-
Notifications
You must be signed in to change notification settings - Fork 0
/
reader.py
555 lines (467 loc) · 20.8 KB
/
reader.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
#!/usr/bin/env python
import numpy as np
import pandas as pd
from os import path
from scipy.integrate import simps
import re
from pathlib import Path
from argparse import ArgumentParser
CYC_TYPES = {'charge', 'discharge', 'cycle'}
RATES = np.array([1/160, 1/80, 1/40, 1/20, 1/10, 1/5, 1/4, 1/3, 1/2, 1, 2, 3, 4, 5])
C_RATES = ['C/160', 'C/80', 'C/40', 'C/20', 'C/10', 'C/5', 'C/4', 'C/3', 'C/2', '1C', '2C', '3C', '4C', '5C']
class ParseNeware():
def __init__(self, newarefile, all_lines=None, ref_cap=None, tmppath=''):
'''parse raw neware datafile into cycle, step, and record data\
and put into pd df'''
if all_lines is not None:
lines = []
self.newarefile = newarefile
for line in all_lines:
lines.append(line.decode('unicode_escape'))
else:
self.newarefile = newarefile[:-4]
with open(newarefile, 'r', encoding='unicode_escape') as f:
lines = f.readlines()
# Replace single space between words with "_" to create column labels.
cyclabels = re.sub(r'(\w+) (\w+)', r'\1_\2', lines[0])
cyclabels = re.sub(r'(\w+) (\w+)', r'\1_\2', cyclabels)
cyclabels = re.sub(r' ', r'', cyclabels)
clabels = cyclabels.strip().split()
steplabels = re.sub(r'(\w+) (\w+)', r'\1_\2', lines[1])
steplabels = re.sub(r'(\w+) (\w+)', r'\1_\2', steplabels)
steplabels = re.sub(r' ', r'', steplabels)
slabels = steplabels.strip().split()
reclabels = re.sub(r'(\w+) (\w+)', r'\1_\2', lines[2])
reclabels = re.sub(r'(\w+) (\w+)', r'\1_\2', reclabels)
reclabels = re.sub(r' ', r'', reclabels)
rlabels = reclabels.strip().split()
cyclnlen = len(clabels)
# print('Found {} cycle labels.'.format(cyclnlen))
steplnlen = len(slabels)
# print('Found {} step labels.'.format(steplnlen))
reclnlen = len(rlabels)
# print('Found {} record labels.'.format(reclnlen))
# Parse out units from column labels and create dictionary of units
# for cycle, step, and record data.
self.cycunits = dict()
newclabels = []
cycheader = ''
for l in clabels:
try:
m = re.search(r'\(.*\)', l)
newlab = l[:m.start()]
if newlab == 'Specific_Capacity-Dchg':
newlab = 'Specific_Capacity-DChg'
if newlab == 'RCap_Chg':
newlab = 'Specific_Capacity-Chg'
if newlab == 'RCap_DChg':
newlab = 'Specific_Capacity-DChg'
self.cycunits[newlab] = l[m.start()+1:m.end()-1]
newclabels.append(newlab)
cycheader = cycheader + '\t{}'.format(newlab)
except:
self.cycunits[l] = None
newclabels.append(l)
cycheader = cycheader + '\t{}'.format(l)
self.stepunits = dict()
stepheader = 'Cycle_ID'
newslabels = ['Cycle_ID']
for l in slabels:
try:
m = re.search(r'\(.*\)', l)
newlab = l[:m.start()]
self.stepunits[newlab] = l[m.start()+1:m.end()-1]
newslabels.append(newlab)
stepheader = stepheader + '\t{}'.format(newlab)
except:
self.stepunits[l] = None
newslabels.append(l)
stepheader = stepheader + '\t{}'.format(l)
self.recunits = dict()
recheader = 'Cycle_ID\tStep_ID'
newrlabels = ['Cycle_ID', 'Step_ID']
for l in rlabels:
try:
m = re.search(r'\(.*\)', l)
newlab = l[:m.start()]
if newlab == 'Vol':
newlab = 'Voltage'
if newlab == 'Cap':
newlab = 'Capacity'
if newlab == 'CmpCap':
newlab = 'Capacity_Density'
if newlab == 'Cur':
newlab = 'Current'
self.recunits[newlab] = l[m.start()+1:m.end()-1]
newrlabels.append(newlab)
recheader = recheader + '\t{}'.format(newlab)
except:
self.recunits[l] = None
newrlabels.append(l)
recheader = recheader + '\t{}'.format(l)
# Create header line for cycle, step, and record data
self.cyc = ['{}\n'.format(cycheader)]
self.step = ['{}\n'.format(stepheader)]
self.rec = ['{}\n'.format(recheader)]
