and the below introduces research situation both domestic and external.
(1) as far back as the nineties in last century, in order to realize the conformal of antenna and Flight Vehicle Structure, the U.S. has taken the lead in carrying out relevant research in the world.Design the skin antenna conformal with wing such as Boeing, and studied the situation of change of wing microstrip antenna radiance in Fatigue Failure Process.NASA has also developed a kind of wing of growing the voyage unmanned plane, and its micro-strip antenna array, solar battery array and wing structure combine together fully.Flight experiment shown distortion, wave the front stress that brings out can exert an influence to electrical property.These results document " Lockyer A J; Alt K H; Coughlin D P, et al.Design and development of a conformal load-bearing smart-skin antenna:overview of the AFRL smart skin structures technology demonstration (S3TD) .In:Proceedings of SPIE; 1999:410-424. " in report is arranged.Their research has found that Service Environment load such as pneumatic, vibration and the tired front stress that brings out can exert an influence to electrical property, however they but do not provide its impact relation is described mathematical model to instruct the dynamo-electric Integrated design of structure, electromagnetism.
(2) Dutch national space laboratory has been studied the Changing Pattern with the conformal microstrip antenna electrical property under vibration environment of unmanned plane wing, Germany has proposed to be applied to the microstrip antenna framework on the intelligent battlebus, and points out that mechanical-physical amount (such as rigidity, intensity, breaking limit, manufacturing stress, thunderbolt environmental stress etc.) can cause the disorderly and inner level signal variation of antenna electromagnetic beam.These results are in document " Paul J.Callus.Conformal load-bearing antenna structure for Australian defence force aircraft [R] .Australia:Australian Government Department of Defence, 2007 ". in report is arranged.Although they by experiment method found that physical construction on the impact relation of microstrip antenna electrical property, does not provide the electrical and mechanical comprehensive method for designing that guarantees simultaneously structural behaviour and the comprehensive optimum of electrical property.
(3) research team of Korea S Pu item University of Science and Technology adopts adhering method that microstrip antenna is embedded in the composite structure.In order to improve antenna gain, Chisang You uses transmission line theory to provide the method that improves antenna gain.This research has report at document " ChisangYou; Woonbong Hwang.Design of load-bearing antenna structures by embedding technology of microstrip antenna in composite sandwich structure.Composite Structures.2005,71 (3-4): 378-382. ".Subsequently, Chisang You utilizes the mathematical statistics instrument to study based on gain optimum structure and electromagnetism collaborative design method.The method has report in document " Chisang You; Staiculescu; LaraMartin; etal.A novel hybrid electrical/mechanical optimization technique using time-domain modeling; finite element method and statistical tools for co-design and optimization of RF-integrated mechanical structures.International Journal of Numerical Modeling:Electronic Network; DevicesandFields.2007,21 (1-2): 91-101. ".Yet, this method for designing is just considered the optimum of structural behaviour and gain performance, the optimal design that does not have implementation structure and Antenna Far Field directional diagram, and this collaborative design method is only for passive interlayer microstrip antenna, and is not suitable for the active interlayer microstrip antenna designs that weaponry of new generation is badly in need of.
(4) domestic Donghua University is embedded into microstrip antenna in the three-dimensional woven fabric by textile technology, exemplar by experiment, and they have studied different weaving manners, compound substance, dielectric material etc. to the impact relation of antenna power electrical property.This research has report at document " Fujun Xu; Lan Yao; Da Zhao; et al.Performance and impact damage of a three dimensionally integrated microstrip feeding antenna structure.Composite Structures.2010,93 (1): 193-197. ".Inst. of Composite material, Harbin Polytechnic Univ. has developed buries microstrip antenna rhythmo structure exemplar, and has studied dielectric parameter and honeycomb thickness to the rule that affects of antenna power electrical property.This result document " Dai Fuhong, Wang Guangning. bury the power Electrical Analysis of microstrip antenna honeycomb sandwich construction. compound substance journal, 2011,28 (2): 231-234. " report arranged.Although domesticly carried out correlative study, yet their research object all is passive interlayer microstrip antenna, does not study the active interlayer microstrip antenna designs method of implementation structure and electromagnetism combination property optimum.
Above technical scheme has the following disadvantages: 1) can affect the power electrical property of interlayer antenna although the factors such as Service Environment, honeycomb, panel, bonding, dielectric material have been found in above-mentioned research, yet they do not set up these factors to the mathematical model of antenna electric performance impact, cause and can not just can obtain optimum structure and electromagnetism synthesis result in the initial design stage.2) existing document lacks active interlayer microstrip antenna structure and the comprehensive method for designing of electromagnetism, usually adopt at present the dynamo-electric integrated approach that separates in the engineering, namely according to electrical performance indexes, utilize respectively antenna synthesis technology and mechanics principle to design the exciting current of radiating element and physical dimension, shape and the topology etc. of structure, then check respectively intensity and the antenna electric performance of antenna structure, this dynamo-electric interlayer microstrip antenna that comprehensively is difficult to after guaranteeing to design that separates satisfies the overall performance optimum.3) because the type antenna has integrated level height, material heterogeneity and multidisciplinary property, prior art is at first used dynamo-electric separate design method, then the debugging by exemplar realizes the power electrical property expected, this dynamo-electric the separation and method for designing by rule of thumb, cause the product development cycle long, yield rate is low, cost is high, can not obtain optimum structure and electromagnetic performance to satisfy the needs of Practical.
Summary of the invention
The object of the invention is to the dynamo-electric Integrated design problem for active interlayer microstrip antenna, provide a kind of implementation structure and electromagnetism comprehensive data-driven method, in order to obtain simultaneously the comprehensive optimum active interlayer microstrip antenna of structure and electromagnetic performance, for the design of the electrical and mechanical comprehensive of the type antenna lays the foundation, avoiding existing dynamo-electric separates and the deficiency of designing technique by rule of thumb, reduce development cost, the consistance of enhancing product performance.
The embodiment of the invention is achieved in that the data-driven method for designing that a kind of active interlayer microstrip antenna structure and electromagnetism are comprehensive, it is characterized in that the comprehensive data-driven method for designing of this active interlayer microstrip antenna structure and electromagnetism may further comprise the steps:
The first step according to given electrical performance indexes, is determined radio-frequency enabled layer material, physical dimension and front radiating element layout in the interlayer microstrip antenna;
Second step according to the physical dimension of above-mentioned radio-frequency enabled layer, is determined the physical dimension of upper and lower panel and honeycomb;
The 3rd step is by front deformation data, the maximum stress σ of mechanical analysis acquisition antenna structure MaxWith maximum deformation quantity v MaxData;
In the 4th step, pre-service front deformation data obtains each radiating element site error △ x that military service load causes Ij(β, F), △ y Ij(β, F) and △ z Ij(β, F), they are functions of parameter of structure design β and military service magnitude of load F;
In the 5th step, according to above-mentioned each radiating element site error, utilize the coupling model of data-driven to calculate the antenna array electrical property of considering under front cell position error and the sandwich irradiation unit joint effect;
In the 6th step, calculate the impact of considering that panel and honeycomb gain on the interlayer microstrip antenna;
In the 7th step, go on foot the interlayer microstrip antenna array electric field far field data E that calculates according to the 5th A(θ, φ) makes up interlayer microstrip antenna structure and electromagnetism Integrated Optimization Model to determine structural design variable and little exciting current amplitude and phase place with radiating element;
The 8th step, utilize optimized algorithm to find the solution Integrated Optimization Model, whether judged result restrains, if do not have, be updated to the design variable initial value finding the solution the result who obtains, and turned back to for the 3rd step, restart finding the solution next time, otherwise its result satisfies optimum structure parameter and the exciting current of power electrical property;
The 9th step is according to antenna electric field far field data E obtained above A(θ, φ) determines minor level and beam position electrical performance indexes, and calculates the gain of interlayer microstrip antenna;
The tenth step, according to above-mentioned radiating element exciting current amplitude and the phase place that comprehensively obtains, utilize the feeding network in the active interlayer microstrip antenna of HFSS Software for Design, last, utilize integrated forming technique to make this antenna.
Further, according to given electrical performance indexes, determine that radio-frequency enabled layer material, physical dimension and the front radiating element layout method of integrated microstrip antenna is:
According to electrical property design objective such as gain, secondary lobe, centre frequency, beam angle and beam angle, at first utilize existing antenna theory to determine physical dimension such as the length L of little shape with radiating element, number and array layout and micro-strip antenna array m, width W mIn order to reduce manufacture difficulty, specify the length of radio-frequency enabled layer the same with micro-strip antenna array with width; Height H in the radio-frequency enabled layer mThickness by feeding network circuit, signal controlling and the treatment circuit of microstrip antenna dielectric-slab, integrated T/R assembly decides; The material of radio-frequency enabled layer is selected teflon and LTCC (LTCC), and little band radiation cell array is etched on the dielectric-slab, realizes the electrical property that needs.
Further, according to the physical dimension of radio-frequency enabled layer, determine that the physical dimension method of upper and lower panel and honeycomb is:
The physical dimension of upper and lower panel and honeycomb is decided by the installing space of weapon platform usually; Select the length L of above and below flaggy and keriotheca and radio-frequency enabled layer mAnd width W mIdentical.
Further, obtain front deformation data, the maximum stress σ of antenna structure by mechanical analysis MaxWith maximum deformation quantity v MaxData, the specific implementation process is as follows:
3a) set up interlayer microstrip antenna mechanics analysis model under Service Environment load such as aerodynamic loading or the temperature loading effect:
Kδ=F
In the formula, K is structural stiffness matrix, and it is the function of parameter of structure design β and physical parameter, and δ represents finite element nodal displacement array, and F is the panel load array, and its expression aerodynamic loading and temperature loading also can make up the load column that obtains by them;
3b) according to the physical dimension of above-mentioned definite covering, honeycomb and radio-frequency enabled layer, utilize the Ansys command stream to set up initial interlayer microstrip antenna Static Analysis Model of Micro-machined, its detailed process is as follows:
