基于计算流体力学的太阳风研究与仿真
|
摘要
随着空间信息技术、大气物理学等学科的迅速发展,人类对宇宙的探索不断深入,宇宙中的各种潜在的对人类危害也逐渐显露。太阳风就是其中一种人类了解甚少而危害巨大的宇宙现象,它能干扰人类无线通信、引起地磁暴、破坏人类空间站等。我国嫦娥探月工程,带回了宇宙空间中的大量太阳风数据与信息,这些信息将为我国研究太阳风与空间信息技术带来巨大的助益。如何利用这些数据建立太阳风系统的数理模型并进行系统仿真,成为当前研究的重大课题并具有十分重要的科学意义和应用价值。
建立太阳风仿真系统的关键技术在于对太阳风数据的研究,总结太阳风的物理特性,结合磁流体力学、天体物理学等相关理论,建立真实可靠的太阳风运动模型,并使用计算机技术进行计算与仿真。目前,对太阳风数据的研究取得初步进展,但是建立太阳风的准确数理模型攻克许多难关,主要问题是太阳风本身的规模巨大、行程遥远、影响因素众多,准确数据仍难以取得。任何科学在研究之初都可以通过简化相关数据来进行基础研究,建立理想条件下的太阳风运动数理模型并应用离散数学方法在计算机上求解,是当前太阳风研究的主要课题,其中各项关键技术与相关学科的研究也势在必行。
本文针对建立太阳风初步数理模型并在计算机上进行仿真,提出应用当前相对成熟的计算流体动力学相关理论,建立瞬态太阳风流体数理模型,通过工程上常用同位网格下的SIMPLE(Semi-Implicit Method for Pressure-Linked Equations)算法进行Navier-Stokes方程与连续方程的求解,并运用图形学相关理论知识、以OGRE图形引擎为平台、运用GPU渲染技术进行计算机渲染与实现。该方法虽然不能建立完全正确的太阳风数理模型,也不能获得真实的仿真结果,但是它提出了一种太阳风初步研究和简化仿真的方法,为以后更进一步研究太阳风系统有重大的指导意义。论文主要研究工作和创新性工作如下:
(1)研究了太阳风的流体性质。
(2)研究了计算流体力学的相关理论基础以及SIMPLE系列算法。
(3)提出了一种基于计算流体动力学的太阳风数理模型的简化方法。
(4)深入研究OGRE图形引擎、图形学知识以及GPU编程。
(5)提出应用CPU计算太阳风数理模型协同GPU进行渲染的方式来加速仿真效率的方法。
With fast developed of the spatial information technology, atmospheric physics and other subjects, human’s exploration of the universe are deeper and deeper, a variety of potential hazards to human beings are gradually emerged in the universe. Solar wind is one of the great harm phenomena of humans but little is known in the universe, it can interfere with human communication, causing geomagnetic storms, the destruction of human space station and so on. China's ChangE lunar exploration program, which brought back a lot of space data and information about the solar wind that will brought great benefit to human’s spatial technology. How to use these data to establish the solar wind systems and build the mathematical simulation model is become a major topic of current research and has important scientific significance and application value.
The key technology of establishing the solar wind simulation system is the research of the solar wind data and summary of the solar wind's physical characteristics. Combined with magnetic fluid mechanics and astrophysics and other related theory, establish and simulate the real and reliable solar wind motion model is viable. Currently, research on solar wind data made initial progress, but to establish the accuracy of the solar wind mathematical model should overcome many difficulties, the main problem is the scale of the solar wind itself, travel far away and lots of affecting factors. The beginnings of study any science are available through simplified data to conduct basic research. The establishment of ideal conditions of the solar wind movement mathematical model and the solution on computer is the main topics in current solar wind research. The research of the key technologies and related disciplines is imperative.
This paper established the solar wind preliminary mathematical model and computer simulation, proposes the application of the current relatively mature computational fluid dynamics-related theory, the transient solar wind body mathematical model, through engineering often collocated grid under the SIMPLE (Semi-Implicit Method for Pressure-Linked Equations) algorithm for Navier-Stokes equations and continuity equation solving, and use of graphics related theoretical knowledge to OGRE graphics engine as the platform, the use of GPU rendering technology, computer rendering and implementation. Although this method can’t establish entirely correct mathematical model of the solar wind, and could not obtain a real simulation results, but it presents a preliminary study on the solar wind and simplified simulation method to further study the solar wind for the future system of great guiding significance. Thesis of innovative research results and contributions are as follows:
(1) Do the research of solar wind fluid properties.
(2) Studied the basis theory of computational fluid dynamics and the SIMPLE series algorithm.
(3) Proposed based on computational fluid dynamics model of the solar wind simplified mathematical methods.
(4) Do the deep research of OGRE graphics engine, Graphic theory and GPU programming knowledge.
