3 Components Data¶

In [2]:

import numpy as np
import time

from ml4pd import components
import dist_utils
from ml4pd.aspen_units import Distillation
from ml4pd.streams import MaterialStream

input_molecules = dist_utils.get_mol_labels()
components.set_components(input_molecules['name'].to_list())

raw_data = dist_utils.get_benchmark_data(comp=3, dates='220520')
data = dist_utils.add_name_columns(raw_data, input_molecules[['name', 'mol']])
data = dist_utils.add_flow_perc(data)
data = dist_utils.rename_flowrate_columns(data)

import numpy as np import time from ml4pd import components import dist_utils from ml4pd.aspen_units import Distillation from ml4pd.streams import MaterialStream input_molecules = dist_utils.get_mol_labels() components.set_components(input_molecules['name'].to_list()) raw_data = dist_utils.get_benchmark_data(comp=3, dates='220520') data = dist_utils.add_name_columns(raw_data, input_molecules[['name', 'mol']]) data = dist_utils.add_flow_perc(data) data = dist_utils.rename_flowrate_columns(data)

Benchmark Data Info¶

In [3]:

input_molecules.T

input_molecules.T

Out[3]:

	0	1	2	3	4	5	6	7	8	9	10	11	12	13
name	acetone	2-butanone	3-pentanone	3-hexanone	4-heptanone	4-octanone	5-nonanone	5-decanone	6-undecanone	6-dodecanone	6-tridecanone	7-tetradecanone	8-pentadecanone	water
mol	C3-ONE	C4-ONE	C5-ONE	C6-ONE	C7-ONE	C8-ONE	C9-ONE	C10-ONE	C11-ONE	C12-ONE	C13-ONE	C14-ONE	C15-ONE	H2O
smiles	CC(=O)C	CCC(=O)C	CCC(=O)CC	CCCC(=O)CC	CCCC(=O)CCC	CCCCC(=O)CCC	CCCCC(=O)CCCC	CCCCCC(=O)CCCC	CCCCCC(=O)CCCCC	CCCCCCC(=O)CCCCC	CCCCCCCC(=O)CCCCC	CCCCCCCC(=O)CCCCCC	CCCCCCCC(=O)CCCCCCC	O

In [4]:

raw_data.describe().T

raw_data.describe().T

Out[4]:

	count	mean	std	min	25%	50%	75%	max
temp_dist	12096.0	190.849120	101.579671	1.589174e+01	1.123273e+02	172.803162	263.550527	416.603500
temp_bott	12096.0	214.908201	101.659791	3.521194e+01	1.337667e+02	196.732403	290.248983	434.794647
temp_feed	12096.0	202.916131	100.331884	3.517856e+01	1.246547e+02	186.038635	276.903909	416.534837
duty_condensor	12096.0	-43006.023505	91269.101137	-4.961898e+05	-3.402247e+04	-14337.635300	-870.295447	-0.031283
duty_reboiler	12096.0	42892.224124	91875.576269	-7.060835e-01	7.637737e+02	14249.645500	34407.838675	497114.572000
flowrate_bott_A	12096.0	0.065716	0.077360	0.000000e+00	1.792839e-08	0.038984	0.126891	0.256620
flowrate_bott_B	12096.0	0.237842	0.181067	0.000000e+00	6.484256e-04	0.261879	0.404521	0.473678
flowrate_bott_C	12096.0	0.312911	0.160322	3.554722e-06	1.374662e-01	0.386805	0.424657	0.482577
flowrate_feed_A	12096.0	0.147181	0.069994	5.263158e-02	7.142857e-02	0.150685	0.200000	0.257143
flowrate_feed_B	12096.0	0.426410	0.034997	3.714286e-01	4.000000e-01	0.424658	0.464286	0.473684
flowrate_feed_C	12096.0	0.426410	0.034997	3.714286e-01	4.000000e-01	0.424658	0.464286	0.473684
flowrate_dist_A	12096.0	0.081737	0.079201	1.399720e-06	6.271408e-03	0.059441	0.146408	0.476189
flowrate_dist_B	12096.0	0.188571	0.182394	3.710545e-06	7.883367e-03	0.146078	0.387500	0.581166
flowrate_dist_C	12096.0	0.113224	0.158992	1.791002e-12	5.018738e-04	0.006774	0.268119	0.473675
ratio_reflux	12096.0	5.287344	7.952540	2.000000e-02	1.524571e-01	1.064687	6.199574	19.000000
ratio_boilup	12096.0	6.545480	8.810405	2.000000e-02	2.000000e-02	0.616441	19.000000	19.000000
pressure_atm	12096.0	8.942689	11.755571	2.000000e-01	8.380382e-01	3.285377	11.390028	29.000000
feed_pressure	12096.0	8.942690	11.755571	2.000010e-01	8.380392e-01	3.285378	11.390029	29.000001
vapor_fraction	12096.0	0.000000	0.000000	0.000000e+00	0.000000e+00	0.000000	0.000000	0.000000
no_stages	12096.0	25.000000	9.690028	1.200000e+01	1.850000e+01	25.000000	31.500000	38.000000
feed_stage	12096.0	5.000000	0.000000	5.000000e+00	5.000000e+00	5.000000	5.000000	5.000000