3b1) determine effective material parameters such as density, elastic modulus and the Poisson ratio of covering, honeycomb and microstrip antenna layer, wherein, the regular hexagon honeycomb core utilizes the equivalence of Y model to be orthotropic plate, being calculated as follows of its equivalent physical parameter:
E cx = E cy = 4 3 E s ( 1 - 3 t c 2 l c 2 ) t c 3 l c 3
E cz = 2 3 E s t c l c
G cxy = 3 3 E s ( 1 - t c 2 l c 2 ) t c 3 l c 3
G cxz = G cyz = 3 3 r G s t c l c
In the formula, E Cx, E CyAnd E CzRepresent respectively honeycomb along x, y, the equivalent elastic modulus of z direction, G Cxy, G CxzAnd G CyzRepresent respectively along xy xz, the equivalent shear modulus of yz direction, E sThe elastic modulus of expression cellular material, G sBe the modulus of shearing of cellular material, t c, l cBe respectively wall thickness and the length of side of regular hexagon honeycomb, r is correction factor, depends on manufacturing process, and theoretical value gets 1;
3b2) in Ansys, define the cell type of each layer use, upper and lower panel uses Solid45 solid element type among the present invention, and adhesive linkage has used Inter205 boundary element type to simulate;
3b3) apply military service load, utilize the distortion of antenna structure front, maximum stress and maximum deformation quantity data under the impact of Ansys software acquisition military service load.
Further, pre-service front deformation data obtains each radiating element site error Δ x that military service load causes Ij(β, F) Δ y Ij(β, F) and Δ z Ij(β, F), they are functions of parameter of structure design β and military service magnitude of load F, its specific implementation process is as follows:
(4a) from the above-mentioned front deformation data that applies after the military service loading analysis, extract position coordinates Γ after each radiating element distortion={ (x Ij, y Ij, z Ij), i=1,2 ..., M, j=1,2 ..., N}, wherein, x Ij, y IjRepresent ij the front horizontal coordinate after the radiating element distortion, z IjFront height coordinate after the expression distortion, M and N represent respectively the little band radiating element of interlayer along x, the sum of y direction;
(4b) never apply the position coordinates that extracts each radiating element expectation in the finite element model of military service load Γ o = { ( x ij o , y ij o , z ij o , ) i = 1,2 , . . . , M , j = 1,2 , . . . , N } , Wherein, The horizontal coordinate that represents ij radiating element center expectation, The Desired Height coordinate of expression radiating element center, their desired locations coordinate is determined by the antenna synthesis technology according to electrical performance indexes;
The position coordinates Γ that (4c) expects according to each radiating element obtained above oWith the position coordinates Γ after the distortion of each radiating element, calculate under the load under arms ij radiating element with respect to the site error of desired locations, it is calculated as follows:
Δ x ij ( β , F ) = x ij - x ij o
Δ y ij ( β , F ) = y ij - y ij o
Δ z ij ( β , F ) = z ij - z ij o
In the formula, Δ x Ij(β, F), Δ y Ij(β, F) and Δ z IjThe coordinate position variable quantity of ij radiating element of (β, F) expression, they are functions of parameter of structure design β and military service magnitude of load F, and load is larger, and location variation is larger.
Further, according to above-mentioned each radiating element site error, utilize the coupling model of data-driven to calculate the antenna array electrical property E that considers under front cell position error and the sandwich irradiation unit joint effect A(θ, φ):
In the formula, M and N represent that respectively the spacing between each radiating element is respectively d along horizontal coordinates x axle and the axial little band radiating element number of y xWith d y, I MnAnd Φ MnThe exciting current amplitude and the phase place that represent respectively mn radiating element, k=2 π λ 0Expression free space wave constant, λ 0Expression free space wavelength, Represent respectively the controlling antenna wave beam to point orientation, j represents imaginary part, Far field, active sandwich irradiation unit under the structural factor x impact;
The modeling method that above-mentioned formula usage data drives is calculated The hybrid modeling method specific implementation process of data-driven is as follows:
5a) at first utilize existing microstrip antenna radiating element computing formula, calculate little with radiating element in the radiating element far field of not considering under covering, honeycomb, coating and the bonded structure factor affecting Its computing formula is as follows:
In the formula, L d, W dAnd h dThe length, width and the dielectric-slab thickness that represent respectively the rectangular radiation unit;
5b) utilize data-driven model to calculate the thickness of covering, honeycomb, coating and bonded structure factor to the correction of radiating element far field influences, this data-driven model is described below:
ΔF E(x)=Re(ΔF(x))+jIm(ΔF(x))
In the formula, structural factor x=[x 1, x 2, x 3, x 4] TExpression is by plate thickness x 1, bonding thickness x 2, honeycomb thickness x 3With coating thickness x 4The vector that forms, Δ F E(x) the little correction with the radiating element far field of expression interlayer, it is a plural number, is comprised of real part Re (Δ F (x)) and imaginary part Im (Δ F (x)) two parts, these two parts all are the nonlinear functions of structural factor x;
5c) according to structural factor x=[x 1, x 2, x 3, x 4] TNumerical value, calculate the little band radiating element of the interlayer far field comprise covering, honeycomb, coating and bonding factor affecting Its mathematic(al) representation is as follows:
In the computing formula of above-mentioned interlayer microstrip antenna radiating element far field, step 5b) data-driven model in, its specific implementation step is as follows:
5b1) for little band radiating element, before design interlayer micro-strip antenna array, use the little band radiating element experiment of L interlayer of uniform experiments method for designing processing exemplar, this exemplar can reflect that different panels, honeycomb, coating and thickness of adhibited layer are to the little influence degree with radiating element of interlayer;
5b2) utilize antenna near-field test macro and three-dimensional coordinate testing tool, measure the radiating element far field of L experiment exemplar of above-mentioned manufacturing, obtain L data the sample set { (x in different structure factor x and corresponding radiating element far field i, F i), x i∈ R, F i∈ C, i=1 ..., L}, wherein, vector x=[x 1, x 2, x 3, x 4] TThe expression structural factor, F represents corresponding radiating element far field data;
5b3) according to above-mentioned actual measurement radiating element far field data F and 5a) middle traditional little band radiating element far field data of calculating Calculate the variable quantity in radiating element far field under covering, honeycomb, coating and the bonded structure factor affecting:
5b4) utilize this formula to calculate respectively the experiment exemplar of L different structure and the variable quantity in corresponding radiating element far field, and carry out the normalized of data, and then obtain experimental data collection Ψ={ (x i, Δ F i), x i∈ R, Δ F i∈ C, i=1 ..., L}, R and C represent respectively set of real numbers and set of complex numbers;
5b5) the data set Ψ after the normalization is divided into two subset Ψ 1And Ψ 2, and with Ω={ (x i, t i), i=1 ..., L} comes unified representation this two subsets, wherein t iRepresent real part and imaginary data among the complex vector located Δ F;
5b6) for sample set Ω={ (x i, t i), i=1 ..., L} utilizes the nuclear machine learning algorithm to set up respectively structural factor x to the meta-model of real part Re among the Δ F (Δ F (x)) and imaginary part Im (Δ F (x)), and its meta-model is unified to use following formula to describe:
t ( x ) = Σ i = 1 L ω i k ( x , x i ) + ω 0
In the formula, t (x) can represent real part Re (Δ F (x)) and the imaginary part Im (Δ F (x)) among the Δ F, and it is the nonlinear function of x, k (x, x i) the expression kernel function, the introducing of kernel function has solved the regression problem of nonlinear data, has avoided the difficulty of direct searching Nonlinear Mapping, and L represents the number of data sample, ω iWeights corresponding to expression kernel function, ω 0It is bias term;
5b7) above-mentioned steps 5b6) meta-model unknown parameter ω in iAnd ω 0Find the solution, two kinds of machine learning algorithms that it is characterized in that using Linear Programming Support Vector Regression and associated vector to return, its solution procedure is as follows:
(1) if uses support vector regression solve un-known parameters ω iAnd ω 0, its solution procedure is as follows:
1. at first specify kernel function, then according to normalized data sample subset Ω={ (x i, t i), i=1 ..., L} uses 5 times to deliver for a check card method definite kernel parameter, compromise constant C and error margin ε;
2. according to preassigned kernel function type, nuclear parameter, compromise constant C and error margin ε, use the linear programming algorithm to find the solution following support vector regression algorithm, obtain parameter ω 0With slack variable ξ j:
Find : ω i + , ω i - , ξ j , ω 0
Min : Σ i = 1 L ( ω i + + ω i - ) + 2 C Σ j = 1 L ξ j
s . t . t j - Σ i = 1 L ( ω i + - ω j - ) k ( x i , x j ) - ω 0 ≤ ϵ + ξ j Σ i = 1 L ( ω i + - ω j - ) k ( x i , x j ) + ω 0 - t j ≤ ϵ + ξ j ω i + ≥ 0 , ω i - ≥ 0 ξ j ≥ 0 , ( ∀ j = 1,2 , · · · , L )
3. try to achieve according to above-mentioned Unknown parameter ω in the formula computational data model below utilizing i:
ω i = ω i + - ω i -
4. the above-mentioned ω that obtains that finds the solution iAnd ω 0In the substitution formula, obtain interlayer microstrip antenna structure factor to little data model with the real part Re among the variation delta F of radiating element far field (Δ F (x)) or imaginary part Im (Δ F (x)) impact;
(2) if use associated vector regression algorithm solution procedure 5b6) in unknown parameter ω iAnd ω 0, then need to consider the impact of noise, namely step 5b6) in meta-model again be expressed as:
t(x)=y(x)+ε
y ( x ) = Σ i = 1 L ω i k ( x , x i ) + ω 0
In the formula, the prediction output of noise effect is not considered in y (x) expression, and ε represents output noise, and it is that 0 variance is σ that this noise satisfies average 2Normal distribution, the prediction output that t (x) expression is actual, satisfying average is that y (x), variance are σ 2And separate normal distribution, its probability distribution is described as p (t i)=N (t iY (x i), σ 2), wherein, operator N () expression average is μ and variances sigma 2Normal distribution, it is defined as follows:
N ( t | μ , σ 2 ) = 1 2 π σ exp ( - ( t - μ ) 2 2 σ 2 )
According to normalized data sample subset Ω={ (x i, t i), i=1 ..., L}, and utilize the method for solving that has the Method Using Relevance Vector Machine algorithm now, obtain the unknown parameter ω in the meta-model i, ω 0And σ 2
5b8) according to plural number by the principle that real part and imaginary part form, make up real part and two meta-models of imaginary part correction of above-mentioned acquisition, obtain structural factor to the data model of radiating element far field variable quantity impact.
Further, calculate the impact of considering that panel and honeycomb gain on the interlayer microstrip antenna, its computing formula is as follows:
S = Γ 01 - Γ 01 e - 2 j k 1 ( x 1 + x 4 ) + Γ 01 2 e - 2 j k o ( 2 x 2 + x 3 + H m ) - e - 2 j [ k 1 ( x 1 + x 4 ) - k o ( 2 x 2 + x 3 + H m ) ] 1 - Γ 01 2 e - 2 j k 1 ( x 1 + x 4 ) + Γ 01 e - 2 j k o ( 2 x 2 + x 3 + H m ) - Γ 01 e - 2 j [ k 1 ( x 1 + x 4 ) - k o ( 2 x 2 + x 3 + H m ) ]
In the formula, S represents to consider panel, honeycomb, bonding and coating thickness impact and the satisfied interlayer microstrip antenna reflection coefficient of opening under the condition, Wave-length constant in the expression panel, The part reflection coefficient of expression from the free space to the panel, The wave constant in the free space, ω fExpression free space angular frequency, ε 0And μ 0Represent respectively dielectric parameter and magnetic permeability in the free space, ε rAnd μ rRepresent respectively dielectric parameter and relative permeability in the panel, H mThe thickness of expression radio-frequency enabled layer.
