readme.radtr

XPDC1是针对静电模型中的等离子体模拟程序！通过INP文件输入参数有较好的通用性

                                                      November 2000

         INSTRUCTIONS FOR USING RADIATION TRANSPORT SIMULATION
	             /HAE JUNE LEE/
  
  In order to simulate the radiation transport in xpdc1 code, you should
  input radiation transport input parameters. It is located at the end of
  each input files, and the meaning of each parameter is explained in the
  manual. 

  When you run xpdc1 code with RT_flag ON (nonzero value), The code will
  ask you of the filename of A_km matrix. If you have it, just type the
  filename, and if not, type "no" or other meaningless characters.
  Be sure that the word you type should not be same as any filename in 
  the working directory. In this case, the code calculate the matrix
  elements and show it in the DIAGNOSTICS panel with a title of `A_ij'.
  Once it is calculated, you can save it in a text file. Click `print' 
  box of the diagnostic window and select your file type as ASCII, and 
  input the filename in the `File name' window.
  For example, if the input file name is 'test.inp', you may type the
  filename as 'test.Akm'. Don't forget that it has to be an ASCII file.

  If you make and save an A_km matrix, it is very effective when you run
  many simulations with change of parameters which are not related to 
  radiation transport, such as axial currents or initial plasma densities. 
  But if you change the system length, gas pressure, or temperature, the
  radiation transport properties may be changed by it. Be sure to check 
  the displayed `delta_nu' and 'k0', and cell size to make it sure if you
  should make the A_km matrix again or not.
  I made the code to be able to use arbitrary k0 for the test of opacity
  dependence; therefore the displayed k0 can be different from the input 
  parameter k0. The code use the inputed value instead of displayed value.

  Besides, you should be careful to choose the maximum and minimum values of
  a frequency domain. In this model, you should use the same absolute value
  of x_max and x_min for a symmetry and you can select only the Lorentz
  or the Doppler lineshape. But it can be extended to more general cases
  in the future. In order to choose proper value of x_max and x_min, 
  you better take a look at the "Lineshape(f)" diagnostic which shows the
  lineshape. When the multiplication of k0 and minimum value of lineshape,
  if should be much less than 1 for precise simulation results. In addition,
  if you use large value of x_max or x_min (e.g. for the Lorentz lineshape), 
  you should also use large x_bin to reduce numerical error. But it becomes
  a very intensive calculation for the case. Therefore, it takes long 
  simulation time to get an A_km for the high pressure case where the Lorentz 
  lineshape should be considered.

  This version doesn't include parallel processing for the RT part, yet.
  In order to simulate high pressure case with a large number of super
  particles, you should turn-off RT_flag and simulate with the parallel 
  version to reach a steady state, and then simulate with RT_flag ON using
  a single processor with the saved dump file. It will take long time 
  to reach a steady state for the radiative and the metastable excited 
  states too, but it's the only way for now.