Energy 70 (2014) 605e611

Contents lists available at ScienceDirect

Energy
journal homepage: www.elsevier.com/locate/energy

Criteria for smart grid deployment in Brazil by applying the Delphi

method
Joaquim J.M. Galo*, Maria N.Q. Macedo, Luiz A.L. Almeida, Antonio C.C. Lima
Department of Industrial Engineering Postgraduate, Federal University of Bahia, Street Aristides Novis, No. 2, 6th Floor, Federação, Salvador,
Bahia 40.210-630, Brazil

a r t i c l e i n f o a b s t r a c t

Article history: The deployment of smart grids has become a global trend in the most diverse electrical systems. This
Received 3 December 2013 strategy aims to increase the control and efficiency of electrical systems. However, this deployment
Received in revised form requires major changes to systems, for example, the information of a telecommunications infrastructure
18 March 2014
and the provision of adequate equipment, human resources and financial resources. In Brazil, which is
Accepted 14 April 2014
Available online 10 May 2014
territorially large and contains a power distribution system with centralised generation, the deployment
of smart grids tends to be time-consuming and could require decades to complete, as well as a large
monetary investment. Therefore, the elaboration and implementation of well-defined criteria are
Keywords:
Criteria
necessary for the planning steps of this deployment to minimise the investment cost.
Delphi method This article aims to develop a methodology for selecting electrical systems that will be implemented in
Deployment smart grids in Brazil by applying the Delphi method based on technical, financial and human resources
Smart grid and environmental applications. This methodology will be used to evaluate electrical systems in their
current operating conditions. The results obtained in this research are the creation of a priority factor to
determine the order of Smart grid deployment to achieve the best cost benefit in electrical systems
evaluated.
Ó 2014 Elsevier Ltd. All rights reserved.

1. Introduction
Smart grids represent an important evolution in the design and
operation of electrical systems because they integrate automation
solutions, different asset optimisations, process improvement,
monitoring and control, the integration of multiple sources of en-
ergy (renewable) and the development of new products and ser-
vices. This integration is already being implemented in several
countries via communication infrastructure systems, open stan-
dards and information technology.
The goal of this modernisation process is to provide more
dynamism in the networks of electrical systems and increase the
amount of available information to increase the transparency of
activities for both consumers and regulators. This process also aims
to enhance the quality and quantity of services offered to the
consumer through differentiated tariffs and dynamics that stimu-
late the optimisation of consumption. However, this process
* Corresponding author.
E-mail addresses: joaquimgalo@uol.com.br, joaquimgalo2@gmail.com
(J.J.M. Galo).
requires the installation of smart metres and increasing the reli-
ability and stability of electrical systems.
The complexity of electrical systems is positively correlated with
the challenges posed to all agents of a given system to be successful
in the deployment of smart grids in a country. However, the ex-
pected benefits for a company as a result of this new paradigm
compensate for the required effort during the initial
implementation.
In different countries, the majority of smart grid projects have
begun with pilot projects or deployment in specific areas. Evalu-
ating real cases before initiating global implementation is critical,
independent of the deployment strategy employed. The samples
should be representative, and the benefits offered should cover the
entire company to leverage further results. The integration of
generation, transmission, distribution and consumers must be
sought in gathering such results. All regulatory aspects and
necessary infrastructures must be considered in deploying smart
grid projects, and the involvement of consumers from the begin-
ning is essential for success [1e7].
This article proposes criteria for smart grid deployment in Bra-
zil’s electrical power system by applying the Delphi method and the
creation of a priority index. Section 2 presents aspects of smart grid