Further, according to the interlayer microstrip antenna array electric field far field data E of above-mentioned calculating A(θ, φ) makes up interlayer microstrip antenna structure and electromagnetism Integrated Optimization Model to determine structural design variable x=[x 1, x 2, x 3, x 4, t c, l c] TWith little exciting current amplitude I with radiating element MnWith phase place Φ Mn:
Find:x,I mnmn
Min:||E A(θ,φ)-E *(θ,φ)||
s . t . S ≤ [ ϵ ] v max ≤ [ v ] σ max ≤ [ σ ] x l ≤ x ≤ x h
In the formula, E *The given expectation electric field far field of (θ, φ) expression, v MaxAnd σ MaxRepresent respectively maximum deformation quantity and maximum stress, they determine x by the Ansys analysis result in the 3rd step l, x hMinimum and the maximal value of expression structural design variable, [ε], [v] and [σ] represent respectively reflection coefficient, maximum deformation quantity and the maximum stress of permission, they are given by design objective, t c, l cThe length of side and the thickness of honeycomb wall that represent respectively the regular hexagon cellular unit, x 1, x 2, x 3, x 4The parameter meaning as previously described.。
Further, according to the antenna electric field far field data E that obtains A(θ, φ) determines minor level and beam position electrical performance indexes, and utilizes following formula to calculate the gain of interlayer microstrip antenna:
Another purpose of the embodiment of the invention is to provide a kind of active interlayer microstrip antenna, mainly formed by top panel, honeycomb, radio-frequency enabled layer and lower panel, wherein upper and lower panel and honeycomb belong to the encapsulation function layer, have mechanics carrying and heat insulation safeguard function, the plate surface also will apply camouflage coating in the above usually; The radio-frequency enabled layer mainly is comprised of the feeding network of micro-strip antenna array, integrated T/R assembly, merit parallel circuit and liquid cooling passage, beam signal controlled processing unit etc., mainly adopts low-temperature co-burning ceramic material; By using the integrated molding manufacturing process, the radio-frequency enabled layer of integrated microstrip antenna is embedded in the platform structure of panel and honeycomb composition, and then implementation structure and function solenoid is integrated.
The present invention compared with prior art has following advantage:
1) the present invention utilizes the data-driven modeling method to set up structure and the electromagnetic coupled model of active interlayer microstrip antenna, this coupling model can be analyzed front radiating element site error that military service load causes and the interlayer microstrip antenna electrical property under the sandwich irradiation cell influence, realize structure and electromagnetism integrated analysis, overcome the deficiency that current dynamo-electric separation method is difficult to realize dynamo-electric Integrated design.
2) the present invention sets up the factors such as panel, honeycomb, bonding and coating to the data model of sandwich irradiation unit far field influences with two kinds of nuclear machine learning algorithms, not only accuracy is higher for the data model that these algorithms are set up, and model is sparse, reduce the computation complexity of structure and electromagnetism Synthetical Optimization model, improved solution efficiency.
(3) structure of the present invention and electromagnetism comprehensive designing method not only utilize existing software such as Ansys to realize the accurate analysis of antenna structure mechanical property, and optimal design when can realize active interlayer microstrip antenna structure and electromagnetism, avoiding existing dynamo-electric separates and the deficiency of designing technique by rule of thumb, shortened the lead time, reduce its development cost, improved the power electrical property of product.
Description of drawings
Fig. 1 is that active interlayer microstrip antenna structure of the present invention forms synoptic diagram;
Fig. 2 is structure of the present invention and electromagnetism comprehensive designing method process flow diagram;
Fig. 3 is that radio-frequency enabled layer of the present invention forms and the physical dimension synoptic diagram;
Fig. 4 is the size Expressing of keriotheca of the present invention and cellular unit;
Fig. 5 is the size Expressing of panel of the present invention;
Fig. 6 is microstrip antenna array Column Layout of the present invention and little band radiating element synoptic diagram;
Fig. 7 is radiating element data-driven hybrid modeling method of the present invention;
Fig. 8 is the structural design variable of comprehensive designing method of the present invention;
Fig. 9 is interlayer microstrip antenna exemplar CAD figure of the present invention;
Figure 10 is the finite element analysis model of interlayer microstrip antenna of the present invention;
Figure 11 is interlayer microstrip antenna structure deformation pattern of the present invention;
Figure 12 be case of the present invention comprehensively obtain normalization direction of an electric field figure;
Figure 13 is the far field electric field comparison diagram of case of the present invention.
Embodiment
In order to make purpose of the present invention, technical scheme and advantage clearer, below in conjunction with drawings and Examples, the present invention is further elaborated.Should be appreciated that specific embodiment described herein only in order to explain the present invention, is not intended to limit the present invention.
The structure that Fig. 1 has provided a kind of active interlayer microstrip antenna of the present invention's design forms synoptic diagram, it mainly is comprised of top panel 1, honeycomb 2, radio-frequency enabled layer 3 and lower panel 4 etc., wherein upper and lower panel and honeycomb belong to the encapsulation function layer, have mechanics carrying and heat insulation safeguard function, the plate surface also will apply camouflage coating in the above usually; The radio-frequency enabled layer mainly is comprised of the feeding network of micro-strip antenna array, integrated T/R assembly, beam signal controlled processing unit etc., mainly adopts LTCC (LTCC) material.By using the integrated molding manufacturing process, the radio-frequency enabled layer of integrated microstrip antenna is embedded in the platform structure of panel and honeycomb composition, and then implementation structure and function solenoid is integrated.With the traditional antenna contrast, the height that active interlayer microstrip antenna has structure/circuit merges characteristics, has reduced antenna weight and space hold rate, has realized the conformal and light-weighted unification of antenna and structure.
2 couples of the present invention are described in further detail referring to accompanying drawing.
The first step according to given electrical performance indexes, is determined radio-frequency enabled layer material, physical dimension and the front radiating element layout of integrated microstrip antenna.
In this step, according to electrical property design objective such as gain, secondary lobe, centre frequency, beam angle and beam angle, at first utilize existing antenna theory to determine physical dimension such as the length L of little shape with radiating element, number and array layout and micro-strip antenna array m, width W mIn order to reduce manufacture difficulty, the present invention specifies the length of radio-frequency enabled layer the same with micro-strip antenna array with width.The structure that accompanying drawing 3 has provided the radio-frequency enabled layer forms and physical dimension, wherein, and height H in the radio-frequency enabled layer mThickness by feeding network circuit (comprising T/R assembly, merit parallel circuit and liquid cooling passage), signal controlling and the treatment circuit (comprising beam signal treatment circuit, supervisory circuit and power circuit) of microstrip antenna dielectric-slab, integrated T/R assembly decides.The material of radio-frequency enabled layer is selected teflon such as RT/duroid5880 and LTCC (LTCC), by existing manufacturing process, little band radiation cell array is etched on the dielectric-slab, realizes the electrical property that needs.
Second step according to the physical dimension of above-mentioned radio-frequency enabled layer, is determined the physical dimension of upper and lower panel and honeycomb.
In this step, the physical dimension of upper and lower panel and honeycomb is decided by the installing space of weapon platform usually.Yet, describing for making things convenient for the subsequent design method, the present invention selects the length L of above and below flaggy and keriotheca and radio-frequency enabled layer mAnd width W mIdentical, such as attached Figure 4 and 5, wherein, accompanying drawing 4 has also provided regular hexagon cellular unit size such as length of side l cWith thickness of honeycomb wall t c, they affect the mechanical property of interlayer microstrip antenna.In addition, plate thickness x 1, bonding thickness x 2, honeycomb thickness x 3With coating thickness x 4Also affect the power electrical property of interlayer microstrip antenna, their design is then determined by following step.
The 3rd step is by front deformation data, the maximum stress σ of mechanical analysis acquisition antenna structure MaxWith maximum deformation quantity v MaxEtc. data, the specific implementation process is as follows:
3a) set up interlayer microstrip antenna mechanics analysis model under Service Environment load such as aerodynamic loading or the temperature loading effect:
Kδ=F \*MERGEFORMAT(1)
In the formula, K is structural stiffness matrix, and it is that parameter of structure design β is (such as the length L of panel, honeycomb and radio-frequency enabled layer mAnd width W m) and the function of physical parameter (such as elastic modulus, Poisson ratio and density), δ represents finite element nodal displacement array, F is the panel load array, its expression aerodynamic loading and temperature loading, the load column that also can be obtained by they combinations.
3b) according to the physical dimension of above-mentioned definite covering, honeycomb and radio-frequency enabled layer, utilize the Ansys command stream to set up initial interlayer microstrip antenna Static Analysis Model of Micro-machined, its detailed process is as follows:
3b1) determine effective material parameters such as density, elastic modulus and the Poisson ratio of covering, honeycomb and microstrip antenna layer, wherein, the regular hexagon honeycomb core utilizes the equivalence of Y model to be orthotropic plate, being calculated as follows of its equivalent physical parameter:
E cx = E cy = 4 3 E s ( 1 - 3 t c 2 l c 2 ) t c 3 l c 3
E cz = 2 3 E s t c l c
\*MERGEFORMAT(2)
G cxy = 3 3 E s ( 1 - t c 2 l c 2 ) t c 3 l c 3
G cxz = G cyz = 3 3 r G s t c l c
\*MERGEFORMAT(3)
In the formula, E Cx, E CyAnd E CzRepresent respectively honeycomb along x, y, the equivalent elastic modulus of z direction, G Cxy, G CxzAnd G CyzRepresent respectively along xy xz, the equivalent shear modulus of yz direction, E sThe elastic modulus of expression cellular material, G sBe the modulus of shearing of cellular material, t c, l cBe respectively wall thickness and the length of side of regular hexagon honeycomb, r is correction factor, depends on manufacturing process, and theoretical value gets 1.
3b2) in Ansys, define the cell type of each layer use, upper and lower panel uses Solid45 solid element type among the present invention, and adhesive linkage has used Inter205 boundary element type to simulate;
3b3) apply military service load, utilize the data such as the distortion of antenna structure front, maximum stress and maximum deformation quantity under the impact of Ansys software acquisition military service load;
In the 4th step, pre-service front deformation data obtains each radiating element site error Δ x that military service load causes Ij(β, F) Δ y Ij(β, F) and Δ z Ij(β, F), they are functions of parameter of structure design β and military service magnitude of load F, its specific implementation process is as follows:
(4a) from the above-mentioned front deformation data that applies after the military service loading analysis, extract position coordinates Γ after each radiating element distortion={ (x Ij, y Ij, z Ij), i=1,2 ..., M, j=1,2 ..., N}, wherein, wherein, x Ij, y IjRepresent ij the front horizontal coordinate after the radiating element distortion, z IjFront height coordinate after the expression distortion, M and N represent the little band radiating element of interlayer along x, the sum of y direction;