(5) Proposes the application of CPU computing mathematical model of the solar wind and GPU rendering collaborative way to accelerate the simulation efficiency.
建立太阳风仿真系统的关键技术在于对太阳风数据的研究,总结太阳风的物理特性,结合磁流体力学、天体物理学等相关理论,建立真实可靠的太阳风运动模型,并使用计算机技术进行计算与仿真。目前,对太阳风数据的研究取得初步进展,但是建立太阳风的准确数理模型攻克许多难关,主要问题是太阳风本身的规模巨大、行程遥远、影响因素众多,准确数据仍难以取得。任何科学在研究之初都可以通过简化相关数据来进行基础研究,建立理想条件下的太阳风运动数理模型并应用离散数学方法在计算机上求解,是当前太阳风研究的主要课题,其中各项关键技术与相关学科的研究也势在必行。
本文针对建立太阳风初步数理模型并在计算机上进行仿真,提出应用当前相对成熟的计算流体动力学相关理论,建立瞬态太阳风流体数理模型,通过工程上常用同位网格下的SIMPLE(Semi-Implicit Method for Pressure-Linked Equations)算法进行Navier-Stokes方程与连续方程的求解,并运用图形学相关理论知识、以OGRE图形引擎为平台、运用GPU渲染技术进行计算机渲染与实现。该方法虽然不能建立完全正确的太阳风数理模型,也不能获得真实的仿真结果,但是它提出了一种太阳风初步研究和简化仿真的方法,为以后更进一步研究太阳风系统有重大的指导意义。论文主要研究工作和创新性工作如下:
(1)研究了太阳风的流体性质。
(2)研究了计算流体力学的相关理论基础以及SIMPLE系列算法。
(3)提出了一种基于计算流体动力学的太阳风数理模型的简化方法。
(4)深入研究OGRE图形引擎、图形学知识以及GPU编程。
(5)提出应用CPU计算太阳风数理模型协同GPU进行渲染的方式来加速仿真效率的方法。
With fast developed of the spatial information technology, atmospheric physics and other subjects, human’s exploration of the universe are deeper and deeper, a variety of potential hazards to human beings are gradually emerged in the universe. Solar wind is one of the great harm phenomena of humans but little is known in the universe, it can interfere with human communication, causing geomagnetic storms, the destruction of human space station and so on. China's ChangE lunar exploration program, which brought back a lot of space data and information about the solar wind that will brought great benefit to human’s spatial technology. How to use these data to establish the solar wind systems and build the mathematical simulation model is become a major topic of current research and has important scientific significance and application value.
The key technology of establishing the solar wind simulation system is the research of the solar wind data and summary of the solar wind's physical characteristics. Combined with magnetic fluid mechanics and astrophysics and other related theory, establish and simulate the real and reliable solar wind motion model is viable. Currently, research on solar wind data made initial progress, but to establish the accuracy of the solar wind mathematical model should overcome many difficulties, the main problem is the scale of the solar wind itself, travel far away and lots of affecting factors. The beginnings of study any science are available through simplified data to conduct basic research. The establishment of ideal conditions of the solar wind movement mathematical model and the solution on computer is the main topics in current solar wind research. The research of the key technologies and related disciplines is imperative.
This paper established the solar wind preliminary mathematical model and computer simulation, proposes the application of the current relatively mature computational fluid dynamics-related theory, the transient solar wind body mathematical model, through engineering often collocated grid under the SIMPLE (Semi-Implicit Method for Pressure-Linked Equations) algorithm for Navier-Stokes equations and continuity equation solving, and use of graphics related theoretical knowledge to OGRE graphics engine as the platform, the use of GPU rendering technology, computer rendering and implementation. Although this method can’t establish entirely correct mathematical model of the solar wind, and could not obtain a real simulation results, but it presents a preliminary study on the solar wind and simplified simulation method to further study the solar wind for the future system of great guiding significance. Thesis of innovative research results and contributions are as follows:
(1) Do the research of solar wind fluid properties.
(2) Studied the basis theory of computational fluid dynamics and the SIMPLE series algorithm.
(3) Proposed based on computational fluid dynamics model of the solar wind simplified mathematical methods.
(4) Do the deep research of OGRE graphics engine, Graphic theory and GPU programming knowledge.
(5) Proposes the application of CPU computing mathematical model of the solar wind and GPU rendering collaborative way to accelerate the simulation efficiency.
引文
[1]张兴波,章公亮,祁贵仲.太阳风速度的周期结构分析[J].空间科学学报,1986,6(4):252-257.
[2]龚文,王庆丰,冯强.基于DirectX的三维实时显示平台的实现[J].应用研究,2010,38(15):61-64.
[3]吴其芬,李桦著.磁流体动力学[M].北京:国防科技大学出版社,2007.
[4]邸锐.OGRE 3D游戏开发框架指南[M].北京:电子工业出版社,2010.
