def make_toml(folder):
files = []
extlist = ["mpt", "csv", "txt", "xlsx", "ecdh"]
#Loop over all elements in current directory
for entry in os.listdir(folder):
# If it is a file and has the correct extension
if os.path.isfile(entry) and entry.split(".")[-1] in extlist:
# Add it to filelist
files.append(entry)
toml_str = """# "Data path", "active mass" and "nickname" (set active mass to 1 if data is normalized with regards to active mass)
files = [\n"""
if len(files)>0:
for file in files:
toml_str += '\t["' + file + '","1.0"],\n'
else:
LOG.warning("Could not find any files in the current folder!")
toml_str += '\t[" ","1.0"],\n'
toml_str += "]\n\n"
toml_str += gen_string(cfg_dict)
with open("ecdh.toml", "w") as f:
f.write(toml_str)
LOG.info("Wrote example configuration to 'ecdh.toml' with %.0f files found"%len(files))
def reduce_data(self, datatreatment):
"""
Takes the data and deletes points not matching the requirement.
Useful if you measured with too high sampling rate and have gigantic datafiles
Example: you sampled for every 1mV change, and every second, then with dV = 0.010 and dt = 10, all points that has less than a difference of 10mV AND less than 10s will be deleted.
"""
LOG.debug("cell.reduce_data() is running, reducing file: {}".format(
self.name))
LOG.info(
"Reducing file: {}. Follow loading bar below, this might take some time."
.format(self.name))
try:
dt = int(datatreatment["dt"])
except:
dt = 10
try:
dV = float(datatreatment["dV"])
except:
dV = 0.01
#dt = 10 #10 s
#dV = 0.01 # 10mV
dI = 0.01 #0.01 mA, 10uA
LOG.debug(f"Reduction parameters: dt: {dt}, dV: {dV}, dI: {dI}.")
#'time/s','Ewe/V', '<I>/mA'
last_t = self.df['time/s'].iloc[0]
last_V = self.df['Ewe/V'].iloc[0]
kill_index = []
max_i = len(self.df)
loadbar = 10
print("|----------|")
print("|", end="", flush=True)
for i, row in self.df.iloc[1:].iterrows():
curr_t = row.iloc[0]
curr_V = row.iloc[1]
if abs(last_t - curr_t) < dt and abs(last_V - curr_V) < dV:
kill_index.append(i)
else:
last_t = curr_t
last_V = curr_V
if i == int(max_i / loadbar):
print("-", end="", flush=True)
loadbar -= 1
if loadbar < 1:
loadbar += 1 #Wow this needs fixing
print("|")
filename, ext = os.path.splitext(self.fn)
self.df.drop(kill_index, inplace=True)
self.df.to_csv(path_or_buf=filename + "_reduced.ecdh")
def smooth_data(self, datatreatment):
"""
Takes the data and removes outliers
Useful if you have a couple of short circuits now and then
"""
LOG.info(f"Removing outliers on {self.name}.")
import pandas as pd
window = int(len(self.df) / 10000)
smoothing_df = pd.DataFrame()
smoothing_df['median'] = self.df['Ewe/V'].rolling(window).median()
smoothing_df['std'] = self.df['Ewe/V'].rolling(window).std()
expmode = self.df.experiment_mode
#filter setup
self.df = self.df[(self.df['Ewe/V'] <= smoothing_df['median'] +
3 * smoothing_df['std'])
& (self.df['Ewe/V'] >= smoothing_df['median'] -
3 * smoothing_df['std'])].ewm(alpha=0.9).mean()
self.df.experiment_mode = expmode
def run():
import sys
if len(sys.argv) < 3: #Then no folder is specified, look for toml in local folder.
if os.path.isfile("./ecdh.toml"):
path = "./ecdh.toml"
else:
LOG.error("Could not find an ecdh.toml file in the current directory! \nOptions: \n1. Run ecdh with the argument 'run' followed by the path of your .toml file. \n2. Initiate toml file in this directory with the init argument.")
sys.exit()
elif os.path.isfile(sys.argv[2]):# File was inserted Read toml config
path = sys.argv[2]
else:
LOG.error("Cannot find the .toml configuration file!")
sys.exit()
LOG.debug("Reading config file: '{}'".format(path))
# Read in configuration file
config = read_config(path)
settings = config["settings"]
# Merge cycle range into specific cycles
if settings['cycle_range']:
try:
cyclerange = np.linspace(settings['cycle_range'][0], settings['cycle_range'][1], settings['cycle_range'][1]-settings['cycle_range'][0] + 1).astype(int)
if type(settings['specific_cycles']) is bool:
settings['specific_cycles'] = cyclerange.tolist()
else:
settings['specific_cycles'] += cyclerange.tolist()
LOG.info(f"Specific cycles: {settings['specific_cycles']}")
except Exception as e:
LOG.warning(f"Could not use the cycle range, Error: {e}")
datatreatment = config["datatreatment"]
# Check that files are found
files = check_files(config["files"])
if len(files) == 0:
import sys
LOG.error("Could not load any datafiles. Exiting. Check that the filepaths are typed correctly in the configuration file.")