(4b) never apply the position coordinates that extracts each radiating element expectation in the finite element model of military service load Γ o = { ( x ij o , y ij o , z ij o , ) i = 1,2 , . . . , M , j = 1,2 , . . . , N } , Wherein, The horizontal coordinate that represents ij radiating element center expectation, The Desired Height coordinate that represents ij radiating element center, their desired locations coordinate utilizes the antenna synthesis technology to determine according to electrical performance indexes;
The position coordinates Γ that (4c) expects according to each radiating element obtained above oWith the position coordinates Γ after the distortion of each radiating element, calculate under the load under arms ij radiating element with respect to the site error of desired locations, it is calculated as follows:
Δ x ij ( β , F ) = x ij - x ij o
Δ y ij ( β , F ) = y ij - y ij o
Δ z ij ( β , F ) = z ij - z ij o
\*MERGEFORMAT(4)
In the formula, Δ x Ij(β, F), Δ y Ij (β, F) and Δ z IjThe coordinate position variable quantity of ij radiating element of (β, F) expression, they are functions of parameter of structure design β and military service magnitude of load F, and load is larger, and location variation is larger.
The change of these radiating element positions has reflected the impact of impact shock that Weapons platform construction unavoidably is subject to, temperature and the load such as pneumatic when fast reserve.These load can cause malformation, and then cause the microstrip antenna front cell position in the embedded structure to change.The change of antenna radiation unit position will cause the radiating element phase error, and then cause aerial array antenna pattern performance to produce larger variation, that is to say that change and the minor level that can cause beam position increase, wherein, beam position directly has influence on the accuracy of antenna location, for synthetic-aperture radar, directly have influence on the radar imagery quality; And the increase of minor level can cause the reduction of Anti-jamming Ability for Radar.
In the 5th step, according to above-mentioned each radiating element site error, utilize the coupling model of data-driven to calculate the antenna array electrical property E that considers under front cell position error and the sandwich irradiation unit joint effect A(θ, φ):
\*MERGEFORMAT(5)
In the formula, M and N represent respectively along horizontal coordinates x axle and the axial little band radiating element number of y, accompanying drawing 6 (a) has provided little band radiating element and has arranged by equidistant rectangular grid battle array, this array antenna has MN radiating element, be positioned on the oxy plane, the spacing between each radiating element is respectively d xWith d y, each unit uses widely used rectangular microstrip radiating element in the engineering, and accompanying drawing 6 (b) has provided a kind of rectangular microstrip radiative unit structure, L among the figure d, W dAnd h dThe length, width and the dielectric-slab thickness that represent respectively the rectangular radiation unit; I MnAnd Φ MnThe exciting current amplitude and the phase place that represent respectively mn radiating element; K=2 π λ 0Expression free space wave constant, λ 0Expression free space wavelength, Represent respectively the controlling antenna wave beam to point orientation; J represents imaginary part, The far field, active sandwich irradiation unit under the structural factor x impact is considered in expression.
In the above-mentioned formula, it is mainly characterized in that the modeling method calculating that usage data drives Accompanying drawing 7 provides the hybrid modeling method synoptic diagram of the data-driven of the present invention's proposition, and with reference to accompanying drawing 7, its specific implementation process is as follows:
5a) at first utilize existing microstrip antenna radiating element computing formula, calculate little with radiating element in the radiating element far field of not considering under covering, honeycomb, coating and the structural factor impact such as bonding Its computing formula is as follows:
\*MERGEFORMAT(6)
In the formula, L d, W dAnd h dThe length, width and the dielectric-slab thickness that represent respectively the rectangular radiation unit are such as accompanying drawing 6 (b).
5b) utilize data-driven model to calculate the thickness of covering, honeycomb, coating and the structural factor such as bonding to the correction of radiating element far field influences, this data-driven model is described below:
ΔF E(x)=Re(ΔF(x))+jIm(ΔF(x))
\*MERGEFORMAT(7)
In the formula, structural factor x=[x 1, x 2, x 3, x 4] TExpression is by plate thickness x 1, bonding thickness x 2, honeycomb thickness x 3With coating thickness x 4The vector that forms, Δ F E(x) the little correction with the radiating element far field of expression interlayer, it is a plural number, is comprised of real part Re (Δ F (x)) and imaginary part Im (Δ F (x)) two parts, these two parts all are the nonlinear functions of structural factor x.
5c) according to structural factor x=[x 1, x 2, x 3, x 4] TNumerical value, calculate the little band radiating element of the interlayer far field comprise covering, honeycomb, coating and bonding factor affecting Its mathematic(al) representation is as follows:
\*MERGEFORMAT(8)
In the computing formula of above-mentioned interlayer microstrip antenna radiating element far field, it is mainly characterized in that step 5b) in data-driven model, its specific implementation step is as follows:
5b1) for little band radiating element, before design interlayer micro-strip antenna array, use the little band radiating element experiment of L interlayer of uniform experiments method for designing processing exemplar, this exemplar can reflect that different panels, honeycomb, coating and thickness of adhibited layer are to the little influence degree with radiating element of interlayer;
5b2) utilize antenna near-field test macro and three-dimensional coordinate testing tool, measure the radiating element far field of L experiment exemplar of above-mentioned manufacturing, obtain L data the sample set { (x in different structure factor x and corresponding radiating element far field i, F i), x i∈ R, F i∈ C, i=1 ..., L}, wherein, vector x=[x 1, x 2, x 3, x 4] TThe expression structural factor, F represents corresponding radiating element far field data, R and C represent respectively set of real numbers and set of complex numbers;
5b3) according to above-mentioned actual measurement radiating element far field data F and 5a) middle traditional little band radiating element far field data of calculating Calculate the variable quantity in the radiating element far field under covering, honeycomb, coating and the bonding factor affecting:
\*MERGEFORMAT(9)
5b4) utilize this formula to calculate respectively the experiment exemplar of L different structure and the variable quantity in corresponding radiating element far field, and carry out the normalized of data, and then obtain experimental data collection Ψ={ (x i, Δ F i), x i∈ R, Δ F i∈ C, i=1 ..., L};
5b5) the data set Ψ after the normalization is divided into two subset Ψ 1And Ψ 2, and with Ω={ (x i, t i), i=1 ..., L} comes this two subset of unified representation, and wherein ti represents real part or the imaginary data among the complex vector located Δ F;
5b6) for sample set Ω={ (x i, t i), i=1 ..., L} utilizes the nuclear machine learning algorithm to set up respectively structural factor x to the meta-model of real part Re among the Δ F (Δ F (x)) and imaginary part Im (Δ F (x)), and its meta-model is unified to use following formula to describe:
t ( x ) = Σ i = 1 L ω i k ( x , x i ) + ω 0 \*MERGEFORMAT(10)
In the formula, real part Re (Δ F (x)) or imaginary part Im (Δ F (x)) among t (x) the expression Δ F, it is the nonlinear function of x, k (x, x i) be the kernel function shown in the table 1, the introducing of kernel function has solved the regression problem of nonlinear data, has avoided the difficulty of direct searching Nonlinear Mapping, and L represents the number of data sample, ω iWeights corresponding to expression kernel function, ω 0It is bias term;
The kernel function that table 1 the present invention uses
5b7) two meta-models of correction of combination real part and imaginary part can obtain step 5b) in structural factor on the data model of far field, sandwich irradiation unit variable quantity impact;
Unknown parameter ω in the above-mentioned meta-model iAnd ω 0, can use following two kinds of nuclear machine learning algorithms to find the solution, the concrete solution procedure of these two kinds of algorithms is as follows.
(1) if the unknown parameter ω in the use support vector regression solving model iAnd ω 0, its solution procedure is as follows:
1. at first specify kernel function, then according to normalized data sample subset Ω={ (x i, t i), i=1 ..., L} uses 5 times of cross validation method definite kernel parameters, compromise constant C and error margin ε;
2. according to preassigned kernel function type, nuclear parameter, compromise constant C and error margin ε, use the linear programming algorithm to find the solution following support vector regression algorithm, obtain parameter to be found the solution With slack variable ξ j:
Find : ω i + , ω i - , ξ j , ω 0
Min : Σ i = 1 L ( ω i + + ω i - ) + 2 C Σ j = 1 L ξ j
s . t . t j - Σ i = 1 L ( ω i + - ω j - ) k ( x i , x j ) - ω 0 ≤ ϵ + ξ j Σ i = 1 L ( ω i + - ω j - ) k ( x i , x j ) + ω 0 - t j ≤ ϵ + ξ j ω i + ≥ 0 , ω i - ≥ 0 ξ j ≥ 0 , ( ∀ j = 1,2 , · · · , L )
\*MERGEFORMAT(11)
3. try to achieve according to above-mentioned Unknown parameter ω in the formula computational data model below utilizing i:
ω i = ω i + - ω i - \*MERGEFORMAT(12)
4. the above-mentioned ω that obtains that finds the solution iAnd ω 0In the substitution formula, obtain interlayer microstrip antenna structure factor to little data model with the real part Re among the variation delta F of radiating element far field (Δ F (x)) or imaginary part Im (Δ F (x)) impact.
(2) if the unknown parameter ω in the use associated vector Regressive Solution model iAnd ω 0, its implementation procedure is as follows:
Associated vector returns and support vector regression all belongs to the nuclear machine learning algorithm, the contrast support vector regression, and it is the simplest that associated vector returns the model that obtains.In addition, associated vector returns can not only provide data model, also can provide the probability distribution of model prediction output.Therefore, if consider the impact of prediction output noise ε, the model shown in the formula also can be expressed as:
y ( x ) = Σ i = 1 L ω i k ( x , x i ) + ω 0
\*MERGEFORMAT(13)
t(x)=y(x)+ε
\*MERGEFORMAT(14)
In the formula, the prediction output of noise effect is not considered in y (x) expression, and ε represents the output noise error, and it satisfies average is that 0 variance is σ 2Normal distribution, the prediction output that t (x) expression is actual, they satisfy average is that y (x), variance are σ 2And separate normal distribution, its probability distribution is described as p (t i)=N (t iY (x i), σ 2), wherein, operator N () expression average is μ and variances sigma 2Normal distribution, it is defined as follows:
N ( t | μ , σ 2 ) = 1 2 π σ exp ( - ( t - μ ) 2 2 σ 2 )
\*MERGEFORMAT(15)
According to above-mentioned analysis, formula and shown in model also can use vector representation to be
t(x)=Φω+ε \*MERGEFORMAT(16)
In the formula, during ω by weights ω iThe vectorial ω that forms=[ω 0, ω 1, ω 2ω L] T, Φ is that the dimension by each structural factor x substitution kernel function gained is the design matrix of L * (L+1):
\*MERGEFORMAT(17)
Utilize normalized data sample subset Ω={ (x i, t i), i=1 ..., L} finds the solution unknown parameter ω in the data model and the variance of output noise ε is σ 2, its concrete solution procedure is as follows:
1. the in advance type of kernel function shown in the option table and kernel function, initialization prediction output noise variances sigma 2, the super parameter alpha of weights=[α 0, α 1, α 2..., α L] TAnd maximum iteration time;
2 according to the design matrix Φ that provides previously and training data sample t, calculates weights ω=[ω 0, ω 1, ω 2ω N] TPosterior probability distribute:
p ( ω | t , α , σ 2 ) = ( 2 π ) - L + 1 2 | Σ | - 1 2 exp [ - ( ω - μ ) T - Σ - 1 ( ω - μ ) 2 ]
