Brussels. Step 1.e Micro-level Non-Residential model¶

In [1]:

import datetime; print(datetime.datetime.now())

2018-03-26 02:14:48.583647

Notebook abstract

A simple micro-level building stock model. The consumption model defined for the building stock works in theory exactly like the other micro level consumption model. The difference between this model and the income, electricity and water demand models is that we don’t have a micro-level consumption data set in order to extract regression coefficients. In order to define a consumption model we use predefine building typologies.

Prior non-residential model¶

In [2]:

import pandas as pd

In [3]:

TypesB = pd.read_csv('data/TypesB.csv', index_col=[0,1])

In [4]:

TypesB

Out[4]:

		heat	cool	elec	sqm	entities	p
group	typ
comm	Services	114.646120	42.900227	85.453653	3600.0	7610	0.383472
	Food services	820.440000	125.188571	298.847619	525.0	3020	0.152179
	Lodging	79.161081	23.435315	74.837117	11100.0	259	0.013051
	Office building	55.291667	31.980083	92.729167	12000.0	8099	0.408113
	Warehous/Storage	101.099570	1.860645	41.330108	4650.0	130	0.006551
indu	Chemical - Pharmaceutical/Medical	183.492971	9.645261	198.687800	105.0	40	0.002016
	Fabricated metal products	33.160702	1.528521	228.669574	105.0	95	0.004787
	Food	483.129187	275.699073	255.629133	39.0	362	0.018241
	Furniture	39.597494	2.899449	172.696441	105.0	45	0.002268
	Machinery	27.536797	24.687100	230.904416	105.0	78	0.003930
	Plastic/Rubber products	31.946667	1396.809143	516.179810	105.0	18	0.000907
	Textiles - Apparel	82.170370	16.185185	243.263426	36.0	89	0.004485

In [5]:

sub_typ = TypesB.loc[:, ['heat', 'cool', 'sqm', 'entities']]

In [6]:

sub_typ_sum = sub_typ.loc[:, ['heat', 'cool']].mul(sub_typ.sqm.mul(sub_typ.entities), axis=0).sum()

In [7]:

sub_typ_sum.div(sub_typ_sum.sum())

Out[7]:

heat    0.689349
cool    0.310651
dtype: float64

In [8]:

input_sd = 0.01

In [9]:

nrb_elec = pd.DataFrame(columns=['co_mu', 'co_sd', 'p', 'dis', 'lb', 'ub'])

In [10]:

nrb_elec.loc['BuildingSqm', 'co_mu'] = ",".join([str(i) for i in TypesB.loc[:, 'sqm']])
nrb_elec.loc['BuildingSqm', 'co_sd'] = ",".join([str(i * input_sd) for i in TypesB.loc[:, 'sqm']])
nrb_elec.loc['BuildingSqm', 'dis'] = "Deterministic;n;Categorical"

nrb_elec.loc['BuildingHeat', 'co_mu'] = ",".join([str(i) for i in TypesB.loc[:, 'heat']])
nrb_elec.loc['BuildingHeat', 'co_sd'] = ",".join([str(i * input_sd) for i in TypesB.loc[:, 'heat']])
nrb_elec.loc['BuildingHeat', 'dis'] = "Deterministic;BuildingSqm;Categorical"

nrb_elec.loc['BuildingCool', 'co_mu'] = ",".join([str(i) for i in TypesB.loc[:, 'cool']])
nrb_elec.loc['BuildingCool', 'co_sd'] = ",".join([str(i * input_sd) for i in TypesB.loc[:, 'cool']])
nrb_elec.loc['BuildingCool', 'dis'] = "Deterministic;BuildingSqm;Categorical"

nrb_elec.loc['BuildingElec', 'co_mu'] = ",".join([str(i) for i in TypesB.loc[:, 'elec']])
nrb_elec.loc['BuildingElec', 'co_sd'] = ",".join([str(i * input_sd) for i in TypesB.loc[:, 'elec']])
nrb_elec.loc['BuildingElec', 'dis'] = "Deterministic;BuildingSqm;Categorical"

nrb_elec.loc[:, 'p'] = ",".join([str(i * input_sd) for i in TypesB.loc[:, 'p']])
nrb_elec.loc[:, 'lb'] = 0

In [11]:

nrb_elec.to_csv('data/table_elec_nr.csv')

In [12]:

nrb_elec

Out[12]:

	co_mu	co_sd	p	dis	ub
BuildingSqm	3600.0,525.0,11100.0,12000.0,4650.0,105.0,105....	36.0,5.25,111.0,120.0,46.5,1.05,1.05,0.39,1.05...	0.00383471907281,0.00152179390275,0.0001305114...	Deterministic;n;Categorical	NaN
BuildingHeat	114.64612041392596,820.44,79.16108108108108,55...	1.1464612041392597,8.204400000000001,0.7916108...	0.00383471907281,0.00152179390275,0.0001305114...	Deterministic;BuildingSqm;Categorical	NaN
BuildingCool	42.90022677542567,125.18857142857142,23.435315...	0.4290022677542567,1.2518857142857143,0.234353...	0.00383471907281,0.00152179390275,0.0001305114...	Deterministic;BuildingSqm;Categorical	NaN
BuildingElec	85.45365281064835,298.84761904761905,74.837117...	0.8545365281064835,2.9884761904761907,0.748371...	0.00383471907281,0.00152179390275,0.0001305114...	Deterministic;BuildingSqm;Categorical	NaN