sys.exit()
LOG.success("Running ECDH: Found {} datafiles!".format(len(files)))
# Define plot specifications
plot = Plot(numfiles=len(files), **settings)
# Run the data reading + plot generation
cells = []
for f in files:
try:
am_mass = f[1]
except:
am_mass = None
try:
nickname = f[2]
except:
nickname = None
cell = Cell(f[0], am_mass, nickname, plot=plot, specific_cycles = settings['specific_cycles'])
cell.get_data()
#cell.edit_GC()
#cell.treat_data(settings)
cell.plot()
#cells.append(cell)
if datatreatment['reduce_data']:
cell.reduce_data(datatreatment)
if datatreatment['smooth_data']:
cell.smooth_data(datatreatment)
if datatreatment['print_capacities']:
if not os.path.isfile("capacity_intervals.json"):
with open("capacity_intervals.json", "w") as f:
f.close()
import json
new_json = cell.get_capacities(datatreatment)
with open("capacity_intervals.json",'r+') as file:
# First we load existing data into a dict.
try:
file_data = json.load(file)
except:
file_data = []
# Join new_data with file_data inside emp_details
file_data.append(new_json)
# Sets file's current position at offset.
file.seek(0)
# convert back to json.
json.dump(file_data, file, indent = 4)
if 'savefig' in settings:
plot.draw(save = settings['savefig'])
else:
plot.draw()
def check_df(df):
"""Check if the dataframe has:
- cycle number (sometimes the Z1 counter f***s up and just says 0)
- If it is GC data: then calculate capacity
- removing data where the cell is resting
If anything is wrong, it tries to fix it"""
if df.experiment_mode == 1: #Then its GC
if df['cycle number'].eq(0).all() or df['cycle number'].max() < 100: #If all cycle numbers are 0, then maybe Z1 counter was not iterated properly.
LOG.info("We only found one cycle in '{}', and suspect this to be false. Checking now if there should be more cycles.".format(df.name))
#We fix this by counting our own cycles.
#Keeping track of signs of current (positive or negative) and cycle number
prev_sign = True
sign = True
cycle_number = 1
new_cycle_indexes = []
for i,current in df['<I>/mA'].items():
if current > 0:
sign = True
df['charge'].at[i] = True
elif current < 0:
sign = False
df['charge'].at[i] = False
if prev_sign is False and sign is True:
#Changing from a discharge to a charge means new cycle
prev_sign = True
cycle_number += 1
#df['cycle number'].at[i-1] = cycle_number
new_cycle_indexes.append(i-1)
elif prev_sign is True and sign is False:
#Changing from a charge to a discharge
prev_sign = False
new_cycle_indexes.append(i-1)
# In place editing of cycle number
df['cycle number'].at[i] = cycle_number
#Remove rows where a new experiment start (BatSmall has f****d datalogging here, where the current and voltage is the same as the prev step, but the capacity restarts)
#df.drop(new_cycle_indexes, axis = 0, inplace = True)
if cycle_number > 1:
LOG.info("Found {} cycles in {}".format(cycle_number, df.name))
else:
LOG.info("There seems to only be one cycle. Sorry for wasting your time.")
else: #charge bool then isnt fixed
for i,current in df['<I>/mA'].items():
if current > 0:
sign = True
df['charge'].at[i] = True
elif current < 0:
sign = False
df['charge'].at[i] = False
def get_capacities(self, datatreatment):
"""
Returns capacity in a potential interval for each cycle.
Appends to filename "./capacities.txt"
"""
if not len(datatreatment['print_capacities']) % 2 == 0:
LOG.critical(
"You wanted to get the potential-interval capacities, but you inserted an odd number of boundaries. You need two boundaries for each range, eg: [3.5, 4.4, 4.4, 5.0] will be two ranges from 3.5->4.4 and 4.4->5.0. The capacities will not be printed."
)
else:
intervals = []
for i, pot in enumerate(datatreatment['print_capacities']):
if i % 2 != 0: #if the index number is odd, we are making a range of it
intervals.append(
(datatreatment['print_capacities'][i - 1], pot))
LOG.info(
"Found intervals for capacity print: {}".format(intervals))
if self.GCdata:
data = self.GCdata
elif self.CVdata:
data = self.CVdata
else:
self.edit_GC()
import numpy as np
def _find_charge(chg, interval):
start, stop = interval
cap, pot = chg
pot = np.asarray(pot)
idxstart = (np.abs(pot - start)).argmin()
idxstop = (np.abs(pot - stop)).argmin()
capdiff = cap[idxstop] - cap[idxstart]
return abs(capdiff)
# Format list like: [chg, dchg], chg = [cap in interval 1, cap in interval 2, cap in interval n]
caps = {self.name: {}}
for interval in intervals:
caps[self.name][str(interval)] = {
"charge": [],
"discharge": []
}
if self.specific_cycles:
for i, cycle in enumerate(data):
if i in self.specific_cycles:
chg, dchg = cycle
for interval in intervals:
caps[self.name][str(interval)]["charge"].append(
_find_charge(chg, interval))
caps[self.name][str(interval)]["discharge"].append(
_find_charge(dchg, interval))
else:
for i, cycle in enumerate(data):
chg, dchg = cycle
for interval in intervals:
caps[self.name][str(interval)]["charge"].append(
_find_charge(chg, interval))
caps[self.name][str(interval)]["discharge"].append(
_find_charge(dchg, interval))
return caps