\*MERGEFORMAT(18)
Wherein, the computing formula of the average μ of weights ω and covariance Σ is as follows:
μ=σ -2ΣΦ Tt
\*MERGEFORMAT(19)
Σ=(σ -2Φ TΦ+A) -1
\*MERGEFORMAT(20)
In the formula, the formulation of matrix Ω and A is as follows:
Ω=σ 2I+ΦA -1ΦT
\*MERGEFORMAT(21)
\*MERGEFORMAT(22)
In the formula, I representation unit matrix, other parameter such as top introduction.
3. according to above-mentioned super parameter alpha=[α 0, α 1, α 2..., α L] TWith covariance matrix Σ, calculate intermediate quantity parameter γ i:
γ i=1-α iΣ i,i
\*MERGEFORMAT(23)
In the formula, Σ I, iI diagonal angle unity element among the covariance matrix Σ of posteriority weights ω, γ i∈ [0,1] a kind ofly estimates, the ω that its expression is estimated by training data iCredibility is worked as α iWhen very large, weights ω iBecause priori Constraint, can cause γ i=0, otherwise, α worked as iLess, γ i=1;
4. utilize following formula to estimate excess of export parameter vector α=[α 0, α 1, α 2..., α L] TWith prediction output variance σ 2:
α i new = γ i μ i 2 \*MERGEFORMAT(24)
( σ 2 ) new = | | t - Φμ | | 2 L - Σ i = 0 L γ i
\*MERGEFORMAT(25)
In the formula, 2) NewExpression utilizes super parameter and the prediction output noise variance after above-mentioned formula iteration is upgraded, μ respectively iThe average of i unit among the expression weight vector ω;
5. order Then return step 2. in, restart iterative next time, when satisfying max (log α i)≤10 -3The time or iterations when reaching preassigned number of times, iterative finishes.
6. after above-mentioned iteration finishes, obtain super parameter alpha=[α 0, α 1, α 2..., α L] T, prediction output variance σ 2, weights ω average μ and covariance Σ, wherein some unit of super parameter vector α will be tending towards infinitely great, according to formula, corresponding weights ω iAverage be 0, this means that corresponding basis function can prune the settled α of middle finger of the present invention i〉=10 9In time, prune basis function, and then the implementation model rarefaction.
7. ω in the above-mentioned average μ substitution of finding the solution, obtain new data x *Corresponding output t *Data drive model and output t *Probability distribution:
y(x *)=μ Tφ(x *)
\*MERGEFORMAT(26)
p ( t * | t ) = N ( t * | y ( x * ) , σ * 2 )
\*MERGEFORMAT(27)
In the formula, new data x *Corresponding basis function vector φ (x *)=[1, k (x *, x 1), k (x *, x 2) ..., k (x *, x L)] T, prediction output t *Satisfying average is y (x *) and variance be Gaussian distribution, wherein 6. the expression above-mentioned steps is finally found the solution and is obtained the estimating noise variances sigma 2
The 6th step, calculate the impact of considering that panel and honeycomb gain on the interlayer microstrip antenna, it has reflected electromagnetic wave penetrate capability, its computing formula is as follows:
S = Γ 01 - Γ 01 e - 2 j k 1 ( x 1 + x 4 ) + Γ 01 2 e - 2 j k o ( 2 x 2 + x 3 + H m ) - e - 2 j [ k 1 ( x 1 + x 4 ) - k o ( 2 x 2 + x 3 + H m ) ] 1 - Γ 01 2 e - 2 j k 1 ( x 1 + x 4 ) + Γ 01 e - 2 j k o ( 2 x 2 + x 3 + H m ) - Γ 01 e - 2 j [ k 1 ( x 1 + x 4 ) - k o ( 2 x 2 + x 3 + H m ) ]
\*MERGEFORMAT(28)
In the formula, S represents to consider panel, honeycomb, bonding and coating thickness impact and the satisfied interlayer microstrip antenna reflection coefficient of opening under the condition, Wave-length constant in the expression panel, The part reflection coefficient of expression from the free space to the panel, The wave constant in the free space, ω fExpression free space angular frequency, ε rAnd μ rRepresent respectively dielectric parameter and relative permeability in the panel, ε 0And μ 0Dielectric parameter and magnetoconductivity in the difference free space, parameter x 1, x 2, x 3, x 4And H mSee accompanying drawing 8.
In the 7th step, go on foot the interlayer microstrip antenna array electric field far field data E that calculates according to the 5th A(θ, φ) makes up interlayer microstrip antenna structure and electromagnetism Integrated Optimization Model to determine structural design variable x=[x 1, x 2, x 3, x 4, t c, l c] TWith little exciting current amplitude I with radiating element MnWith phase place Φ Mn:
Find:x,I mnmn
Min:||E A(θ,φ)-E *(θ,φ)
s . t . S ≤ [ ϵ ] v max ≤ [ v ] σ max ≤ [ σ ] x l ≤ x ≤ x h
\*MERGEFORMAT(29)
In the formula, E *The given expectation electric field far field of (θ, φ) expression, v MaxAnd σ MaxRepresent respectively maximum deformation quantity and maximum stress, they determine x by the Ansys analysis result in the 3rd step l, x hMinimum and the maximal value of expression structural design variable, [ε], [v] and [σ] represent respectively reflection coefficient, maximum deformation quantity and the maximum stress of permission, and they are given by design objective, and other parameter is as previously described.
The 8th step, utilize existing optimized algorithm to find the solution above-mentioned Integrated Optimization Model, whether judged result restrains, if do not have, this is found the solution the result who obtains be updated to the design variable initial value, and turned back to for the 3rd step, restart finding the solution next time, otherwise its result satisfies optimum structure parameter and the exciting current of power electrical property.
The 9th step is according to antenna electric field far field data E obtained above A(θ, φ) determines the electrical performance indexes such as minor level and beam position, and utilizes following formula to calculate the gain of interlayer microstrip antenna:
\*MERGEFORMAT(30)
The tenth step, according to above-mentioned radiating element exciting current amplitude and the phase place that comprehensively obtains, utilize the feeding network in the active interlayer microstrip antenna of HFSS Software for Design, last, utilize integrated forming technique to make this antenna.
Advantage of the present invention can further specify by following S frequency range interlayer microstrip antenna.
The interlayer microstrip antenna of this part take centre frequency as 2.5GHz verified the validity of said structure and electromagnetism comprehensive designing method as case.The design objective of this interlayer microstrip antenna is as follows: gain is greater than 15dB, and band is wider than 50MHz, and secondary lobe is greater than 12dB, and bendind rigidity is greater than 160N/mm.
According to above-mentioned electrical performance indexes, at first design microstrip antenna layout and the physical dimension of radio-frequency enabled layer.In this exemplar, we have designed 8 little band radiating elements, its long and wide 800mm and 200mm of being respectively.Accompanying drawing 9 has provided the CAD of this checking exemplar, has also provided the original depth of panel, honeycomb and radio-frequency enabled layer among the figure, and they are respectively 2.5mm, 22.5mm and 3mm, and these thickness are not final design result.
Utilize the inventive method, above-mentioned cad model is imported in the Ansys software set up its geometric model, as shown in Figure 10.In this case, Selection Floater of the present invention, honeycomb and radio-frequency enabled functional layer are used Solid45 solid element type, and the bonding interface of interlayer adopts Inter205 boundary element type to simulate.When grid division, length direction is got 40 of nodes, and Width is got 9 of nodes, and thickness direction is got 10 of nodes.Apply distributed load 1000N at this model line of symmetry Nodes, model two symmetrical base nodes apply the displacement constraint of three directions, as shown in Figure 10.In addition, in Ansys, also need to arrange some physical parameters, in this case, upper and lower panel uses fiberglass epoxy resin composite material plate, the radio-frequency enabled layer uses the RT/duroid5880 material, the regular hexagon paper honeycomb that keriotheca uses, and the initial value of its wall thickness and the length of side is t c=0.6mm and l c=5mm, the elastic modulus of honeycomb are E s=3600MPa and modulus of shearing are G s=1900MPa utilizes the present invention formula that provides and the equivalent parameters that can calculate panel, keriotheca and radio-frequency enabled layer in the 3rd step, as shown in table 2.These parameters are input in the Ansys software, and then imposed load just can be analyzed the mechanical property of this checking exemplar, obtains the data such as front deflection, maximum stress and maximum deformation quantity.Accompanying drawing 11 has provided a kind of malformation analysis result, after the distortion front represents imposed load among the figure, and the deformation of antenna array structure, and desirable front represents not have the front structure of imposed load expectation.
The equivalent parameters that table 2 mechanical analysis is used
According to data-driven method for designing of the present invention, comprehensively go out the structural parameters x=[x in the present case 1, x 2, x 3, x 4, t c, l c] TWith each little exciting current amplitude I with radiating element MnWith phase place Φ Mn, make it satisfy expectation electric field far field given in advance, see the expectation electric field curve in the accompanying drawing 11.When finding the solution formula, in the present case value of [ε], [v] and [σ] be 1,0.1mm and 300MPa, the minimum zone x of x lWith maximum magnitude x hValue be respectively x l=[1,0.01,5,0.01,0.1,4] TMm, x h=[10,2,50,1,2,8] TMm.Use the particle swarm optimization algorithm formula, be met the design result of expection power electrical performance indexes.Obtain structural factor x=[2.31 after finding the solution, 0.32,7.25,0.56,0.74,5.3] T8 little exciting current amplitude and phase places with radiating element are as shown in table 3, normalization far-field pattern such as accompanying drawing 12 after comprehensive, as we can see from the figure, far-field pattern after utilizing structure and electromagnetism comprehensive is almost identical with the expectation electric field, secondary lobe after particularly comprehensive is 14.34dB, satisfies the secondary lobe requirement of expection.
Radiating element exciting current amplitude and phase place after table 3 is comprehensive
According to above-mentioned current excitation after comprehensive, utilize formula to calculate its gain.Accompanying drawing 13 has provided the far field electric field of this interlayer microstrip antenna after comprehensive, can find maximum electric field intensity namely to gain from figure and be 15.9dB, and by the Ansys mechanical analysis, can calculate the bendind rigidity of the inventive method after comprehensive is 176N/m.Table 4 has provided the electrical performance indexes and the contrast situation of desired design index that obtains behind this case antenna comprehensive, can see from table, utilizes the active interlayer microstrip antenna of the inventive method design to satisfy the design objective of expecting fully.
The complex optimum result of the active interlayer microstrip antenna of table 4
In order further to verify the validity of the inventive method, this case has also used traditional dynamo-electric separate design method respectively from the subject angle of structure and electromagnetism, designs this case antenna.Accompanying drawing 13 has provided the far field direction of an electric field figure behind two kinds of method synthesis, as we can see from the figure, the gain that the dynamo-electric separation method of the ratio of gains that the inventive method comprehensively obtains obtains surpasses 0.5dB, moreover, by mechanical analysis, the maximum distortion of the inventive method antenna array structure under load is 0.035mm, and the maximum distortion of dynamo-electric separation method is 0.15mm, it means that bendind rigidity is poor, and the ratio of rigidity classic method of the antenna structure that this explanation the inventive method comprehensively obtains will be got well.
Experiment by present case, optimal design when can find that the present invention not only can realize active interlayer microstrip antenna structure and electromagnetism, avoid existing dynamo-electric and separate and the deficiency of method for designing by rule of thumb, shortened the lead time, improved mechanics and the electromagnetic performance of product.
The above only is preferred embodiment of the present invention, not in order to limiting the present invention, all any modifications of doing within the spirit and principles in the present invention, is equal to and replaces and improvement etc., all should be included within protection scope of the present invention.