http://dx.doi.org/10.1016/j.energy.2014.04.033
0360-5442/Ó 2014 Elsevier Ltd. All rights reserved.
606 J.J.M. Galo et al. / Energy 70 (2014) 605e611
Fig. 1. Equivalent duration of interruption per consumer unit (DEC).
(Source: ANEEL [8]).
deployment in Brazil. Section 3 presents a survey of the variables
that can affect the deployment of a smart grid. Section 4 presents
the criteria for selection of the relevant variables and the repre-
sentation thereof as indicators. Section 5 discusses the application
of the Delphi method and the steps to be followed to evaluate the
electrical systems that must be established for smart grid deploy-
ment. Section 6 provides concluding remarks and
recommendations.
2. Aspects of smart grid deployment in Brazil
The Brazilian energy market is based predominantly on cen-
tralised generation. The electric energy sector has always been
essentially managed by the actions and initiatives of utilities and
regulators of the Regulatory Agency (ANEEL e Electric Energy Na-
tional Agency) and the national system operator (ONS).
In Brazil, the electric generation and transmission systems
employed by electric utilities are already supervised by automated
systems that use digital technology to monitor consumption pro-
cesses in virtually all major centres. These systems offer several
features such as tele-supervision, remote control and remote
sensing by an SCADA system (system control and data acquisition).
These systems indicate the operating conditions of all real-time and
automated systems that can track large volumes of cargo.
The reality of the distribution system in Brazil (which features a
voltage of less than 34.5 kV) is very different. Due to its complexity
and a high number of consumers (approximately 68 million), the
automated deployment of these systems is still in its infancy and
their management is still performed conventionally. For example,
power is measured manually in approximately 95% of consumer
units via electromechanical metres, which contributes to the
vagueness of measurements and poor monitoring of loads [8].
The graph in Fig. 1 shows the monitoring of an important indi-
cator of quality, Equivalent Duration of Interruption per Consumer
Unit (DEC)/year in Brazil.
The monitoring of this indicator is appropriate because it in-
dicates that the average time of interruption was 18.35 h in 2010
and 18.40 and 18.65 h in 2011 and 2012, respectively. This indicator
has grown in recent years, which suggests a need for greater system
control.
The current scenario of electric systems in Brazil presents the
following features:

Obsolescence of the assets of the distribution system.

High level of technical and non-technical losses.

Generation and transmission systems are automated (SCADA).

The equivalent duration of interruption per customer-annual
(DEC) has grown in recent years (a decrease was observed af-
ter the automation of generation and transmission).

System recovery is slow and precarious.

Slow modernisation of the networks and deployment of smart
grids.
One of the main factors that must be considered for smart grid
deployment in Brazil is the possibility of preventing fraud and
illegal connections. Each year, utilities lose approximately $5 billion
due to this problem [9]. Conversely, the need for high initial in-
vestments and the definition of clear government regulations are
barriers to the expansion of this market. However, this motivation
does not apply to the entire territory, which has resulted in the
heterogeneous deployment of pilot projects, various project types
and smart grid technologies.
The Electric Energy National Agency (ANEEL) and the Ministry of
Mines and Energy (MME), in conjunction with national dealers,
manufacturers, research institutes, etc., are seeking to create a
national smart grid model that meets the profile of the Brazilian
consumer and the existing infrastructure, with all of its virtues and
shortcomings.
Although the market still awaits a clear definition of regulatory
agencies to structure their offerings and set business plans, some of
the resources of the smart grid are currently being tested by utili-
ties in the form of pilot projects. The following are examples of such
projects being conducted in various Brazilian states:

Sete Lagoas (Minas Gerais): The CEMIG concession has been
conducting a project called “Cities of the future” since 2009,
which analyses the capacity and the benefits of the adoption of a
smart grid architecture based on tests of the electrical system of
the city. The project examines the feasibility of expansion for the
concession area of CEMIG and evaluates the products, services
and solutions available in the market. New metres and modern
 J.J.M. Galo et al. / Energy 70 (2014) 605e611
telecommunications systems are being installed to allow for the
local dealership to monitor the consumption levels of the pop-
ulation and identify any power interruptions in real time [10].

Aparecida (São Paulo): The EDP Group company Bandeirante
(EDP Energias of Portugal Group) is implementing a pilot project
called INOVCITY (Social innovation and mobility electric). With
an expected investment of $10 million, the company foresees
the feasibility of a set of technologies that will enable greater
efficiency and quality in the provision of customer services,
including smart metering, efficient public lighting, micro-
generation with renewable energy sources and electrical
mobility, as well as a set of energy efficiency actions and edu-
cation of local communities. Moreover, the CPFL Group is
invested in the area of network automation and application of
intelligent network concepts in the city of São Paulo [11,12].

