【摘要】：現(xiàn)代飛機結構的輕量化需求使得機翼柔度變大,靜氣彈效應愈加顯著。傳統(tǒng)工程結構設計優(yōu)化往往只考慮結構本身的強度剛度性能,并未直接考慮靜氣彈效應帶來的諸如飛機升力效率、副翼效率、焦點位置變化等影響,這使得無法獲得滿足不同學科需求的最佳結構。隨著現(xiàn)代飛機結構更精細化設計以及更深層次挖掘結構潛能的需求,結構設計變量的數(shù)量和各學科模型的規(guī)模越來越大,結構靜氣彈多學科優(yōu)化方法在處理這種大規(guī)模優(yōu)化問題時面臨著求解效率低、計算代價大以及導數(shù)信息獲取困難等挑戰(zhàn),這限制了飛機結構的進一步精細優(yōu)化設計。本文旨在通過改進亞聲速工程面元法求解精度,提高靜氣彈求解效率以及構造靜氣彈設計敏度算法等,與大規(guī)模結構優(yōu)化方法結合,力圖解決大規(guī)模結構靜氣彈多學科優(yōu)化面臨的技術難點,主要研究工作如下:為著實現(xiàn)機翼彈性氣動載荷的高效率與高精度計算,針對高精度計算流體力學(CFD)方法計算耗時過長的矛盾,本文采用工程處理觀點,提出了一種改進的亞聲速工程面元法—分段精細修正面元法。技術途徑為:采用多個迎角下的剛性機翼高精度CFD氣動力數(shù)據(jù),進行工程面元法的分段修正,獲取多段修正因子矩陣;同時,將機翼彈性變形的下洗分段,利用所獲取的修正因子矩陣提高機翼彈性氣動載荷的計算精度與效率。為進一步提高分段精細修正面元法計算精度,本文提出了一種面元網(wǎng)格劃分優(yōu)化算法,該算法以機翼面元展向和弦向劃分數(shù)目為優(yōu)化變量,以靜氣彈計算中機翼在一較大彈性變形下的高精度CFD氣動力數(shù)據(jù)為基礎,在ISGHIT軟件平臺上對面元網(wǎng)格實現(xiàn)最優(yōu)劃分,使得最優(yōu)面元網(wǎng)格劃分下的修正面元法彈性氣動載荷計算結果與高精度CFD結果更為接近。為提高氣動與結構耦合界面的數(shù)據(jù)傳遞精度與效率,本文采用數(shù)值精度高、適應性好的基于徑向基函數(shù)(RBF)數(shù)據(jù)傳遞方法,并對其緊支半徑以及傳遞節(jié)點及其數(shù)量提出了選擇規(guī)則,提高了RBF方法的計算效率。工程中的靜氣彈性能往往采用簡單定義,不能完全反映靜氣彈效應給飛機氣動效率及操穩(wěn)特性帶來的影響。本文采用靜氣彈性能的精確定義并利用復步長求導方法,結合分段精細修正面元法構造了求解升力效率、副翼效率、焦點弦向位置變化率的高效算法,完成了算法程序設計。其中,提出一種雙重迭代計算副翼效率的方法,不但能準確求解副翼效率還能計算飛機定常滾轉速率,充分反映了機翼彈性和副翼偏角給飛機滾轉機動性能帶來的影響。為了給予設計者更多的參考信息,本文還提出了彈性升力迎角補償算法以及彈性滾轉速率副翼偏角補償算法,以求得定載與定速滾轉情況下的迎角補償量與副翼偏角補償量;同時,設計了機翼發(fā)散速度與反效速度的低階估算算法�；谔荻刃畔㈩惖膬�(yōu)化算法是大規(guī)模結構數(shù)值優(yōu)化常用的一種高效算法,為了給予這類優(yōu)化方法導數(shù)信息支持,本文利用所提出的分段精細修正面元法構造了靜氣彈設計敏度半解析算法,并采取多項措施提高其計算效率。為適應大規(guī)模結構靜氣彈多學科優(yōu)化程序的模塊化組織,本文編寫了求解高效、讀寫規(guī)范的靜氣彈求解程序模塊,并通過OPENMP并行技術進一步提高了程序計算效率。通過M6機翼靜氣彈性能算例考察,并與高精度CFD數(shù)據(jù)、NASTRAN軟件計算結果比較,表明本文提出的多項靜氣彈性能求解算法以及程序設計具有精度高、效率好的技術優(yōu)勢。最后,綜合分析了結構靜氣彈多學科優(yōu)化的原理與特點,并利用.MASS文件實現(xiàn)不同工況下不同集中質(zhì)量加載。文中總結了工程中常用的結構設計約束,闡述了本課題組開發(fā)的大規(guī)模結構優(yōu)化程序中的數(shù)值優(yōu)化算法以及約束篩選、變量降維、文件組織、并行處理等核心技術。在該程序基礎上,本文完成了靜氣彈性能約束的并入集成,形成了大規(guī)模結構靜氣彈多學科優(yōu)化程序。在此工作基礎上,本文對一飛翼無人機結構采用699個設計變量,施加4種強度設計工況、2種飛行工況以及近10種約束,進行了結構靜氣彈多學科數(shù)值優(yōu)化計算,并對優(yōu)化結果進行了分析校驗。結果表明本文采用的大規(guī)模結構靜氣彈多學科優(yōu)化方法計算高效,并減少結構重量15.66%,優(yōu)化結果滿足工程約束。另外,通過5種不同的約束組合進行結構多學科優(yōu)化結果比較,分析了各約束對結構重量的影響,以助于發(fā)掘結構設計規(guī)律。
[Abstract]:The light-weight demand of modern aircraft structure makes the wing soft degree become large, and the static-gas bomb effect is more significant. The optimization of the traditional structural design often only takes into account the strength and rigidity performance of the structure itself, and does not directly take into account the influence of the static-gas effect, such as the lift efficiency of the airplane, the aileron efficiency, the change of the focus position, and the like, which makes it impossible to obtain the best structure to meet the needs of different disciplines. With the more refined design of the modern aircraft structure and the need of the deeper excavation of the structural potential, the number of structural design variables and the scale of each subject model are becoming larger and larger, and the multi-disciplinary optimization method of the structure static gas bomb is faced with low solution efficiency in the process of processing the large-scale optimization problem, The computational cost and the difficulty of obtaining the derivative information have limited the further fine-tuning design of the aircraft structure. The purpose of this paper is to solve the technical difficulties faced by the multi-disciplinary optimization of large-scale structure static-gas bomb by improving the accuracy of the sub-sonic engineering surface element method, improving the solution efficiency of the static-gas bomb and the design of a static-gas bomb, and combining with the large-scale structure optimization method. The main research work is as follows: In order to realize the high efficiency and high precision calculation of the elastic pneumatic load of the wing, the problem that the time consuming is too long is calculated for the high-precision computational fluid dynamics (CFD) method. In this paper, an improved sub-segment fine correction surface element method for subsonic engineering is presented. The technical method comprises the following steps of: adopting high-precision CFD aerodynamic data of a rigid wing at a plurality of angles of attack, performing section correction of the engineering surface element method, acquiring a multi-section correction factor matrix, and simultaneously, and the calculation accuracy and the efficiency of the wing elastic pneumatic load are improved by utilizing the acquired correction factor matrix. in ord to further improve that calculation accuracy of the segment fine correction surface element method, a surface element mesh partition optimization algorithm is proposed, On the basis of the high-precision CFD aerodynamic data of the wing under a large elastic deformation, the optimal division of the meta-grid is realized on the ISGHIT software platform, so that the calculation of the elastic aerodynamic