Claims (10)

1. the comprehensive data-driven method for designing of an active interlayer microstrip antenna structure and electromagnetism is characterized in that the comprehensive data-driven method for designing of this active interlayer microstrip antenna structure and electromagnetism may further comprise the steps:
The first step according to given electrical performance indexes, is determined radio-frequency enabled layer material, physical dimension and front radiating element layout in the interlayer microstrip antenna;
Second step according to the physical dimension of above-mentioned radio-frequency enabled layer, is determined the physical dimension of upper and lower panel and honeycomb;
The 3rd step is by front deformation data, the maximum stress σ of mechanical analysis acquisition antenna structure MaxWith maximum deformation quantity v MaxData;
In the 4th step, pre-service front deformation data obtains each radiating element site error Δ x that military service load causes Ij(β, F), Δ y Ij(β, F) and Δ z Ij(β, F), they are functions of parameter of structure design β and military service magnitude of load F;
In the 5th step, according to above-mentioned each radiating element site error, utilize the coupling model of data-driven to calculate the antenna array electrical property of considering under front cell position error and the sandwich irradiation unit joint effect;
In the 6th step, calculate the impact of considering that panel and honeycomb gain on the interlayer microstrip antenna;
In the 7th step, go on foot the interlayer microstrip antenna array electric field far field data E that calculates according to the 5th A(θ, φ) makes up interlayer microstrip antenna structure and electromagnetism Integrated Optimization Model to determine structural design variable and little exciting current amplitude and phase place with radiating element;
The 8th step, utilize optimized algorithm to find the solution Integrated Optimization Model, whether judged result restrains, if do not have, be updated to the design variable initial value finding the solution the result who obtains, and turned back to for the 3rd step, restart finding the solution next time, otherwise its result satisfies optimum structure parameter and the exciting current of power electrical property;
The 9th step is according to antenna electric field far field data E obtained above A(θ, φ) determines minor level and beam position electrical performance indexes, and calculates the gain of interlayer microstrip antenna;
The tenth step, according to above-mentioned radiating element exciting current amplitude and the phase place that comprehensively obtains, utilize the feeding network in the active interlayer microstrip antenna of HFSS Software for Design, last, utilize integrated forming technique to make this antenna.
2. the comprehensive data-driven method for designing of active interlayer microstrip antenna structure as claimed in claim 1 and electromagnetism, it is characterized in that, according to given electrical performance indexes, determine that radio-frequency enabled layer material, physical dimension and the front radiating element layout method of interlayer microstrip antenna is:
According to electrical property design objective such as gain, secondary lobe, centre frequency, beam angle and beam angle, at first utilize existing antenna theory to determine physical dimension such as the length L of little shape with radiating element, number and array layout and micro-strip antenna array m, width W mIn order to reduce manufacture difficulty, specify the length of radio-frequency enabled layer the same with micro-strip antenna array with width; Height H in the radio-frequency enabled layer mThickness by feeding network circuit, signal controlling and the treatment circuit of microstrip antenna dielectric-slab, integrated T/R assembly decides; The material of radio-frequency enabled layer is selected teflon and LTCC, and little band radiation cell array is etched on the dielectric-slab, realizes the electrical property that needs.
3. the comprehensive data-driven method for designing of active interlayer microstrip antenna structure as claimed in claim 1 and electromagnetism is characterized in that, according to the physical dimension of radio-frequency enabled layer, determines that the physical dimension method of upper and lower panel and honeycomb is:
The physical dimension of upper and lower panel and honeycomb is decided by the installing space of weapon platform usually; Select the length L of above and below flaggy and keriotheca and radio-frequency enabled layer mAnd width W mIdentical.
4. the comprehensive data-driven method for designing of active interlayer microstrip antenna structure as claimed in claim 1 and electromagnetism is characterized in that, obtains front deformation data, the maximum stress σ of antenna structure by mechanical analysis MaxWith maximum deformation quantity v MaxData, the specific implementation process is as follows:
3a) set up interlayer microstrip antenna mechanics analysis model under Service Environment load such as aerodynamic loading or the temperature loading effect:
Kδ=F
In the formula, K is structural stiffness matrix, and it is the function of parameter of structure design β and physical parameter, and δ represents finite element nodal displacement array, and F is the panel load array, and its expression aerodynamic loading and temperature loading also can make up the load column that obtains by them;
3b) according to the physical dimension of above-mentioned definite covering, honeycomb and radio-frequency enabled layer, utilize the Ansys command stream to set up initial interlayer microstrip antenna Static Analysis Model of Micro-machined, its detailed process is as follows:
3b1) determine effective material parameters such as density, elastic modulus and the Poisson ratio of covering, honeycomb and microstrip antenna layer, wherein, the regular hexagon honeycomb core utilizes the equivalence of Y model to be orthotropic plate, being calculated as follows of its equivalent physical parameter:
E cx = E cy = 4 3 E s ( 1 - 3 t c 2 l c 2 ) t c 3 l c 3
E cz = 2 3 E s t c l c
G cxy = 3 3 E s ( 1 - t c 2 l c 2 ) t c 3 l c 3
G cxz = G cyz = 3 3 r G s t c l c
In the formula, E Cx, E CyAnd E CzRepresent respectively honeycomb along x, y, the equivalent elastic modulus of z direction, G Cxy, G CxzAnd G CyzRepresent respectively along xy xz, the equivalent shear modulus of yz direction, E sThe elastic modulus of expression cellular material, G sBe the modulus of shearing of cellular material, t c, l cBe respectively wall thickness and the length of side of regular hexagon honeycomb, r is correction factor, depends on manufacturing process, and theoretical value gets 1;
3b2) in Ansys, define the cell type of each layer use, upper and lower panel uses Solid45 solid element type, and adhesive linkage has used Inter205 boundary element type to simulate;
3b3) apply military service load, utilize the distortion of antenna structure front, maximum stress and maximum deformation quantity data under the impact of Ansys software acquisition military service load.
5. the comprehensive data-driven method for designing of active interlayer microstrip antenna structure as claimed in claim 1 and electromagnetism is characterized in that, pre-service front deformation data obtains each radiating element site error Δ x that military service load causes Ij(β, F) Δ y Ij(β, F) and Δ z Ij(β, F), they are functions of parameter of structure design β and military service magnitude of load F, its specific implementation process is as follows:
(4a) from the above-mentioned front deformation data that applies after the military service loading analysis, extract position coordinates Γ after each radiating element distortion={ (x Ij, y Ij, z Ij), i=1,2 ..., M, j=1,2 ..., N}, wherein, x Ij, y IjRepresent ij the front horizontal coordinate after the radiating element distortion, z IjFront height coordinate after the expression distortion, M and N represent respectively the little band radiating element of interlayer along x, the sum of y direction;
(4b) never apply the position coordinates that extracts each radiating element expectation in the finite element model of military service load Γ o = { ( x ij o , y ij o , z ij o ) , i = 1,2 , . . . , M , j = 1,2 , . . . , N } , Wherein, The horizontal coordinate that represents ij radiating element center expectation, The Desired Height coordinate of expression radiating element center, their desired locations coordinate is determined by the antenna synthesis technology according to electrical performance indexes;
(4c) according to each radiating element desired locations coordinate Γ obtained above oWith the position coordinates Γ after the distortion of each radiating element, calculate under the load under arms ij radiating element with respect to the site error of desired locations, it is calculated as follows:
Δ x ij ( β , F ) = x ij - x ij o
Δ y ij ( β , F ) = y ij - y ij o
Δ z ij ( β , F ) = z ij - z ij o
In the formula, Δ x Ij(β, F), Δ y Ij(β, F) and Δ z IjThe coordinate position variable quantity of ij radiating element of (β, F) expression, they are functions of parameter of structure design β and military service magnitude of load F, and load is larger, and location variation is larger.
6. the comprehensive data-driven method for designing of active interlayer microstrip antenna structure as claimed in claim 1 and electromagnetism, it is characterized in that, according to above-mentioned each radiating element site error, utilize the coupling model of data-driven to calculate the antenna array electrical property E that considers under front cell position error and the sandwich irradiation unit joint effect A(θ, φ):
In the formula, M and N represent that respectively the spacing between each radiating element is respectively d along horizontal coordinates x axle and the axial little band radiating element number of y xWith d y, I MnAnd Φ MnThe exciting current amplitude and the phase place that represent respectively mn radiating element, k=2 π λ 0Expression free space wave constant, λ 0Expression free space wavelength, Represent respectively the controlling antenna wave beam to point orientation, j represents imaginary part, Far field, active sandwich irradiation unit under the structural factor x impact;
The modeling method that above-mentioned formula usage data drives is calculated The hybrid modeling method specific implementation process of data-driven is as follows:
5a) at first utilize existing microstrip antenna radiating element computing formula, calculate little with radiating element in the radiating element far field of not considering under covering, honeycomb, coating and the bonding factor affecting Its computing formula is as follows:
In the formula, L d, W dAnd h dThe length, width and the dielectric-slab thickness that represent respectively the rectangular radiation unit;