Buzios (Rio de Janeiro): The Ampla concession has an ongoing
project that primarily aims to rationalise consumption and
reduce waste with electricity in general to make the city sus-
tainable, rational and efficient. Various investments are involved
to affect these smart improvements in Buzios. The total invest-
ment was calculated for an average initiative of R $40 million
must end in 2014. Investments were made in transmission lines
and power distribution, the installation of smart metres and
generating electricity from more sustainable sources, such as
wind and solar. In addition, LED lamps were installed for low
consumption at several different points of the city [13,14].

Parintins (Amazonas): The Eletrobrás Group has an ongoing
project in which sensors have been installed at various points in
the distribution network, which allow for the immediate
intervention of the supplier in the event of system crashes.
Furthermore, this project allows for the monitoring of data
provided by metres, such as time consumption and cost of kWh
at different times of the day. The utility can operate at different
rates throughout the day to stimulate consumption outside the
peak hours of demand. Eletrobras invested R$21 million in the
replacement of all metres, the automation of feeders, the mea-
surement and monitoring of distribution and the establishment
of a system to improve the use of a 25-MW generator in the park
of Parintins. All information collected in the field will be sent to
a central intelligence that will be able to intervene in real time to
correct defects and failures of the system and interact with the
consumer [15].
In other states of the country, concessions are exchanging
electromechanical metres for digital metres and are developing
studies to implement several pilot projects.
3. Survey of the variables that can affect the deployment of
smart grid
Most sources agree that smart grids must be implemented in
places where they add value to the electric system via productivity
gains and improvements in energy efficiency. One of the main gains
must be in the area of operational efficiency, which includes im-
provements in the control of life assets, facilitating the identifica-
tion of faults or defects, avoiding unnecessary offsets and the lower
loading of the system, among others. In the area of energy effi-
ciency, the reduction of technical and non-technical losses will
enable a more effective distribution of energy because the gener-
ation is up to the final customer.
The practices involved in the deployment of smart grid archi-
tecture and the results that will be achieved are strategic and of
interest to various parties, including but not limited to electrical
power utilities, regulators, consumers, public authorities, suppliers,
research institutions, funders and development agents.
607
Brazil does not possess the resources for full deployment in the
short term. Thus, deployment should occur in steps that are plan-
ned in detail over an extended period. The first electrical systems
selected to initiate the deployment process should be evaluated
carefully to seek a smaller investment for the expected benefit
because the continuity of the process is necessary for the genera-
tion of new resources. For example, an indiscriminate exchange of
all metres would not be productive because the electrical networks
of Brazilian distribution share a poor telecommunications system
and obsolete assets based on the current infrastructure. Thus, this
indiscriminate change would incur high investments and small
expected returns.
ANEEL has sought to create a national model that meets the
profile of Brazilian consumers and the existing infrastructure by
considering its virtues and shortcomings, such as fraud in the
supply. The path proposed by this model is long and must be well
planned to involve all actors in the process.
A survey of the current conditions of electrical systems is
necessary to select the variables involved in smart grid
deployment. This survey examines technical considerations,
economic characteristics, installation locations, available human
resources, environmental conditions, potential partnerships and
the socio-economic conditions of the consumers, as listed below
[16e45]:
1. Technical characteristics

Importance of the loads of the system;

The system load factor;

System power factor;

System load level;

Level of technical losses;

Level of non-technical losses (commercial);

Electric system infrastructure (automation, protection and
measurement);

Telecommunications and information infrastructure;

Failure or defect rate;

Technical resources for operation;

Technical resources for maintenance;

Asset lifetime (ageing);

Potential for the development of energy efficiency programs;

Inclusion potential of distributed generation sources (wind,
solar and biomass);

