Residential%20 Electrical%20 Energy%20 Consumption%20 Profile%20in%20 Brazil

Residential Electrical Energy Overview - the Brazilian
Consumption Profile in Brazil Economy post -1994
Mônica Barros Brazilian government started in June 1994,
and economic plan (named “Plano Real”)
Reinaldo Castro Souza
that dramatically reduced monthly inflation
DEE, PUC-RIO from 80% to about 1%.

August 1997 Before the advent of “Plano Real”, lower
income classes had no protection against
daily inflation and currency devaluations,
since they had limited access to baking
services and products.
info@mbarros.com 1 info@mbarros.com 2

Overview - the Brazilian Overview - the Brazilian
Economy post -1994 Economy post -1994
At the onset of “Plano Real”, minimum This, together with a stable currency,
wage almost doubled in real terms (from caused a massive income transfer to the
roughly US$ 60 to US$ 100 monthly). poorest individuals in society.

The radical fall in inflation rates also Even though credit restrictions have been
contributed to increase, in real terms, the imposed by Brazil’s Central Bank and
disposable income of poor families, since interest rates are among the highest in the
now their money has the same purchasing world, access to credit is relatively easy,
power at the beginning or at the end of the especially in the electronic goods and
month. automotive sectors.

Overview - the Brazilian Overview - the Electric Sector
Economy post -1994 in Brazil
All of these factors, together with Most of the power plants are hydroelectric
increasing electronics imports, caused a plants, whose construction takes a very
substantial impact on electrical energy long period of time (around 10 years, in
consumption, especially in the residential some cases).
sector. The electric sector in Brazil has been
Electricity rates (which are still under going through dramatic changes since
government control) have been raised 1995.
above inflation rates, but this has not State controlled companies (energy
prevented consumption from experiencing producers and distributors) are being sold
unprecedented growth. to private groups.

Overview - the Electric Sector Overview - the Electric Sector
in Brazil in Brazil
The explosive growth in electrical energy A survey on residential electricity
consumption in Brazil for the past 3 years consumption habits and holding of
has made demand analysis fundamental electrical appliances was done in 1988.
for planning and control.
Due to technological advances and the
Several efforts are currently being made to economic changes just mentioned, this
create a residential consumer profile in 1988 research is obviously outdated.
different areas of the country.


in Brazil in Brazil
This presentation is part of an ongoing This profile will be able to identify
consulting project developed for electricity spending habits and aid in the
Eletrobrás, the Brazilian Electric Sector implementation of Demand Side
Holding Company. Management (DSM) policies.

The objective of this project is to create a Effective implementation of DSM policies
profile of residential consumers in all is crucial at this moment, since Brazil is
areas of Brazil. on the verge of an electrical energy
collapse, due to unexpected and
unprecedented consumption growth.

in Brazil in Brazil
Currently, we are in the process of In 1990, residential consumption
implementing surveys throughout Brazil. corresponded to 20% of total electrical
energy consumed.
Residential consumption is a major
concern for electrical power companies in In 1996, this participation grew to 27%,
Brazil, since its share in total consumption and in the years 2000-2002, it is estimated
has been growing fast since 1990. at 33%.


Overview - the Electric Sector Sampling Scheme used in the
in Brazil survey
The sample surveys currently in progress Due to the diversity in social and
are important for two reasons: economic indicators throughout the
Demand Side Management country, an ordinary sample plan based
Identification of factors that can serve on the number of residential consumers in
as explanatory variables in forecasting each town or city is not appropriate, even
models for residential consumption when analyzing individual states.

We propose an alternative sampling plan,
where stratification is based on clustering.


Sampling Scheme used in the Available Data for each town or
survey city
These clusters are created from the notion Total consumption
of an “electrical distance” which Average household consumption
compares consumption in each town with Total number of households whose
average values for each utility company. average monthly consumption falls into
each of the 10 categories: 0-30 KWh, 31-50
These clusters will serve as strata in a KWh, 51-100 KWh, 101-150 KWh, 151-200
stratified sampling procedure, in order to KWh, 201-300 KWh, 301-400 KWh, 401-500
reduce “within stratum” variance. KWh, 501-1000 KWh, above 1000 KWh.


We construct some additional
Cluster Analysis
variables, namely :
Standardized Consumption = total town Based on percentages of households in
consumption standardized so that the whole each of the 10 categories.
sample of towns in each state is a variable with
mean zero and variance one. We start the procedure by forming n
Electrical Distance = Euclidean distance clusters, where n is roughly 10 % of the
computed from the percentages of households number of towns in the state.
in each category for a give town ( in Algorithm used: Euclidean distances,
comparison with percentages for the entire
single linkage clustering.
state).
Percentages of households in each of the 10
consumption categories.