load of the modified plane element method under the optimal plane element mesh is more close to the high-precision CFD result. In order to improve the data transfer precision and efficiency of the coupling interface between the pneumatic and the structure, this paper adopts the radial basis function (RBF) data transmission method with high numerical precision and good adaptability, and puts forward the selection rules for the compact radius and the number of the transmission nodes and the transmission nodes. and the calculation efficiency of the RBF method is improved. The static and static elastic energy in the project often adopts a simple definition, and can not completely reflect the influence of the static-gas elastic effect on the aerodynamic efficiency and the operation stability of the aircraft. In this paper, the precise definition of the static and elastic energy is used and the complex step method is used, and a high-efficiency algorithm for solving the change of the lift efficiency, the aileron efficiency and the position change rate of the focal chord is constructed in combination with the segmented fine correction surface element method, and the algorithm programming is completed. In this paper, a double-iteration method for calculating the aileron efficiency is proposed, which not only can accurately solve the aileron efficiency but also can calculate the steady roll rate of the airplane, and fully reflects the influence of the wing elasticity and the aileron deflection angle to the rolling mobility of the airplane. In order to give the designer more reference information, an elastic lift angle-of-attack compensation algorithm and an elastic roll-rate aileron-off-angle compensation algorithm are presented in this paper to determine the angle-of-attack compensation and the amount of the aileron-off-angle compensation in the case of fixed-load and constant-speed rolling, and at the same time, A low-order estimation algorithm for the speed of divergence and the velocity of the wing is designed. The optimization algorithm based on the gradient information class is a kind of high-efficiency algorithm used in large-scale structure numerical optimization. In order to support the derivative information of this kind of optimization method, this paper constructs a half-resolution algorithm for the design of static-gas bomb by using the proposed segmentation fine correction surface element method. and a plurality of measures are taken to improve the calculation efficiency. In order to adapt to the modular organization of the multi-subject optimization program of large-scale static-gas bomb, this paper has developed a static-gas bomb solution program module for high-efficiency and read-write specifications, and further improves the program calculation efficiency through OPENNMP parallel technology. Through the study of the static and elastic energy of the M6 wing, and compared with the high-precision CFD data and the NASTRAN software, it is shown that the multiple static-gas elastic energy-solving algorithm and the programming have the advantages of high precision and high efficiency. In the end, the principle and characteristics of the multi-disciplinary optimization of the structure static-gas bomb are comprehensively analyzed and used. The mass loading of different concentration under different working conditions is realized by the MASS file. In this paper, the structure design constraints commonly used in the project are summarized, the numerical optimization algorithm in the large-scale structure optimization program developed by the research group, and the core technologies such as constraint selection, variable drop-down, file organization and parallel processing are described. On the basis of this program, this paper completes the integration of the static and gas elastic energy constraints, and forms a large-scale structure static-gas multi-disciplinary optimization program. On the basis of this work, 699 design variables are used in the structure of a flying-wing unmanned aerial vehicle, four strength design conditions, two flight conditions and nearly 10 constraints are applied, and the multi-subject numerical optimization calculation of the structure static-gas bomb is carried out, and the optimization result is analyzed and verified. The results show that the large-scale structure static-gas multi-disciplinary optimization method is efficient and reduces the structural weight by 15.66%, and the optimization results meet the engineering constraints. In addition, the influence of each constraint on the weight of the structure is analyzed through the comparison of five different constraint combinations to the structural multi-subject optimization results, so as to help to find out the law of the structural design.
【學位授予單位】：西北工業(yè)大學
【學位級別】：博士
【學位授予年份】：2016
【分類號】：V279
，

本文編號：2380195