Critical connection or border (important).
2. Economic

Invoicing of the electrical system;

Supply cost (kW);

Cost of maintenance;

Cost of operation.
3. Installation location

Underground installation;

Distance from the centre of operation;

Distance from research centres;

Borders other concessionaires;

Region with government social policies;

Region with historic cities;

Region with tourist towns (coastline, natural beauties, wa-
tering place, etc.);

Region with natural resources (agriculture, water, minerals);

Region with interests of large companies;

Region with harsh weather conditions;

Region with forecasted growth (demographic, financial
investment);
4. Human resources

Availability of skilled professionals in the area;

Technical staff with potential for training in the area.
608 J.J.M. Galo et al. / Energy 70 (2014) 605e611
5. Environmental conditions

Fossil fuel energy sources (greenhouse gas emissions, carbon
footprint);

Regional climate conducive to natural disasters for the electric
system;

Environmental incentives program.
6. Partnerships

Research institution (university, research centre);

Telecommunications operator;

Manufacturers of equipment or related area;

Software developers;

Electric utilities (national or international with projects
developed in the area).
7. Socioeconomic conditions of consumers

Purchasing power of the community;

HDI of the community;

History of involvement in community projects;

Region with social programs in progress;

Community dedicated to the incorporation of new services.
8. *Special cases

New Installations;

Existing installation update (reform).
4. Criteria for selecting the relevant variables and their
representation as indicators
Due to the high investment, long time and rapid technological
developments required, a smart grid must be deployed in well-
planned steps. The current operating conditions of electrical sys-
tems should be studied and evaluated to select the best deploy-
ment sequence based on the assumption that system investment
will be lower than the expected benefit.
This article proposes a method for selecting electrical systems to
be deployed with a lower cost per benefit expected via the calcu-
lation of an index that will determine the priority order of smart
grid deployment in the analysed systems. Fig. 2 shows the various
steps that must be followed in the deployment process. The first
step consists of the preparation of guidelines and criteria to be
followed because the deployment process is usually long and must
also be constantly re-evaluated; thus, certain updates will be
necessary [16e45].
The most relevant variables should be selected from the vari-
ables mentioned in Section 3, and their number should be reduced
to select only the most significant ones for the deployment process.
These variables should be correlated with the remaining variables.
Fig. 2. Steps for smart grid deployment.
The next step consists of the representation of variables as in-
dicators, from which quantification can be established (attribution
of comparative values for selection).
An indicator provides significant information about a particular
area of interest and adequately represents a given situation [32]. An
indicator corresponds to a selected variable and can primarily
synthesise a complex set of information to retain only the essential
meaning of the analysed aspects [46,47].
The indicator term is a selected parameter that is considered in
isolation or in combination with others to reflect the conditions of a
system. Normally, an indicator is used as a pre-treatment for
original data.
The selected indicators are diverse to provide a wide scope for
the evaluation of systems because the implementation relies on
strengths (such as electric system infrastructure) and weaknesses
that can be improved by the implementation of a smart grid (level
of losses).
Based on these criteria, 20 indicators and 1 factor of priority
(compensation factor in priority projects) were determined:
1. Importance of system (load);
2. Load factor of system;
3. Level of losses of the system;
4. Loading of system and/or low power factor;
5. Forecast growth of load;
6. Electric system infrastructure;
7. Telecommunications infrastructure;
8. Failure rate or defect of the system;
9. Resources for operation and maintenance;
10. Ageing of assets of the system;
11. Potential of energy efficiency;
12. Potential of distributed generation;
13. Environmental impacts of sources that feed the system;
14. Adverse climates and prone to natural disasters;
15. Availability of human resources in the area;
16. Distance of the centre of operation and research centres;
17. High HDI or community with involvement in similar
projects;
18. Partnership with institutions (research centre, university,
telecom operator);
19. System based on the results of research and development
projects in the area;
20. *Special case.
*Special case indicator

New installation will be implemented to greatest extent possible
in accordance with new technologies while observing the
fundamental premise of a minor investment for the designed
benefits.