Time Benchmark¶

In [5]:

times = []
for i in range(0, 10):
    start_time = time.time()
    feed_stream = MaterialStream(stream_type="feed")(
        vapor_fraction=data['vapor_fraction'].to_list(),
        pressure=data['feed_pressure'].to_list(),
        molecules=dist_utils.get_name_columns(data).to_dict('list'),
        flowrates=dist_utils.get_flowrate_columns(data).to_dict('list'),
    )
    
    dist_col = Distillation(
        no_stages = data['no_stages'].to_list(),
        feed_stage =  data['feed_stage'].to_list(),
        pressure = data['pressure_atm'].to_list(),
        reflux_ratio = data['ratio_reflux'].to_list(),
        boilup_ratio = data['ratio_boilup'].to_list(),
        verbose=False
    )

    bott_stream, dist_stream = dist_col(feed_stream)
    
    times.append(time.time() - start_time)

ordered_data = dist_utils.sort_targets_by_weight(data, feed_stream._mw_idx)

average = np.mean(times).round(2)
std = np.std(times).round(2)

print(f"{len(data)} data pts take {average} +/- {std} seconds to predict.")

times = [] for i in range(0, 10): start_time = time.time() feed_stream = MaterialStream(stream_type="feed")( vapor_fraction=data['vapor_fraction'].to_list(), pressure=data['feed_pressure'].to_list(), molecules=dist_utils.get_name_columns(data).to_dict('list'), flowrates=dist_utils.get_flowrate_columns(data).to_dict('list'), ) dist_col = Distillation( no_stages = data['no_stages'].to_list(), feed_stage = data['feed_stage'].to_list(), pressure = data['pressure_atm'].to_list(), reflux_ratio = data['ratio_reflux'].to_list(), boilup_ratio = data['ratio_boilup'].to_list(), verbose=False ) bott_stream, dist_stream = dist_col(feed_stream) times.append(time.time() - start_time) ordered_data = dist_utils.sort_targets_by_weight(data, feed_stream._mw_idx) average = np.mean(times).round(2) std = np.std(times).round(2) print(f"{len(data)} data pts take {average} +/- {std} seconds to predict.")

12096 data pts take 3.22 +/- 0.37 seconds to predict.

Classifier Benchmark¶

In [6]:

ok_idx = np.array(ordered_data[ordered_data['Status'] == 'OK'].index)
dist_utils.plot_confusion_matrix(ordered_data, dist_col.status)

ok_idx = np.array(ordered_data[ordered_data['Status'] == 'OK'].index) dist_utils.plot_confusion_matrix(ordered_data, dist_col.status)

Out[6]:

0.9748175182481752

Flowrates Benchmark¶

In [ ]:

dist_utils.plot_flow(
    y_true=ordered_data.loc[ok_idx],
    y_pred=bott_stream.flow.loc[ok_idx]
)

dist_utils.plot_flow( y_true=ordered_data.loc[ok_idx], y_pred=bott_stream.flow.loc[ok_idx] )

Duty Benchmark¶

In [ ]:

dist_utils.plot_duty(ordered_data, dist_col, ok_idx)

dist_utils.plot_duty(ordered_data, dist_col, ok_idx)

Temperature Benchmark¶

In [ ]:

dist_utils.plot_temp(ordered_data, bott_stream, dist_stream, ok_idx)

dist_utils.plot_temp(ordered_data, bott_stream, dist_stream, ok_idx)

Mean Absolute Errors¶

In [ ]:

dist_utils.evaluate_flow(
    ordered_data.loc[ok_idx], 
    bott_stream.flow.loc[ok_idx], 
    metric='mae'
)

dist_utils.evaluate_flow( ordered_data.loc[ok_idx], bott_stream.flow.loc[ok_idx], metric='mae' )

Out[ ]:

	flowrate_bott_A	flowrate_bott_B	flowrate_bott_C	mean
0	0.006074	0.015565	0.012195	0.011278

In [ ]:

dist_utils.evaluate_duty(
    ordered_data, 
    dist_col, 
    ok_idx, 
    metric='mae'
)

dist_utils.evaluate_duty( ordered_data, dist_col, ok_idx, metric='mae' )

Out[ ]:

	condensor_duty	reboiler_duty	mean
0	4107.908326	4144.421101	4126.164713

In [ ]:

dist_utils.evaluate_temp(
    ordered_data, 
    bott_stream, 
    dist_stream, 
    ok_idx, 
    metric='mae'
)

dist_utils.evaluate_temp( ordered_data, bott_stream, dist_stream, ok_idx, metric='mae' )

Out[ ]:

	bott_temp	dist_temp	mean
0	6.339481	4.837448	5.588465