5b) utilize data-driven model to calculate covering, honeycomb, coating and bonding thickness to the correction of radiating element far field influences, this data-driven model is described below:
ΔF E(x)=Re(ΔF(x))+jIm(ΔF(x))
In the formula, structural factor x=[x 1, x 2, x 3, x 4] TExpression is by plate thickness x 1, bonding thickness x 2, honeycomb thickness x 3With coating thickness x 4The vector that forms, Δ F E(x) the little correction with the radiating element far field of expression interlayer, it is a plural number, is comprised of real part Re (Δ F (x)) and imaginary part Im (Δ F (x)) two parts, these two parts all are the nonlinear functions of structural factor x;
5c) according to structural factor x=[x 1, x 2, x 3, x 4] TNumerical value, calculate the little band radiating element of the interlayer far field comprise under covering, honeycomb, coating and the bonding factor affecting Its mathematic(al) representation is as follows:
In the computing formula of above-mentioned interlayer microstrip antenna radiating element far field, step 5b) data-driven model in, its specific implementation step is as follows:
5b1) for little band radiating element, before design interlayer micro-strip antenna array, use the little band radiating element experiment of L interlayer of uniform experiments method for designing processing exemplar, this exemplar can reflect that different panels, honeycomb, coating and thickness of adhibited layer are to the little influence degree with radiating element of interlayer;
5b2) utilize antenna near-field test macro and three-dimensional coordinate testing tool, measure the radiating element far field of L experiment exemplar of above-mentioned manufacturing, obtain L data the sample set { (x in different structure factor x and corresponding radiating element far field i, F i), x i∈ R, F i∈ C, i=1 ..., L}, wherein, vector x=[x 1, x 2, x 3, x 4] TThe expression structural factor, F represents corresponding radiating element far field data;
5b3) according to radiating element far field data F and the 5a of above-mentioned actual measurement) in traditional little band radiating element far field data of calculating Calculate the variable quantity in radiating element far field under covering, honeycomb, coating and the bonded structure factor affecting:
5b4) utilize this formula to calculate respectively the experiment exemplar of L different structure and the variable quantity in corresponding radiating element far field, and carry out the normalized of data, and then obtain experimental data collection Ψ={ (x i, Δ F i), x i∈ R, Δ F i∈ C, i=1 ..., L}, R and C represent respectively set of real numbers and set of complex numbers;
5b5) the data set Ψ after the normalization is divided into two subset Ψ 1And Ψ 2, and with Ω={ (x i, t i), i=1 ..., L} comes unified representation this two subsets, wherein t iRepresent real part or imaginary data among the complex vector located △ F;
5b6) for sample set Ω={ (x i, t i), i=1 ..., L} utilizes the nuclear machine learning algorithm to set up respectively structural factor x to the meta-model of real part Re among the Δ F (Δ F (x)) and imaginary part Im (Δ F (x)), and its meta-model is unified to use following formula to describe:
t ( x ) = Σ i = 1 L ω i k ( x , x i ) + ω 0
In the formula, t (x) can represent real part Re (Δ F (x)) or the imaginary part Im (Δ F (x)) among the Δ F, and it is the nonlinear function of x, k (x, x i) the expression kernel function, the introducing of kernel function has solved the regression problem of nonlinear data, has avoided the difficulty of direct searching Nonlinear Mapping, and L represents the number of data sample, ω iWeights corresponding to expression kernel function, ω 0It is bias term;
5b7) above-mentioned steps 5b6) meta-model unknown parameter ω in iAnd ω 0Find the solution, two kinds of algorithms that it is characterized in that using support vector regression and associated vector to return, its solution procedure is as follows:
(1) if uses support vector regression solve un-known parameters ω iAnd ω 0, its solution procedure is as follows:
1. at first specify kernel function, then according to normalized data sample subset Ω={ (x i, t i), i=1 ..., L} uses 5 times to deliver for a check card method definite kernel parameter, compromise constant C and error margin ε;
2. according to preassigned kernel function type, nuclear parameter, compromise constant C and error margin ε, use the linear programming algorithm to find the solution following support vector regression algorithm, obtain parameter With slack variable ξ j:
Find : ω i + , ω i - , ξ j , ω 0
Min : Σ i = 1 L ( ω i + + ω i - ) + 2 C Σ j = 1 L ξ j
s . t . t j - Σ i = 1 L ( ω i + - ω j - ) k ( x i , x j ) - ω 0 ≤ ϵ + ξ j Σ i = 1 L ( ω i + - ω j - ) k ( x i , x j ) + ω 0 - t j ≤ ϵ + ξ j ω i + ≥ 0 , ω i - ≥ 0 ξ j ≥ 0 , ( ∀ j = 1,2 , · · · , L )
3. try to achieve according to above-mentioned Unknown parameter ω in the formula computational data model below utilizing i:
ω i = ω i + - ω i -
4. the above-mentioned ω that obtains that finds the solution iAnd ω 0In the substitution formula, obtain interlayer microstrip antenna structure factor to little data model with the real part Re among the variation delta F of radiating element far field (Δ F (x)) or imaginary part Im (Δ F (x)) impact;
(2) if use associated vector regression algorithm solution procedure 5b6) in unknown parameter ω iAnd ω 0, then need to consider the impact of noise, namely step 5b6) in meta-model again be expressed as:
t(x)=y(x)+ε
y ( x ) = Σ i = 1 L ω i k ( x , x i ) + ω 0
In the formula, the prediction output of noise effect is not considered in y (x) expression, and ε represents output noise, and it is that 0 variance is σ that this noise satisfies average 2Normal distribution, the prediction output that t (x) expression is actual, it satisfies average is that y (x), variance are σ 2And separate normal distribution, its probability distribution is described as p (t i)=N (t iY (x i), σ 2), wherein, operator N () expression average is μ and variances sigma 2Normal distribution, it is defined as follows:
N ( t | μ , σ 2 ) = 1 2 π σ exp ( - ( t - μ ) 2 2 σ 2 )
According to normalized data sample subset Ω={ (x i, t i), i=1 ..., L}, and utilize the method for solving that has the Method Using Relevance Vector Machine algorithm now, obtain the unknown parameter ω in the meta-model i, ω 0And σ 2
5b8) according to plural number by the principle that real part and imaginary part form, make up real part and two meta-models of imaginary part correction of above-mentioned acquisition, obtain structural factor to the data model of radiating element far field variable quantity impact.
7. the comprehensive data-driven method for designing of active interlayer microstrip antenna structure as claimed in claim 1 and electromagnetism is characterized in that, calculates the impact of considering that panel and honeycomb gain on the interlayer microstrip antenna, and its computing formula is as follows:
S = Γ 01 - Γ 01 e - 2 j k 1 ( x 1 + x 4 ) + Γ 01 2 e - 2 j k o ( 2 x 2 + x 3 + H m ) - e - 2 j [ k 1 ( x 1 + x 4 ) - k o ( 2 x 2 + x 3 + H m ) ] 1 - Γ 01 2 e - 2 j k 1 ( x 1 + x 4 ) + Γ 01 e - 2 j k o ( 2 x 2 + x 3 + H m ) - Γ 01 e - 2 j [ k 1 ( x 1 + x 4 ) - k o ( 2 x 2 + x 3 + H m ) ]
In the formula, S represents to consider panel, honeycomb, bonding and coating thickness impact and the satisfied interlayer microstrip antenna reflection coefficient of opening under the condition, Wave-length constant in the expression panel, The part reflection coefficient of expression from the free space to the panel, The wave constant in the free space, ω fExpression free space angular frequency, ε 0And μ 0Represent respectively dielectric parameter and magnetic permeability in the free space, ε rAnd μ rRepresent respectively dielectric parameter and relative permeability in the panel, H mThe thickness of expression radio-frequency enabled layer.
8. the comprehensive data-driven method for designing of active interlayer microstrip antenna structure as claimed in claim 1 and electromagnetism is characterized in that, according to the interlayer microstrip antenna array electric field far field data E of above-mentioned calculating A(θ, φ) makes up interlayer microstrip antenna structure and electromagnetism Integrated Optimization Model to determine structural design variable x=[x 1, x 2, x 3, x 4, t c, l c] TWith little exciting current amplitude I with radiating element MnWith phase place Φ Mn:
Find:x,I mnmn
Min:||E A(θ,φ)-E *(θ,φ)||
s . t . S ≤ [ ϵ ] v max ≤ [ v ] σ max ≤ [ σ ] x l ≤ x ≤ x h
In the formula, E *The given expectation electric field far field of (θ, φ) expression, v MaxAnd σ MaxRepresent respectively maximum deformation quantity and maximum stress, they determine x by the Ansys analysis result in the 3rd step l, x hMinimum and the maximal value of expression structural design variable, [ε], [v] and [σ] represent respectively reflection coefficient, maximum deformation quantity and the maximum stress of permission, they are given by design objective, t c, l cThe length of side and the thickness of honeycomb wall that represent respectively the regular hexagon cellular unit.
9. the comprehensive data-driven method for designing of active interlayer microstrip antenna structure as claimed in claim 1 and electromagnetism is characterized in that, according to the antenna electric field far field data E that obtains A(θ, φ) determines minor level and beam position electrical performance indexes, and utilizes following formula to calculate the gain of interlayer microstrip antenna:
10. active interlayer microstrip antenna, it is characterized in that mainly be comprised of top panel, honeycomb, radio-frequency enabled layer and lower panel, wherein upper and lower panel and honeycomb belong to the encapsulation function layer, have mechanics carrying and heat insulation safeguard function, the plate surface also will apply camouflage coating in the above; The radio-frequency enabled layer mainly is comprised of feeding network, the beam signal controlled processing unit of micro-strip antenna array, integrated T/R assembly, merit parallel circuit and liquid cooling passage, mainly adopts low-temperature co-burning ceramic material; By using the integrated molding manufacturing process, the radio-frequency enabled layer of integrated microstrip antenna is embedded in the platform structure of panel and honeycomb composition, and then implementation structure and function solenoid is integrated.
CN201310127319.6A 2013-04-12 2013-04-12 Active interlayer microstrip antenna and the comprehensive data-driven method for designing of electromagnetism and antenna Active CN103353904B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201310127319.6A CN103353904B (en) 2013-04-12 2013-04-12 Active interlayer microstrip antenna and the comprehensive data-driven method for designing of electromagnetism and antenna