# Separate cycle, step, and record data and write to
# file (needs to be changed to tmpfile) to be read
# in as DataFrame with inferred dtypes.
cyc_nlws = len(lines[0]) - len(lines[0].lstrip())
step_nlws = len(lines[1]) - len(lines[1].lstrip())
rec_nlws = len(lines[2]) - len(lines[2].lstrip())
for line in lines[3:]:
l = line.strip().split()
nlws = len(line) - len(line.lstrip())
if nlws == cyc_nlws:
cycnum = l[0]
self.cyc.append(line)
elif nlws == step_nlws:
stepnum = l[0]
self.step.append('{0}{1}'.format(cycnum, line))
else:
self.rec.append('{0}\t{1}{2}'.format(cycnum, stepnum, line[1:]))
# if len(l) == cyclnlen:
# cycnum = l[0]
# cyc.append(line)
# elif len(l) == steplnlen:
# stepnum = l[0]
# step.append('{0}{1}'.format(cycnum, line))
# else:
# rec.append('{0}\t{1}{2}'.format(cycnum, stepnum, line))
tmppath = Path(tmppath)
#with open('cyc.dat', 'w') as f:
with open(tmppath / 'cyc.dat', 'w') as f:
for l in self.cyc:
f.write(l)
#with open('step.dat', 'w') as f:
with open(tmppath / 'step.dat', 'w') as f:
for l in self.step:
f.write(l)
#with open('rec.dat', 'w') as f:
with open(tmppath / 'rec.dat', 'w') as f:
for l in self.rec:
f.write(l)
self.cyc = pd.read_csv(tmppath / 'cyc.dat', sep='\t+', header=0, engine='python')
self.step = pd.read_csv(tmppath / 'step.dat', sep='\t+', header=0, engine='python')
self.rec = pd.read_csv(tmppath / 'rec.dat', sep='\t+', header=0, engine='python')
nstep = self.step['Step_ID'].values[-1]
# self.step['C_Rate'] = ['N/A']*nstep
step_rates = ['N/A']*nstep
# try:
ncyc = self.get_ncyc()
print('Found {} cycles.'.format(ncyc))
cycnums = np.arange(1, ncyc+1)
chg_cur_max = np.zeros(ncyc)
dis_cur_max = np.zeros(ncyc)
chg_rates = []
chg_crates = []
dis_rates = []
dis_crates = []
cyc_id, dcap = self.get_discap()
# ref_cap = np.amax(dcap)
if ref_cap is None:
max_inds = np.argpartition(dcap, -5)[-5:]
ref_cap = np.sum(dcap[max_inds]) / 5
for i in range(ncyc):
cycle = self.rec.loc[self.rec['Cycle_ID'] == cycnums[i]]
# print(cycle.columns)
# print(cycle['Record_ID'])
stepnums = cycle['Step_ID'].unique()
chg = cycle.loc[cycle['Step_ID'] == stepnums[0]]
chg_cur = chg['Current'].values
chg_cur_max = np.amax(np.absolute(chg_cur))
if chg_cur_max > 0.0:
cr = chg_cur_max / ref_cap
ind = np.argmin(np.absolute(RATES - cr))
chgrate = RATES[ind]
if chgrate not in chg_rates:
chg_rates.append(chgrate)
chg_crates.append(C_RATES[ind])
step_rates[stepnums[0]-1] = C_RATES[ind]
dis = cycle.loc[cycle['Step_ID'] == stepnums[-1]]
dis_cur = dis['Current'].values
dis_cur_max = np.amax(np.absolute(dis_cur))
if dis_cur_max > 0.0:
cr = dis_cur_max / ref_cap
ind = np.argmin(np.absolute(RATES - cr))
disrate = RATES[ind]
if disrate not in dis_rates:
dis_rates.append(disrate)
dis_crates.append(C_RATES[ind])
step_rates[stepnums[-1]-1] = C_RATES[ind]
self.step['C_rate'] = step_rates
print('Found charge C-rates: {}'.format(chg_crates))
print('Found discharge C-rates: {}'.format(dis_crates))
self.chg_crates = chg_crates
self.dis_crates = dis_crates
'''