[5]黄静.太阳耀斑非热电子动力学研究[J].天文学进展,2010,28(3):213-228.
[6]邓洪涛,地球内核与磁场[J].武汉大学学报,2010,35(7):854-856.
[7]王福军.计算流体动力学分析——CFD软件原理与应用[M] .北京:清华大学出版社. 2004.
[8]佟桂芳,徐德龙,张强等.一种新改进的SIMPLE算法[J].西安建筑科技大学学报,2001,9,33(3):221~224.
[9]吴子牛.计算流体动力学基本原理[M].北京:科学出版社,2001.
[10]尚晶晶.Direct3D游戏开发技术详解[M].北京:人民邮电出版社,2006.
[11] (美)Hubert Nguyen著.杨柏林,陈根浪,王聪译.GPU精粹3[M].北京:清华大学出版社,2010.
[12] (美)Donald Hearn, M. Pauline Baker著.蔡士杰,宋继强,蔡敏译.计算机图形学[M].北京:电子工业出版社,2005.
[13]叶至军.C++ STL开发技术导引[M].北京:人民邮电大学,2007.
[14] (美)Michael Dickheiser著.孟宪武译.游戏编程精粹6[M].北京:人民邮电出版社,2007.
[15] (美)Steve McConnell著.金戈,汤凌,陈硕,张菲.代码大全第二版[M].北京:电子工业出版社,2006.
[16] (美)Mike McShaffry著.冯兆麟,孔祥一,李华杰译.游戏编程全接触[M].北京:人民邮电出版社,2006.
[17] (美)Scott Meyers著.侯捷译.Effective C++第三版[M].北京:电子工业出版社,2006.
[18] (美)Sartaj Sahni著.汪诗林,孙晓东译.数据结构算法与应用[M].北京:机械工业出版社,2000.
[19] (美)Scott Meyers著.Effective STL[M].北京:中国电力出版社,2003.
[20]王鹏杰,李威,王聪著.DirectX游戏编程[M].北京:机械工业出版社,2010.
[21] Enright Douglas, Marschner Stephen, Fedkiw Ronald. Animation and rendering of complex water surfaces[J].ACM Transactions on Graphicx, 2002, 21(3):736~744.
[22] Hong Jeong-Mo,Kim Chang-Hun.Animation of bubbles in liquid[J].Computer Graphics Forum, 2003, 22(3):253~262.
[23] Lamorlette Arnauld, Foster Nick. Structural modeling of flames for a production environment[J]. ACM Transactions on Graphics,2002, 21(3):729~735.
[24] Peachey Darwyn R. Modeling waves and surf[J].Computer Graphics, 1986, 20(4):65~74.
[25] Fournier Alain, Reeves William T. A Simple model of ocean waves[J]. Computer Graphics, 1986, 20(4):75~84.
[26] Kass Michael, Miller Gavin. Rapid, stable fluid dynamics for computer graphics[J]. Computer Graphics, 1990, 24(4):49~57.
[27] Foster Nick, Metaxas Dimitri. Realistic animation of liquids[J]. Graphical Models and Image Processing, 1996, 58(5):471~483.
[28] Stam Jos, Fiume Eugene. Dipicting fire and other gaseous phenomena using diffusion processes[A]. Computer Graphics Proceedings. 1995, 129~136.
[29] McNamara Antoine, Treuille Adrien, Popovic Zoran, etc. Fluid control using the adjoint method[J].ACM Transactions on Graphics, 2004, 23(3):449~456.
[30]邓宝庆,吴文权.同位网格上不可压流动的反欠松弛压力修正算法[J].工程热物理学报,2002,9,23(5):605~607.
[31]宋道云,刘洪来,方波等.动量差值与完全压力校正算法及交错网格SIMPLE算法的比较[J].华东理工大学学报,2003,4,29(2):124~143.
[32]冯天详.一个高斯-赛德尔方法解方程组的有趣结果[J].重庆三峡学院学报,2010,3,26(3):14~17.
[33]龚敏敏.GPU精粹2[M].北京:清华大学出版社,2007.5.
[34] Srivastava. S.P.Method of Interpretation of Magnetotelluric Data When Source Field Is Considered[J]. Journal of Geophysical Research,1965,4,70(4): 945-954.
[35]王水,李罗权.磁场重连[M].合肥:安徽教育出版社,1998.
[36] Furth PH, Killeen J, Rosenbluth MN. Finite-resistivity insablities of a sheet pinch. Phys Fliuds, 1963,6(4):459~484.
[2]龚文,王庆丰,冯强.基于DirectX的三维实时显示平台的实现[J].应用研究,2010,38(15):61-64.
[3]吴其芬,李桦著.磁流体动力学[M].北京:国防科技大学出版社,2007.
