高能质子照相系统优化设计与减小色散模糊技术研究

High energy GeV proton is an important radiographic probe for dense object in the hydro-test experiments.As a diagnosis technology,it has been studied for many years in U.S and Russia. While, the relevant research in China is still at the primary stage. The relative energy loss of the incident proton gives birth to chromatic aberrations which mainly debases the image quality . Additionally, because the radiography system has several parts such as diffuser, matching lens and Zumbro lens. The parameters of each parts are correlated with each other, which means the parameters can not be optimized individually(one by one). Moreover, the constraints of the design given by the size of Field-of-View and beam tube have complicated influence on the parameters optimization. Thus, It is difficult but very important to optimize the radiography system. In this project, intelligent optimization algorithms will be applied to optimize all parameters with all the constraints and correlations. A novel type called auxiliary lens is firstly proposed which can evidently depress the blurring caused by chromatic aberrations. In addition, the blurring with which is degenerated by a space-variance spread function can be partly removed by algorithm. In this project, a reliable Monte Carlo simulation code will be developed. The study will include two parts. Firstly, based on experiment results, the physics models and cross section data,as well as the source code about GeV proton radiography of MCNP6,Geant4 and Phits will be studied and analyzed, then a suitable physics module will be build up. Secondly, the real magnetic distribution will be constructed in the Monte-Carlo simulation based on finite element method, then the boundary effect of magnetic field will be considered during the optimization. With suitable physics module and real magnetic field , Monte-Carlo code with high simulation efficiency will be obtained

在致密物质及流体动力学过程诊断研究领域，美俄已开展大量GeV高能质子照相理论与实验工作，国内相关研究尚处于初期阶段。目前，质子能损导致的色散是制约照相精度提升的关键问题。由于照相系统参数较多，各部件如扩束器、匹配、Zumbro透镜等相互耦合，视场、束流管道尺寸等对参数的约束复杂，系统整体优化设计也十分重要。本项目采用智能优化算法开展系统多约束、多目标（图像分辨率及重建密度精度等）优化设计。结合首次提出的新型辅助成像透镜概念设计，有望显著减小色散模糊并实现对不同面密度区域的高精度成像。根据质子能损色散失焦导致图像模糊的机理，开发适用于退化函数空变条件的图像重建算法予以去模糊复原。对比MCNP6、Geant4、Phits的质子模型，结合已有实验，优选适用于高能质子照相的物理模块包。采用有限元法构建真实磁场，将磁场边界效应纳入相应的系统优化设计中，结合适用物理包建立高能质子照相快速模拟蒙卡程序。

本项目研究主要围绕高能质子闪光照相展开，针对影响高能质子照相性能关键环节，包含高能质子与物质相互作用（质子在扩束器内的输运特性）、匹配透镜、准直、Zumbro磁透镜等，开展系统性研究。针对照相系统涉及参数多，参数间耦合性强等特点，开展照相系统整体多目标、多约束优化设计理论研究。采用结合智能优化算法-遗传算法、磁场输运数值方法与蒙特卡洛程序包Geant4的程序综合体，建立了一套完整的高能质子照相系统优化设计方案及相关优化算法程序，完成系统各关键部件特性进行独立研究与整体耦合优化设计，实现了对20GeV等能量质子照相系统的最优化设计。首次提出降低色差辅助透镜概念设计。根据蒙卡模拟研究与理论分析提出Zumbro透镜后增加辅助透镜的方案，避免了采用Zumbro透镜时仅能通过提高质子能量的方法降低色差。采用已建立的照相系统整体优化方法，设计得到了20GeV质子照相系统及辅助照相系统，经过蒙卡程序模拟验证，对于不同类型客体，辅助透镜均可有效提升系统对密度分布的测量精度。