# stuff for assinging directly to df instead of writing to file then
# using read_csv(). Issue is dtypes. Pandas infers when reading from file
# but not if assigning df from lists. Once resovled, replace csv write/read.
cyclnlen = len(cyc[0])
if cyclnlen > len(newclabels):
d = cyclnlen - len(newclabels)
for i in range(d):
newclabels.append('NA{}'.format(i+1))
steplnlen = len(step[0])
if steplnlen > len(newslabels):
d = steplnlen - len(newslabels)
for i in range(d):
newslabels.append('NA{}'.format(i+1))
reclnlen = len(rec[0])
if reclnlen > len(newrlabels):
d = reclnlen - len(newrlabels)
for i in range(d):
newrlabels.append('NA{}'.format(i+1))
self.cyc = pd.DataFrame.from_records(cyc, columns=newclabels)
self.step = pd.DataFrame.from_records(step, columns=newslabels)
self.rec = pd.DataFrame.from_records(rec, columns=newrlabels)
'''
if self.recunits['Voltage'] == 'mV':
self.rec['Voltage'] = self.rec['Voltage'] / 1000
self.recunits['Voltage'] = 'V'
# Convert Capacity_Density to mAh/g from mAh/kg. Unit label can be wrong.
cyc2_df = self.rec.loc[self.rec['Cycle_ID'] == 3]
max_cap = cyc2_df['Capacity_Density'].values
if np.amax(max_cap) > 1000:
self.rec['Capacity_Density'] = self.rec['Capacity_Density'] / 1000
self.cyc['Specific_Capacity-DChg'] = self.cyc['Specific_Capacity-DChg'] / 1000
if self.recunits['Capacity_Density'] == 'mAh/kg':
self.recunits['Capacity_Density'] = 'mAh/g'
# The following set of functions are designed to be intuitive
# for people in the lab that want to get particular information
# from cycling data.
def get_rates(self, cyctype='cycle'):
if cyctype not in CYC_TYPES:
raise ValueError('cyctype must be one of {0}'.format(CYC_TYPES))
if cyctype == 'charge':
rates = self.chg_crates
#return self.chg_crates
elif cyctype == 'discharge':
rates = self.dis_crates
#return self.dis_crates
elif cyctype == 'cycle':
#rates = self.chg_crates
rates = [r for r in self.chg_crates if r in self.dis_crates]
#for r in self.dis_crates:
# if r not in rates:
# rates.append(r)
return rates
def get_ncyc(self):
'''
Returns the total number of cycles.
'''
return int(self.cyc['Cycle_ID'].values[-1])
def select_by_rate(self, rate, cyctype='cycle'):
'''
Return record data for all cycles that have a particular rate.
rate: dtype=string.
cyctype: {'cycle', 'charge', 'discharge'}
'''
if rate not in C_RATES:
raise ValueError('rate must be one of {0}'.format(C_RATES))
if cyctype not in CYC_TYPES:
raise ValueError('cyctype must be one of {0}'.format(CYC_TYPES))
if cyctype == 'charge':
if rate not in self.chg_crates:
raise ValueError('There are no {} charges. Select a different rate.'.format(rate))
if cyctype == 'discharge':
if rate not in self.dis_crates:
raise ValueError('There are no {} discharges. Select a different rate.'.format(rate))
selected_cycs = []
stepdata = self.step.loc[self.step['C_rate'] == rate]
rate_stepnums= self.step.loc[self.step['C_rate'] == rate]['Step_ID'].unique()
print(len(rate_stepnums))
cycnums = stepdata['Cycle_ID'].unique()
for i in range(len(cycnums)):
stepnums = self.step.loc[self.step['Cycle_ID'] == cycnums[i]]['Step_ID'].unique()
if cyctype == 'cycle':
if (stepnums[0] in rate_stepnums) & (stepnums[-1] in rate_stepnums):
#if len(stepnums) >= 2:
selected_cycs.append(cycnums[i])
elif cyctype == 'charge':
if stepnums[0] in rate_stepnums:
selected_cycs.append(cycnums[i])
elif cyctype == 'discharge':
if stepnums[-1] in rate_stepnums:
selected_cycs.append(cycnums[i])
#for i in range(len(cycnums)):
#stepnums = stepdata.loc[stepdata['Cycle_ID'] == cycnums[i]].values
#stepnums = stepdata.loc[stepdata['Cycle_ID'] == cycnums[i]]['Step_ID'].unique()
'''
if cyctype == 'cycle':
if len(stepnums) >= 2:
selected_cycs.append(cycnums[i])
elif cyctype == 'charge':
step = self.step.loc[self.step['Step_ID'] == stepnums[0]]
if step['C_rate'].values == rate:
selected_cycs.append(cycnums[i])
elif cyctype == 'discharge':
step = self.step.loc[self.step['Step_ID'] == stepnums[-1]]
if step['C_rate'].values == rate:
selected_cycs.append(cycnums[i])
'''