[4]邸锐.OGRE 3D游戏开发框架指南[M].北京:电子工业出版社,2010.
[5]黄静.太阳耀斑非热电子动力学研究[J].天文学进展,2010,28(3):213-228.
[6]邓洪涛,地球内核与磁场[J].武汉大学学报,2010,35(7):854-856.
[7]王福军.计算流体动力学分析——CFD软件原理与应用[M] .北京:清华大学出版社. 2004.
[8]佟桂芳,徐德龙,张强等.一种新改进的SIMPLE算法[J].西安建筑科技大学学报,2001,9,33(3):221~224.
[9]吴子牛.计算流体动力学基本原理[M].北京:科学出版社,2001.
[10]尚晶晶.Direct3D游戏开发技术详解[M].北京:人民邮电出版社,2006.
[11] (美)Hubert Nguyen著.杨柏林,陈根浪,王聪译.GPU精粹3[M].北京:清华大学出版社,2010.
[12] (美)Donald Hearn, M. Pauline Baker著.蔡士杰,宋继强,蔡敏译.计算机图形学[M].北京:电子工业出版社,2005.
[13]叶至军.C++ STL开发技术导引[M].北京:人民邮电大学,2007.
[14] (美)Michael Dickheiser著.孟宪武译.游戏编程精粹6[M].北京:人民邮电出版社,2007.
[15] (美)Steve McConnell著.金戈,汤凌,陈硕,张菲.代码大全第二版[M].北京:电子工业出版社,2006.
[16] (美)Mike McShaffry著.冯兆麟,孔祥一,李华杰译.游戏编程全接触[M].北京:人民邮电出版社,2006.
[17] (美)Scott Meyers著.侯捷译.Effective C++第三版[M].北京:电子工业出版社,2006.
[18] (美)Sartaj Sahni著.汪诗林,孙晓东译.数据结构算法与应用[M].北京:机械工业出版社,2000.
[19] (美)Scott Meyers著.Effective STL[M].北京:中国电力出版社,2003.
[20]王鹏杰,李威,王聪著.DirectX游戏编程[M].北京:机械工业出版社,2010.
[21] Enright Douglas, Marschner Stephen, Fedkiw Ronald. Animation and rendering of complex water surfaces[J].ACM Transactions on Graphicx, 2002, 21(3):736~744.
[22] Hong Jeong-Mo,Kim Chang-Hun.Animation of bubbles in liquid[J].Computer Graphics Forum, 2003, 22(3):253~262.
[23] Lamorlette Arnauld, Foster Nick. Structural modeling of flames for a production environment[J]. ACM Transactions on Graphics,2002, 21(3):729~735.
[24] Peachey Darwyn R. Modeling waves and surf[J].Computer Graphics, 1986, 20(4):65~74.
[25] Fournier Alain, Reeves William T. A Simple model of ocean waves[J]. Computer Graphics, 1986, 20(4):75~84.
[26] Kass Michael, Miller Gavin. Rapid, stable fluid dynamics for computer graphics[J]. Computer Graphics, 1990, 24(4):49~57.
[27] Foster Nick, Metaxas Dimitri. Realistic animation of liquids[J]. Graphical Models and Image Processing, 1996, 58(5):471~483.
[28] Stam Jos, Fiume Eugene. Dipicting fire and other gaseous phenomena using diffusion processes[A]. Computer Graphics Proceedings. 1995, 129~136.
[29] McNamara Antoine, Treuille Adrien, Popovic Zoran, etc. Fluid control using the adjoint method[J].ACM Transactions on Graphics, 2004, 23(3):449~456.
[30]邓宝庆,吴文权.同位网格上不可压流动的反欠松弛压力修正算法[J].工程热物理学报,2002,9,23(5):605~607.
[31]宋道云,刘洪来,方波等.动量差值与完全压力校正算法及交错网格SIMPLE算法的比较[J].华东理工大学学报,2003,4,29(2):124~143.
[32]冯天详.一个高斯-赛德尔方法解方程组的有趣结果[J].重庆三峡学院学报,2010,3,26(3):14~17.
[33]龚敏敏.GPU精粹2[M].北京:清华大学出版社,2007.5.
[34] Srivastava. S.P.Method of Interpretation of Magnetotelluric Data When Source Field Is Considered[J]. Journal of Geophysical Research,1965,4,70(4): 945-954.
[35]王水,李罗权.磁场重连[M].合肥:安徽教育出版社,1998.
[36] Furth PH, Killeen J, Rosenbluth MN. Finite-resistivity insablities of a sheet pinch. Phys Fliuds, 1963,6(4):459~484.
© 2004-2018 中国地质图书馆版权所有 京ICP备05064691号 京公网安备11010802017129号 地址：北京市海淀区学院路29号 邮编：100083 电话：办公室：(+86 10)66554848；文献借阅、咨询服务、科技查新：66554700 |