Efficient Laser Radiation Symmetry Analysis Based On Modified Conjugate Gradient-iterative Hard Thresholding

Laser indirectly driven Inertial Confinement Fusion(ICF)is one of the key research topics in developed countries in the world.This is because the ICF releases enormous fusion energy that can be used to solve the energy crisis that humans will encounter.The radiation symmetry of the cavity and capsule is the key to the success of the ICF experiment.The calculation efficiency of the radiation flux on the inner wall of the cavity and the surface of the capsule directly affects the evaluation efficiency of radiation symmetry.Therefore,solving the radiation flux balance equation model efficiently is an important prerequisite for ICF.How to improve the efficiency of the radiation flux balance equation model was analyzed and studied in this dissertation.Firstly,since in the steady-state conditions,the radiation energy and absorbed energy of the inner wall of the spherical cavity and the surface of the spherical capsule are equaling,so based on the view factor algorithm,the balance equation model of the radiation flux of the cavity and the capsule is established.Secondly,the mesh element on the surface of the spherical cavity and the surface of the spherical capsule are discretized,arranged in a certain order,and the radiation flux balance equation model is established on any discrete grid surface element.Thirdly,since the radiation flux balance equation model are nonlinear equation model,and in order to meet the precision requirements of ICF simulation experiments,usually,the discrete mesh element need to be subdivided very smaller,which results in a large scale of equation model,if traditional iterative methods are used to solve,it may take a long time to even solve the problem.To solve this problem,combined the compressive sensing method,an efficient method which is for solving the radiation flux based on the modified conjugate gradient-iterative hard threshold was proposed.The main research content of this method is as follows:?In order to improve the efficiency of the radiation flux equilibrium equation model,first of all,the sparse characteristics of the radiation flux on the inner wall of the spherical cavity and the surface of the spherical capsule are found,and the spherical harmonics of the unfolding order and sparse degree are obtained through experiments.Then the Latin hypercube sampling method and the random sampling calculation formulas is used to calculate their observations respectively.Last,the large-scale nonlinear equation model of radiation flux are transformed into small-scale coefficient equation model,which greatly reduces the computational complexity of the radiation flux balance equation model and improves its solution efficiency.?The optimal reconstruction algorithm is also an essential part of the equation model to speed up the solution.In view of the insufficiency of the iterative hard threshold(IHT)and the normalized iterative hard threshold(NIHT)algorithm etc.in the reconstruction of nonlinear compressed sensing problems,combined with the improvement of the Conjugate Gradient-Iterative Hard Thresholding(CG-IHT)algorithm,this paper proposes the Modified Conjugate Gradient-Iterative Hard Thresholding(MCG-IHT)algorithm.And through simulation experiments,it is verified that the proposed reconstruction algorithm can be accelerated 4 to 90 times faster than the traditional iterative method,when the mesh element is larger,its acceleration ratio is also larger.Based on the above research,IRad3 D simulation software is introduced to analyze simulation optimization examples of radiation symmetry analysis.Finally,the validity and reliability of the research results was the verified by the application example.