AI Nuclear Fusion 2022

Table of Contents

1. Introduction
2. Getting Started
3. Usage
4. Todo

Introduction

This project aims to develop a Particle in Cell plasma code. Original code in 2021 by Hilary,Yin Yue and Chloe, extensive improvements by Samuel,Ananth and Vishwa.

Getting Started

Prerequisites

• MSYS2

• to avoid confusion, use either of "mingw64" or "ucrt" and do not mix the two. The following examples make use pf ucrt

• Tools pacman -S base-devel cmake git mingw-w64-ucrt-x86_64-gcc paraview

• Libs (Opencl, OpenMP, gsl, fftw, vtk, nfftw) pacman -S mingw-w64-ucrt-x86_64-opencl-headers mingw-w64-ucrt-x86_64-opencl-clhpp mingw-w64-ucrt-x86_64-opencl-icd mingw-w64-ucrt-x86_64-openmp mingw-w64-ucrt-x86_64-gsl mingw-w64-ucrt-x86_64-fftw mingw-w64-ucrt-x86_64-vtk using float version of nfftw3f. should install from source using ./configure --enable-float https://www-user.tu-chemnitz.de/~potts/nfft/installation.php

• GCC added to PATH in MSYS

  • In the root directory, run export PATH=$PATH:/ucrt64/bin
  • this can also be added in windows "edit system environment variables" , "Path". Add C:\msys64\usr\bin and C:\msys64\ucrt64\bin

Usage

Todo

• Setup particles for hot rod project

• Setup particles for hot plate project

• Setup particles for MagLIF

  • Two sets of plasmas similar to hot rod project
    • 1. Low density plasma cylinder
    • 2. High density thin cylindrical shell around plasma cylinder.
  • Setup external fields.
    • Electric field along the cylinder.

• add in "artificial viscosity" to simulate energy loss/gain

  • F=q(E+vxB)+rv w
  • here viscosity r is negative when there is energy loss and positive when there is energy gain

• add temperature field Te[x][y][z]

  • approximate Te as average KE of particles

• viscosity_field[p][x][y][z]

  • P=Fv=r * v * v, So r=P_perparticle/(vv)
  • Bremstrahlung radiation loss from NRL Plasma formulary, https://tanimislam.github.io/research/NRL_Formulary_2019.pdf
  • Pbr=1.69e-32 * Ne * pow(Te,0.5) Sum(Z * Z * Ni(Z)) in Watts per cm^3
  • cyclotron radiation loss from NRL Plasma formulary
  • Pc=6.21-28 * B * B * Ne * Te in Watts per cm^3
  • power density from fusion from NRL Plasma formulary (Assume to be absorbed within the cell for the time being not realistic as fusion products may have very long range. if cell sizes are small, most of energy from fusion products will be lost from the cell) all in in Watts per cm^3
  • P_DD=3.3e-13 * N_D * N_D * sigma_v_dd
  • P_DT=5.6e-13 * N_D * N_T * sigma_v_DT
  • P_DHe3=2.9e-12 * N_D * N_He3 * sigma_v_DHe3
  • power transfer between electrons and ion

• to get more performance, you might want to recompile the libraries used. for example to install FFTW3 recompiled with OMP enabled (if not on Windows):

wget https://www.fftw.org/fftw-3.3.10.tar.gz tar xvzf fftw-3.3.10.tar.gz cd fftw-3.3.10/ ./configure --enable-threads --enable-openmp --enable-avx --enable-avx2 --enable-avx512 --enable-avx-128-fma --enable-float --with-our-malloc --enable-sse2 make make install

For Windows, we have to compile FFTW into one single shared library:

./configure --enable-threads --enable-sse2 --enable-avx --enable-avx2 --enable-avx-128-fma --with-our-malloc --enable-float --enable-shared --disable-static --with-combined-threads

• Next, we can also recompile NFFT to make use of multithreading:

cd ../nfft-3.5.3/ ./configure --enable-float --enable-openmp make make install

• Resulting library can be linked using -lnfft3f_threads