Case Study: COELCE Case Study: COELCE

COELCE is the energy distributor in the We start by forming 18 clusters, but
State of Ceará, in the Northeastern part of several of those contained less than 3
Brazil. towns or villages.
This clustering procedure is applied to all Thus, a sampling procedure based on
towns in the State, except for the capital each of these clusters would not be cost-
city (Fortaleza), which was subject to a efficient, which lead us to reduce the
separate survey. number of clusters used.
Most towns and villages in the state are This reduction is done until each cluster
characterized by very small average formed contains a “reasonable” number of
electricity consumption. towns.

Case Study : COELCE Case Study : COELCE
10 Clusters based on percentages of
In COELCE’s case we used 10 clusters, households in each category

but 6 of those consisted on 3 or less Cluster num_obs average average
towns, and were later condensed in 2 new 1 50
consumption
55.1
std. dev. distance std. dev.
4.2 0.22 0.03
clusters. 2
3
22
1
51
45.9
3.9
******
0.27
0.31
0.03
******
4 3 60.7 1.5 0.2 0.01
5 4 67 7 0.17 0.04
6 63 68.1 5.6 0.14 0.03
Moreover, the total number of households 7 23 87.4 6.9 0.07 0.02
8 2 72 0.21 0.14 0
in these small clusters is negligible, and 9 3 43.2 2.5 0.35 0.01
10 3 38.8 3.3 0.41 0.03
their combination doesn’t lead to
ENTIRE num_obs average std. dev. average std. dev.
significant losses in precision. SAMPLE consumption distance

174 63.6 13 0.18 0.08

info@mbarros.com 21

Case Study: Rio de Janeiro 5 geographical zones

We conducted a preliminary study in the South (19 neighborhoods) - most affluent,
city of Rio de Janeiro. but includes some shanty towns with
The basic aim was to identify similar totally different consumption patterns.
electricity consumption patterns among
154 neighborhoods that comprise the city. North (26 neighborhoods) - some areas
Originally, the city was divided into five are upper medium class, but generally
zones using a geographical criterion. lower consumption than on the south
Significant differences among each of the zone.
five zones are observed.


Descriptive Statistics - whole
5 geographical zones
sample
West (26 neighborhoods) - mixed, some
new residential areas but others with rural average std.dev. m um
inim m um
axim
characteristics. consumption 192.1 54.8 97 492
p0-50 13.9 5.6 5.0 33.9

Suburban (71 neighborhoods) - low p51-100 17.6 5.3 3.4 32.0
income areas, low energy consumption. p101-150 19.2 3.9 6.9 26.7
p151-300 34.5 6.9 13.4 48.8
Center (15 neighborhoods) - around p301-500 10.6 5.3 1.9 29.2
downtown, some low income p >501 4.2 6.3 0 52.6
neighborhoods.

info@mbarros.com 25

Case Study : Rio de Janeiro Average Consumption by zone
ZONA: centro
We consider only 6 categories of energy
300

consumo médio
250
Center
consumption, namely: 200

150

0-30 KWh
100

CAJU

COSME VELH
CENTRO

ESTACIO

FLAMENGO
CATETE

CIDADE NOV

GAMBOA

GLORIA

LARANJEIRA

MANGUEIRA

SANTA TERE

SANTO CRIS

SAUDE
CATUMBI
31-50 KWh
BAIRRO
51-100 KWh
ZONA: norte
101-150 KWh 400
consumo médio

350

151-300 KWh
300
250
North
200
301-500 KWh 150
100
ALTO B VIS

BANCARIOS

CACUIA

COCOTA

FREGUESIA

JD CARIOCA

JD GUANABA

MARACANA

PAQUETA

PORTUGUESA

RIBEIRA

S CRISTOVA

TAUA

TIJUCA

VILA ISABE
ENGENHO NO

GALEAO

MONERO
ANDARAI

C UNIVERSI

PCA BANDEI

RIO COMPRI

ZUMBI
GRAJAU

PITANGUEIR

PR BANDEIR
above 501 KWh
BAIRRO

info@mbarros.com 27

Average Consumption by zone Average Consumption by zone
ZONA: oeste
280
ZONA: sul
consumo médio

260
240
West 600

consumo médio
220
550
200
180 500
450
South
160 400
140 350
120 300
250

SANTISSIMO
ANIL

CAMPINHO
B.GUARATIB

COSMOS

GARD AZUL
CURICICA
FREGUESIA

GUARATIBA

INHOAIBA

P.GUARATIB

PACIENCIA
PCA SECA

PECHINCHA
S VASCONCE

SEPETIBA

TANQUE
TAQUARA
CAMORIM

CAMPO GRAN
CIDADE DEU

JACAREPAGU

SANTA CRUZ
200
150
100
BAIRRO 50
0

BARRA TIJU

JD BOTANIC

LEBLON

SAO CONRAD
RC BANDEIR

VARG.GRAND
BOTAFOGO
COPACABANA
GAVEA
HUMAITA
IPANEMA
ITANHANGA

JOA
LAGOA

LEME

ROCINHA

URCA

VARG.PEQUE
VIDIGAL
ZONA: suburbana
240
consumo médio

220
200
180
160
Suburban
140
120 BAIRRO
100
BARROS FIL

VIC.CARVAL

VILA PENHA
ABOLICAO
AGUA SANTA
B RIBEIRO

BONSUCESSO

COLEGIO

DEODORO
ENG DENTRO
ENG RAINHA
H GURGEL
INHAUMA
JACARE
JD AMERICA

MAG BASTOS
MANGUINHOS
MARIOPOLIS
OLARIA

PAVUNA
PENHA
PILARES
Q BOCAIUVA
RAMOS
LINS VASCO

RIACHUELO
ROCHA
SAMPAIO

VILA KOSMO
CACHAMBI
CAVALCANTI

VL VALQUEI
COSTA BARR

TODOS SANT
PADRE MIGU

TURIACU

BAIRRO

Cluster Analysis of Cluster Analysis of
neighborhoods neighborhoods
We base the cluster procedure on the a ve ra g e a v e ra g e
C lu s te r # o bs . a v e ra g e s td . d e v. % a b o ve % b e lo w
percentages of households in each of the c o n s u m p tio n 3 0 1 KW h 1 5 1 KW h
11 1 9 7 .0 **** 2 .4 8 4 .3
6 energy consumption categories. 2 8 1 1 8 .8 6 .5 4 .0 7 5 .1
4 28 1 4 9 .3 1 1 .1 7 .3 6 4 .0
We created 12 clusters, of which 6 contain 9
5
2
1
1 7 5 .0
1 8 6 .0
1 9 .8
****
1 5 .4
1 6 .1
6 0 .3
6 6 .9
only one neighborhood. 1 99 1 9 2 .7 2 4 .8 1 4 .1 4 7 .5
10 1 2 4 6 .0 **** 2 7 .1 4 6 .6
2 other clusters contain 2 neighborhoods 8
3
2
9
2 6 5 .0
2 9 7 .0
1 5 .6
1 9 .1
2 9 .2
3 6 .9
2 5 .1
3 1 .8
each. 7
12
1
1
3 9 1 .0
4 0 8 .0
****
****
5 5 .6
5 6 .1
1 6 .8
2 4 .7
6 1 4 9 2 .0 **** 6 4 .9 1 9 .8

W h o le
S a m p le 154 1 9 2 .1 5 4 .8 1 4 .8 5 0 .7

info@mbarros.com 31

Conclusions Conclusions

In both cases, the clustering procedure Moreover, some of the neighborhoods
results in groups that are much more singled out by the cluster procedure are
homogeneous than the entire sample. clear “outliers”, that is, do not represent
the entire population being sampled.
In the Rio de Janeiro case study, even in
cluster 1, which contains roughly 2/3 of In the Rio de Janeiro case study, clusters
the sample, there is a considerable 6, 7 and 12 represent very high income
reduction of variance, when compared areas of the city, as reflected by their
with the whole sample of neighborhoods. energy consumption levels.


Conclusions

Also, cluster 11 indicates the
neighborhood with lowest average
consumption among all 154 sampled.

Surprising as it might be, clusters 6
(highest average consumption) and 11
(lowest average consumption) are
geographically contiguous.

info@mbarros.com 35

Residential%20 Electrical%20 Energy%20 Consumption%20 Profile%20in%20 Brazil

Recommended

Recommended

More Related Content

Similar to Residential%20 Electrical%20 Energy%20 Consumption%20 Profile%20in%20 Brazil

Similar to Residential%20 Electrical%20 Energy%20 Consumption%20 Profile%20in%20 Brazil (20)

More from Monica Barros

More from Monica Barros (13)

Residential%20 Electrical%20 Energy%20 Consumption%20 Profile%20in%20 Brazil