Existing installation (update) will be parsed if the deployment
will be a complete update of the assets or a partial adaptation.
4.1. Definition of the priority factor
Several localisation variables, such as underground installation,
cities (historical, tourism, natural resources), are of interest to large
companies with government social programs and investment
growth prospects. These variables will not be represented as in-
dicators, and a “Yes” value will be gathered as a factor, which is
called the priority factor (Kj), to compensate for priority projects. Kj
should have a maximum value of 1.2, i.e. 20%, to avoid masking the
weight of the other variables involved. In other applications,
Kj ¼ 1.0.
 J.J.M. Galo et al. / Energy 70 (2014) 605e611 609

After relevant indicators have been defined, the next step will
create an index called index deployment (IP). This value can be
calculated by applying a weighted average to 20 indicators and
using the priority factor to compensate for the priority projects, as
indicated by the following equation:

Kj X
N
IP ¼ Wi $Ii (1)
N
i¼1

where Ii are the indicators involved in the assessment; Wi is a

weighting factorda value used to determine the weight of an in-
dicator in the consolidation of the index. The factor quantifies the
degree of importance of an indicator; N is the total number of in-
dicators used in the evaluation (20 indicators in this article); and Kj
is the priority factor for priority projects.
The weighting factor (Wi) and priority factor (Kj) should be
determined by the application of the Delphi method, which will be
described in Section 5.

5. Delphi method and the steps to be followed to evaluate the

electrical systems

The Delphi method is a technique used to find a consensus of

opinions among a group of experts about future events. The tech-
nique was first used 60 years ago when it was developed by re-
searchers Olaf Helmer and Norman Dalker at the Rand Corporation.
The Delphi method is a very useful technique, especially when a
qualitative analysis of a market is desired, because it allows for the
design of future trends, such as technological discontinuities, socio-
economic changes or planning future events. The procedure is
simple: the manager is highly efficient and generates complete
individual ideas via the assumption that the collective consensus is
well defined and parameterised to take advantage of all of the
synergy of the discussion group.
The Delphi method is based on the principle that the predictions
made by a structured group of specialists are more precise
compared to those made by unstructured groups or individuals.
The technique can be adapted for use in face-to-face meetings, in
which case it is referred to as the mini-Delphi or Estimate-Talk-
Estimate (ETE) method. The Delphi method has been widely used
for business forecasting and has certain advantages over other
approaches in making structured predictions in market situations.
The internet has been a significant asset in the Delphi methodology
for promoting group discussions and thus reaching a consensus
[48e50].
The application of the Delphi method has the following ad-
vantages and disadvantages:

Advantages e the technique is suitable for obtaining consensus
among experts, and the process can be performed virtually and
focus on work details or on an entire project.

Disadvantages e the technique is a time-consuming process and
requires skill to interpret the reviews.
Despite the Delphi method be suitable to achieve the consensus
among experts, it has the limitation of being time consuming and
requires skill to interpret the opinions of experts from various fields
with multiple interests.
Fig. 3. Flowchart representation of the Delphi method.
The steps constituting the Delphi method in determining factors
can be described as follows:
Step 1. Initial meeting (in-person or virtual): Discussion of the
method, the objectives of the work, the context, the deadlines
and the calendar of meetings.
Step 2. Definition of two teams to carry out the work and the
selection of their respective coordinators:

Team 1: General team with the participation of experts in
several areas of the company;

Team 2: Team of experts for the smart grid deployment
process.
Step 3. The elaboration of a project to implement the work:

Definition of the project, complete with job descriptions and
descriptions of the method used, the resources, the schedule, etc.

A form containing the list of indicators (with space for new
suggestions) with fields to record the suggestions of the
respective values of weights (weighting factor) as well as a
field for the suggestion of the priority factor.
Suggestions for setting the values of the weighting and priority
factors:
Weighting factor (Wi): this value should be an integer ranging
from 1 to 5 (wide range for sorting) according to the degree of
importance attached to the indicator in smart grid deployment.
Compensation factor of priority projects (Kj): as defined in
Section 4, this value should not be too high to avoid interfering with
the comparison between systems.