# For cyctype='cycle' need to check that first and last step both have same rate.
# For charge/discharge need to check that first/last step are at C_rate=rate
return selected_cycs
def get_discap(self, cycnums=None, normcyc=None, active_mass=None):
'''
Arguments
- normcyc : cycle number to use as normalization.
- active_mass : used to calc specific capacity, in units of grams.
Returns
- cycle numbers
- discharge capacity or specific capacity if active_mass is passed.
'''
if cycnums is not None:
cap = self.cyc[self.cyc['Cycle_ID'].isin(cycnums)]['Cap_DChg']
else:
cap = self.cyc['Cap_DChg']
cycnums = self.cyc['Cycle_ID']
if active_mass is not None:
cap = cap / active_mass
if normcyc is not None:
normcyc = int(normcyc)
#cap = self.cyc[caplabel]
return cycnums, cap / cap[normcyc]
else:
return cycnums, cap
#return self.cyc['Cycle_ID'], self.cyc['Cap_DChg']
def get_chgcap(self, normcyc=None, active_mass=None):
'''
Returns the cycle numbers and charge capacity
'''
cap = self.cyc['Cap_Chg']
if active_mass is not None:
cap = cap / active_mass
if normcyc is not None:
normcyc = int(normcyc)
return self.cyc['Cycle_ID'], cap / cap[normcyc]
else:
return self.cyc['Cycle_ID'], cap
def get_deltaV(self, normcyc=None, cycnums=None):
'''
Return the difference between average charge and discharge voltages
computed by average value theorem (integrate V-Q)
NOTE: This still needs work. Crashes if V-Q curve data is too noisy.
'''
if cycnums is not None:
cycle_nums = cycnums
else:
cycle_nums = np.arange(1, self.get_ncyc() + 1)
# dVfile = '{}-deltaV.csv'.format(self.newarefile)
# if path.isfile(dVfile) is True:
# df = pd.read_csv(dVfile)
# cycnums = df['Cycle_ID'].values
# dV = df['Delta_V'].values
# else:
# if normcyc is not None:
# startcyc = normcyc
bad_inds = []
# cycnums = np.arange(1, self.get_ncyc()+1)
dV = np.zeros(len(cycle_nums), dtype=float)
for i in range(len(cycle_nums)):
Qchg, Vchg = self.get_vcurve(cycnum=i+1, cyctype='charge')
Qdis, Vdis = self.get_vcurve(cycnum=i+1, cyctype='discharge')
if ( (Qchg[-1] - Qchg[0]) < 0.001 ) or ( (Qdis[-1] - Qdis[0]) < 0.001 ):
# bad_inds.append(i)
continue
else:
dVchg = (1/(Qchg[-1] - Qchg[0]))*simps(Vchg, Qchg)
dVdis = (1/(Qdis[-1] - Qdis[0]))*simps(Vdis, Qdis)
dV[i] = dVchg - dVdis
if len(bad_inds) > 0:
dV = np.delete(dV, bad_inds)
cycle_nums = np.delete(cycle_nums, bad_inds)
# df = pd.DataFrame(data={'Cycle_ID': cycnums, 'Delta_V': dV})
# df.to_csv(path_or_buf=dVfile, index=False)
# print(bad_inds)
if normcyc is not None:
return cycle_nums, dV/dV[normcyc]
else:
return cycle_nums, dV
def get_vcurve(self, cycnum=-1, cyctype='cycle', active_mass=None):
'''
Get voltage curve (cap, V) for a specific cycle number (cycnum).
TODO: Need to deal with error handling properly.
'''
if cyctype not in CYC_TYPES:
raise ValueError('cyctype must be one of {0}'.format(CYC_TYPES))
if cycnum == -1:
cycnum = self.get_ncyc() - 1
try:
cycle = self.rec.loc[self.rec['Cycle_ID'] == cycnum]
except:
# raise Exception('Cycle {} does not exist. Input a different cycle number.'.format(cycnum))
print('Cycle {} does not exist. Input a different cycle number.'.format(cycnum))
stepnums = cycle['Step_ID'].unique()
if cyctype == 'charge':
chg = cycle.loc[cycle['Step_ID'] == stepnums[0]]
voltage = chg['Voltage'].values
capacity = chg['Capacity'].values
elif cyctype == 'discharge':
dis = cycle.loc[cycle['Step_ID'] == stepnums[-1]]
voltage = dis['Voltage'].values
capacity = dis['Capacity'].values
elif cyctype == 'cycle':
chg = cycle.loc[cycle['Step_ID'] == stepnums[0]]
Vchg = chg['Voltage'].values
Cchg = chg['Capacity'].values
dis = cycle.loc[cycle['Step_ID'] == stepnums[-1]]
Vdchg = dis['Voltage'].values
Cdchg = dis['Capacity'].values
voltage = np.concatenate((Vchg, Vdchg))
capacity = np.concatenate((Cchg, -Cdchg+Cchg[-1]))
if active_mass is not None:
return capacity / active_mass, voltage
else:
return capacity, voltage
# TODO: try smoothing vcurves before taking dQ/dV instead of
# smoothing after taking derivative.
def get_dQdV(self, cycnum=-1, cyctype='cycle', active_mass=None,
avgstride=None):
'''
Get dQdV for specific cycle. Returns charge and discharge together.
TODO: Add running average.
'''
cchg, vchg = self.get_vcurve(cycnum=cycnum, cyctype='charge')
cend = cchg[-1]
# vchg, cchg = vchg[:-1], cchg[:-1]
delta_cchg = cchg[1:] - cchg[:-1]
delta_vchg = vchg[1:] - vchg[:-1]
inf_inds = np.where(np.absolute(delta_vchg) < 1e-12)
vchg = np.delete(vchg, inf_inds[0] + 1)
cchg = np.delete(cchg, inf_inds[0] + 1)
dQdVchg = (cchg[1:] - cchg[:-1]) / (vchg[1:] - vchg[:-1])
# vchg = (vchg[1:] + vchg[:-1]) / 2
cdchg, vdchg = self.get_vcurve(cycnum=cycnum, cyctype='discharge')
cdchg = -cdchg + cend
# vdchg, cdchg = vdchg[1:], cdchg[1:]
delta_cdchg = cdchg[1:] - cdchg[:-1]
delta_vdchg = vdchg[1:] - vdchg[:-1]
inf_inds = np.where(np.absolute(delta_vdchg) < 1e-12)
vdchg = np.delete(vdchg, inf_inds[0] + 1)
cdchg = np.delete(cdchg, inf_inds[0] + 1)
dQdVdchg = -(cdchg[1:] - cdchg[:-1]) / (vdchg[1:] - vdchg[:-1])
# vdchg = (vdchg[1:] + vdchg[:-1]) / 2
voltage = np.concatenate((vchg[1:], vdchg[:-1]))
dQdV = np.concatenate((dQdVchg, dQdVdchg))
if avgstride is not None:
voltage, dQdV = self.runavg(voltage, dQdV, avgstride)
return voltage, dQdV
# return np.concatenate((vchg[1:], vdchg[:-1])), np.concatenate((dQdVchg, dQdVdchg))
def runavg(self, xarr, yarr, avgstride):
'''
Running average.
'''
window = avgstride*2 + 1
weights = np.repeat(1.0, window)/window
avgdata = np.convolve(yarr, weights, 'valid')
return xarr[avgstride: -avgstride], avgdata
if __name__ == '__main__':
parser = ArgumentParser()
parser.add_argument('newarefile')
args = parser.parse_args()
# this does nothing currently. Need to think about
# what command line utils to include.
nd = ParseNeware(args.newarefile)