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201310127319.6A CN103353904B (en) 2013-04-12 2013-04-12 Active interlayer microstrip antenna and the comprehensive data-driven method for designing of electromagnetism and antenna

Publications (2)

Publication Number Publication Date
CN103353904A true CN103353904A (en) 2013-10-16
CN103353904B CN103353904B (en) 2016-04-27

Family

ID=49310276

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201310127319.6A Active CN103353904B (en) 2013-04-12 2013-04-12 Active interlayer microstrip antenna and the comprehensive data-driven method for designing of electromagnetism and antenna

Country Status (1)

Country Link
CN (1) CN103353904B (en)

Cited By (19)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103618130A (en) * 2013-10-18 2014-03-05 中国电子科技集团公司第十研究所 Embedded smart skin antenna
CN103887605A (en) * 2014-04-04 2014-06-25 西安电子科技大学 Wing antenna integrating structures and functions
CN104615836A (en) * 2015-02-12 2015-05-13 西安电子科技大学 Rapid prediction method for impact of spaceborne microstrip antenna array thermal deformation on electrical performance
CN104765921A (en) * 2015-04-03 2015-07-08 南昌航空大学 Design method for low-noise composite plate structure
CN104820780A (en) * 2015-05-05 2015-08-05 中国飞机强度研究所 Method for computing equivalent Poisson ratio of concave honeycomb structure
CN105426575A (en) * 2015-10-30 2016-03-23 西安电子科技大学 Method for inversion from electrical property of vertical interconnection structure to process parameter thereof
CN105701316A (en) * 2016-02-29 2016-06-22 中国飞机强度研究所 Temperature control point position selection method for test specimen under heat and force combined effect
CN105787160A (en) * 2016-02-17 2016-07-20 西安电子科技大学 Satellite-bone active phased-array antenna structure lightweight design method based on electromechanical coupling
CN104036093B (en) * 2014-06-26 2017-01-18 西安电子科技大学 Large deformation array antenna sidelobe performance prediction method based on array element mutual coupling
CN106354909A (en) * 2016-08-22 2017-01-25 西安电子科技大学 Method for determining structure tolerance of sparsely-configured antenna array based on radiation field and scattering field
CN106971042A (en) * 2017-03-31 2017-07-21 福州大学 A kind of method that legalizes for being used to mix the design of Height Standard element circuit
CN108173565A (en) * 2016-12-07 2018-06-15 中国移动通信集团终端有限公司 For determining the method, apparatus of communication device antenna working condition and communication equipment
CN108304646A (en) * 2018-01-30 2018-07-20 中国电子科技集团公司第三十八研究所 A kind of multidisciplined modeling and design method of multichannel chip TR components
CN108399274A (en) * 2018-01-16 2018-08-14 西安电子科技大学 A kind of comprehensive electromechanical analysis method of conformal bearer antenna
CN108959685A (en) * 2018-04-17 2018-12-07 中国科学院沈阳自动化研究所 A kind of equivalent modeling method of solar wing windsurfing
CN109684650A (en) * 2017-10-18 2019-04-26 大唐移动通信设备有限公司 A kind of method and apparatus of determining centre frequency
CN109714015A (en) * 2018-12-28 2019-05-03 电子科技大学 A kind of laminated low pass filter based on magnetic Jie's composite material
CN112994808A (en) * 2021-05-20 2021-06-18 成都天锐星通科技有限公司 Radio frequency signal internal calibration system and phased array antenna
US11303041B2 (en) 2019-07-24 2022-04-12 Cypress Semiconductor Corporation Machine learning-based antenna array validation, prototyping and optimization

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR20120121721A (en) * 2011-04-27 2012-11-06 포항공과대학교 산학협력단 Antenna structure
CN102998540B (en) * 2012-10-22 2015-01-07 西安电子科技大学 Forecasting method for influences, on electrical performances, of array surface morphology of conformal load-bearing microstrip antenna
CN102890741B (en) * 2012-10-25 2015-04-08 西安电子科技大学 Method for predicting influence of vibration deformation on electric performance of array antenna

Cited By (33)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103618130A (en) * 2013-10-18 2014-03-05 中国电子科技集团公司第十研究所 Embedded smart skin antenna
CN103618130B (en) * 2013-10-18 2015-08-26 中国电子科技集团公司第十研究所 Embedded intelligence skin antenna
CN103887605B (en) * 2014-04-04 2016-08-24 西安电子科技大学 Structure-function integration wing antenna
CN103887605A (en) * 2014-04-04 2014-06-25 西安电子科技大学 Wing antenna integrating structures and functions
CN104036093B (en) * 2014-06-26 2017-01-18 西安电子科技大学 Large deformation array antenna sidelobe performance prediction method based on array element mutual coupling
CN104615836B (en) * 2015-02-12 2017-06-06 西安电子科技大学 The method for quick predicting that the thermal deformation of Test of Space Micro-Strip Array array antenna influences on electrical property
CN104615836A (en) * 2015-02-12 2015-05-13 西安电子科技大学 Rapid prediction method for impact of spaceborne microstrip antenna array thermal deformation on electrical performance
CN104765921A (en) * 2015-04-03 2015-07-08 南昌航空大学 Design method for low-noise composite plate structure
CN104765921B (en) * 2015-04-03 2017-07-28 南昌航空大学 A kind of low noise composite panel construction design method
CN104820780B (en) * 2015-05-05 2018-02-02 中国飞机强度研究所 A kind of method for the equivalent Poisson's ratio for calculating inner concave shape honeycomb
CN104820780A (en) * 2015-05-05 2015-08-05 中国飞机强度研究所 Method for computing equivalent Poisson ratio of concave honeycomb structure
CN105426575A (en) * 2015-10-30 2016-03-23 西安电子科技大学 Method for inversion from electrical property of vertical interconnection structure to process parameter thereof
CN105426575B (en) * 2015-10-30 2018-10-12 西安电子科技大学 By vertical interconnecting structure electrical property to its working process parameter inversion method
CN105787160A (en) * 2016-02-17 2016-07-20 西安电子科技大学 Satellite-bone active phased-array antenna structure lightweight design method based on electromechanical coupling
CN105787160B (en) * 2016-02-17 2018-10-09 西安电子科技大学 The structure lightened design method of spaceborne active phase array antenna based on mechanical-electric coupling
CN105701316A (en) * 2016-02-29 2016-06-22 中国飞机强度研究所 Temperature control point position selection method for test specimen under heat and force combined effect
CN105701316B (en) * 2016-02-29 2018-10-09 中国飞机强度研究所 A kind of testpieces temperature control point position selecting method under power heat integration effect
CN106354909A (en) * 2016-08-22 2017-01-25 西安电子科技大学 Method for determining structure tolerance of sparsely-configured antenna array based on radiation field and scattering field
CN106354909B (en) * 2016-08-22 2019-02-19 西安电子科技大学 Sparse arrangement array antenna structure public affairs method for determining difference towards radiation and scattering
CN108173565B (en) * 2016-12-07 2019-10-11 中国移动通信集团终端有限公司 For determining the method, apparatus and communication equipment of communication device antenna working condition
CN108173565A (en) * 2016-12-07 2018-06-15 中国移动通信集团终端有限公司 For determining the method, apparatus of communication device antenna working condition and communication equipment
CN106971042B (en) * 2017-03-31 2019-10-15 福州大学 A kind of method that legalizes designed for mixing Height Standard element circuit
CN106971042A (en) * 2017-03-31 2017-07-21 福州大学 A kind of method that legalizes for being used to mix the design of Height Standard element circuit
CN109684650A (en) * 2017-10-18 2019-04-26 大唐移动通信设备有限公司 A kind of method and apparatus of determining centre frequency
CN108399274A (en) * 2018-01-16 2018-08-14 西安电子科技大学 A kind of comprehensive electromechanical analysis method of conformal bearer antenna
CN108399274B (en) * 2018-01-16 2021-09-03 西安电子科技大学 Electromechanical comprehensive analysis method for conformal bearing antenna
CN108304646A (en) * 2018-01-30 2018-07-20 中国电子科技集团公司第三十八研究所 A kind of multidisciplined modeling and design method of multichannel chip TR components
CN108304646B (en) * 2018-01-30 2021-07-23 中国电子科技集团公司第三十八研究所 Multidisciplinary modeling and design method of multichannel chip type TR (transmitter-receiver) component
CN108959685A (en) * 2018-04-17 2018-12-07 中国科学院沈阳自动化研究所 A kind of equivalent modeling method of solar wing windsurfing
CN109714015A (en) * 2018-12-28 2019-05-03 电子科技大学 A kind of laminated low pass filter based on magnetic Jie's composite material
CN109714015B (en) * 2018-12-28 2021-10-26 电子科技大学 Laminated low-pass filter based on magnetic dielectric composite material
US11303041B2 (en) 2019-07-24 2022-04-12 Cypress Semiconductor Corporation Machine learning-based antenna array validation, prototyping and optimization
CN112994808A (en) * 2021-05-20 2021-06-18 成都天锐星通科技有限公司 Radio frequency signal internal calibration system and phased array antenna

Also Published As

Publication number Publication date
CN103353904B (en) 2016-04-27

Similar Documents

Publication Publication Date Title
CN103353904B (en) Active interlayer microstrip antenna and the comprehensive data-driven method for designing of electromagnetism and antenna
CN101344564B (en) Active phase array antenna electrical property prediction method based on mechanical, electric and thermal three-field coupling
CN104615836B (en) The method for quick predicting that the thermal deformation of Test of Space Micro-Strip Array array antenna influences on electrical property
CN102998540B (en) Forecasting method for influences, on electrical performances, of array surface morphology of conformal load-bearing microstrip antenna
CN106940739B (en) Method for rapidly predicting influence of vibration on electrical performance of wing conformal phased array antenna
CN105701297B (en) A kind of reflector antenna mechanical-electric coupling design method based on multiple spot Adaptive proxy model
CN105302962B (en) Reflector antenna electromechanics Integrated Optimal Design method based on structure-electromagnetism mixed cell
CN103246781A (en) Array antenna radar cross section reduction method based on space mapping
CN106252873B (en) A kind of Novel Interval Methods of conformal bearer Antenna Far Field power radiation pattern
CN111276822B (en) Antenna pattern main lobe controllable antenna array design method
CN104733850A (en) Holographic modulation based artificial tensor impedance surfaced antenna and implementation method thereof
Zhou et al. Electromechanical co-design and experiment of structurally integrated antenna
CN104933213A (en) Large-scale phased antenna array wide-angle scanning optimization method based on space mapping
CN109670140A (en) A kind of array antenna Electrical Analysis method based on array element feature mode
Zhou et al. Performance of structurally integrated antennas subjected to dynamical loads
CN104063533B (en) A kind of wing cover antenna electric coupling analysis method
CN107357962A (en) A kind of antenna house rib cross-sectional size optimization method based on Adaptive proxy model
CN104732050A (en) Method for estimating distribution of electromagnetism in flying targets made of carbon fiber materials under lightning pulses
CN106934097B (en) Electrical performance-oriented key dynamic mode selection method for spatial mesh antenna
CN106599504B (en) Space mesh antenna dynamic load analysis method based on electromechanical coupling model
CN108446504A (en) Near-field array Antenna measuring table method based on convex optimization
CN103336862A (en) Support vector machine-based structure design method for sandwich microstrip antenna
Li et al. Optimization design of skin antenna based on Bayesian optimization
Li et al. A novel inversion method of manufacturing flaws in the packaging of conformal load-bearing antenna structure
Zhou et al. Design, fabrication, and testing of active skin antenna with 3D printing array framework

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
C14 Grant of patent or utility model
GR01 Patent grant