5.1. Application of the Delphi method to determine factors Step 4. Gather the forms, compile and statistically analyse the
received data (discard the extreme values and apply arithmetic
A flowchart representing the application of the Delphi method mean) and subsequently condense the information into a single
to determine factors is presented in Fig. 3. form without identifying the participants.
610 J.J.M. Galo et al. / Energy 70 (2014) 605e611
Table 1
Matrix for calculating the priority index.
Step 5. Send the new form to all experts requesting a review and
add any new suggestions.
Step 6. Repeat step 4 with the new values to seek a consensus. If
necessary, repeat steps 5 and 6.
Step 7. Prepare a report listing all of the processes performed,
and present the results and targets for implementation. If dif-
ferences arise, new rounds should be performed.
5.2. Priority index calculation applied to the selection of electrical
systems for smart grid deployment
After the consolidation of the list of indicators and the estab-
lishment of the weighting and compensation factors, the deploy-
ment team must generate a table containing this information. The
next step will be to assess the system (selection of electrical sys-
tems) by calculating the priority index.
Verifying that all 20 evaluated indicators apply to the system
(none may be appropriate to the particular system) will be
important in the evaluation of each specific electrical system.
If the indicator applies to a given system, it will then be classi-
fied as low, medium or high (as described below). If it does not
apply, the indicator will be nullified.

Low: If the value is between 0 and 1;

Medium: If the value is between 1.1 and 2;

High: If the value is between 2.1 and 3.
All values are rounded to one decimal place.
The IP for deployment will be calculated using equation (1).
Specifically, the countervailing factor will be multiplied by the
weighted media of the indicators. Table 1 shows an example of this
calculation (array for the calculation of the priority index).
When evaluating various electrical systems, the system that
obtains the highest priority index is best suited to initiate the
deployment of the smart grid.
Detailed information will be important in evaluating each sys-
tem, such as the name of the system, the coding and the substation
and the regional centre of operation to which the system belongs.
After the evaluation of the first electrical systems by the specific
deployment group, the validation of the process will be important.
The results (condensed in a table e Table 1) should be forwarded to
the participants of all groups to allow for new suggestions.
6. Conclusions
The smart grid deployment process implies a profound trans-
formation in energy companies, involving major challenges that
must be faced via the development of a well-defined methodology
based on criteria that must seek greater efficiency and a lower cost
per expected benefit.
Brazil does not possess the resources for a full deployment in the
short term. Thus, this deployment should occur in steps that are
planned in detail over an extended period. The deployment process
should be carefully and continuously evaluated because the conti-
nuity of the process will be necessary for the generation of new
resources.
The choice of the Delphi method introduces an opportunity to
improve the method based on the discussions and experiences of
experts. These experts may incorporate other variables and review
the values of the weighting factors and priority throughout the
deployment process.
Despite the Delphi method be suitable to achieve the consensus
among experts, it has the limitation of being time consuming and
require skill to interpret the opinions of experts from various fields
with multiple interests.
 J.J.M. Galo et al. / Energy 70 (2014) 605e611
The method proposed in this article should serve as a tool to
assist managers of energy companies in making decisions using
criteria that may select systems that feature a lower investment per
expected benefit.
References
[1] DOE e Department of Energy. The smart grid: an introduction,2008. Available
at, http://www.oe.energy.gov/1165.htm [accessed December 2009].
[2] EPRI report. Methodological approach for estimating the benefits and costs of
smart grid demonstration projects. Palo Alto, California 94303e0813, USA:
EPRI; January 2010.
[3] Blumsack S, Fernandez A. Ready or not, here comes the smart grid! Energy
2012;37:61e8.
[4] Faruqui A, Hiledik R. Piloting the smart grid. Electr J Aug./Sept. 2009;22(7).
Available at, http://www.smartgridnews.com/artman/uploads/1/Piloting_the_
smart_grid__05-29-09.pdf [accessed December 2009].
[5] Faruqui A, Sergici S, