發(fā)布時間：2018-05-14 20:37

  本文選題：衛(wèi)星艙 + 布局優(yōu)化��； 參考：《南京信息工程大學(xué)》2016年碩士論文

【摘要】：衛(wèi)星艙布局優(yōu)化問題是一類NP困難問題。它不僅是一類特殊的耦合系統(tǒng)設(shè)計問題,也是一類復(fù)雜的組合優(yōu)化問題。本文以國際商業(yè)通信衛(wèi)星艙為背景,具體研究如何將衛(wèi)星艙組件(有效載荷)分配到衛(wèi)星艙承載板上,并優(yōu)化組件在承載板上的布局。采用Wang-Landau (WL)抽樣算法作為全局優(yōu)化算法,結(jié)合局部搜索和啟發(fā)式策略構(gòu)建了一種混合算法。具體研究內(nèi)容及成果如下：(1)研究已分配組件到基面后的多目標(biāo)衛(wèi)星艙布局優(yōu)化問題。針對該問題,采用Wang-Landau (WL)抽樣算法進行求解。為了加快全局最優(yōu)解的搜索速度,在每次蒙特卡羅掃描生成新的布局時便執(zhí)行基于梯度法的局部搜索(LS)。通過結(jié)合WL抽樣算法,LS和啟發(fā)式布局更新策略,構(gòu)建了一種新的混合布局算法WL-LS。此外,為了改進算法的搜索效率,提出了一種簡單高效的干涉量計算方法,將三維模型轉(zhuǎn)化成二維模型,即長方體簡化為矩形,圓柱體簡化為圓形。在此布局問題中,矩形待布物正交放置。對兩個算例進行了測試。計算結(jié)果顯示,該算法是一種求解多目標(biāo)衛(wèi)星艙布局優(yōu)化問題的有效算法。(2)研究三維簡化衛(wèi)星艙組件分配與布局優(yōu)化問題。針對衛(wèi)星艙組件在承載板之間的分配問題,采用神經(jīng)網(wǎng)絡(luò)算法求解得到一種近似最優(yōu)分配方案。針對三維簡化衛(wèi)星艙組件布局優(yōu)化問題,首先將組件簡化為長方體待布物和圓柱體待布物。在計算干涉量時,又將其簡化為矩形待布物和圓形待布物,其中矩形待布物任意放置。提出了改進的不適合多邊形法,即通過分析矩形或圓形中心和不適合邊界之間的幾何關(guān)系,進行干涉性判斷和干涉量的計算。最后,在使用神經(jīng)網(wǎng)絡(luò)算法得到衛(wèi)星艙組件的分配方案后,運行WL-LS算法對兩個算例進行了測試。計算結(jié)果表明,所提出的算法是解決三維簡化衛(wèi)星艙組件分配與布局優(yōu)化問題的有效算法。
[Abstract]:The optimization of satellite cabin layout is a kind of NP-hard problem. It is not only a special coupled system design problem, but also a complex combinatorial optimization problem. Taking the international commercial communication satellite module as the background, this paper studies how to distribute the satellite module (payload) to the bearing board of the satellite module and optimize the layout of the module on the bearing board. A hybrid algorithm is constructed by using Wang-Landau sampling algorithm as global optimization algorithm and combining local search and heuristic strategy. The main contents and results are as follows: (1) the optimization of multi-target satellite module layout after the allocation of components to the base plane is studied. To solve this problem, Wang-Landau sampling algorithm is used to solve the problem. In order to speed up the search speed of the global optimal solution, the local search based on gradient method is performed every time Monte-Carlo scan generates a new layout. A new hybrid layout algorithm, WL-LSs, is constructed by combining WL sampling algorithm with heuristic layout update strategy. In addition, in order to improve the search efficiency of the algorithm, a simple and efficient method for calculating interferometry is proposed. The three-dimensional model is transformed into a two-dimensional model, that is, the cuboid is reduced to a rectangle and the cylinder to a circular one. In this layout problem, the rectangular object to be placed is orthogonal. Two examples are tested. The results show that the algorithm is an effective algorithm to solve the multi-objective satellite cabin layout optimization problem. In order to solve the allocation problem of satellite module between bearing plates, an approximate optimal allocation scheme is obtained by using neural network algorithm. Aiming at the optimization of 3D simplified satellite module layout, the module is first simplified as cuboid to be distributed and cylinder to be distributed. In the calculation of interference, it is reduced to rectangular and circular objects, in which rectangular objects are placed arbitrarily. An improved unsuitable polygon method is proposed in this paper. By analyzing the geometric relationship between the rectangular or circular center and the unsuitable boundary, the interference judgment and the calculation of the interference quantity are carried out. Finally, after the allocation scheme of satellite module is obtained by using neural network algorithm, two examples are tested by running WL-LS algorithm. The results show that the proposed algorithm is an effective algorithm to solve the problem of 3D simplified satellite module allocation and layout optimization.
【學(xué)位授予單位】：南京信息工程大學(xué)
【學(xué)位級別】：碩士
【學(xué)位授予年份】：2016
【分類號】：V423.4;TP301.6
，

本文編號：1889375