funcs.py

"""
Useful functions for PTetra workshop
Author: Sigvald Marholm, 2022
"""

import pandas as pd
import numpy as np
import os
import re
import langmuir as l

def read_hst(folder_name):
    """ Read PTetra history file in folder """

    file_name = os.path.join(folder_name, 'pictetra.hst')

    column_names = [
        'timestep',
        'time',
        'netot',
        'nitot',
        'Te_eff',
        'pot1',
        'sc_phi_0',
        'sc_q_0',
        'sc_i_0',
        'sc_phi_1',
        'sc_q_1',
        'sc_i_1',
    ]

    df = pd.read_csv(file_name, sep='\s+', skiprows=2, names=column_names, index_col=0)

    # Add new column sc_i_tot for total collected spacecraft current
    df = df.assign(sc_i_tot=lambda a: a.sc_i_0 + a.sc_i_1)

    return df

def parse_parameters(folder_name):
    """
    Parses geometry and plasma parameters from folder name and the pictetra.dat
    file inside it, and returns a geometry and Species object from Langmuir, as
    well as the voltages according to pictetra.dat. Beware that only the
    electron species is included, and that the list of voltages is simply read
    from the sc_fixedPot entry in pictetra.dat. Thus, the voltages will not be
    correct for floating potential simulations.

    Example:

        geometry, electron, voltages = parse_parameters('Sphere_0.5R_3V_3V')
        print(geometry)
        print(electron.debye)

    """

    dat_file = os.path.join(folder_name, 'pictetra.dat')

    with open(dat_file) as file:
        lines = file.read()

    # Read parameters from pictetra.dat
    density = float(re.search(r'\s+ne\s*=\s*(\S*)', lines).groups()[0])
    temperature = float(re.search(r'\s+te\s*=\s*(\S*)', lines).groups()[0])
    voltages = re.search(r'\$begin sc_fixedPot\s*(\S*)\s*(\S*)', lines)
    voltages = list(map(float, voltages.groups()))

    # Create Langmuir species
    electron = l.Electron(n=density, eV=temperature)

    # Read geometry from folder name
    radius = float(re.search(r'_([^_]*)R', folder_name).groups()[0])
    if 'cylinder' in folder_name.lower():
        length = float(re.search(r'_([^_]*)L', folder_name).groups()[0])
        geometry = l.Cylinder(r=radius*electron.debye, l=length*electron.debye)
    else: # sphere
        geometry = l.Sphere(r=radius*electron.debye)

    return geometry, electron, voltages

def average(df, column, relaxation_time):
    """
    Perform exponential moving average on a column in a Pandas dataframe. The
    relaxation time is in seconds.

    Example:

        df = read_hst('Sphere_0.5R_3V_3V')
        sc_i_tot_av = average(df, 'sc_i_tot', 1e-6)
    """
    delta_t = df.time[1]-df.time[0]
    alpha = 1-np.exp(-delta_t/relaxation_time)
    return df[column].ewm(alpha=alpha).mean()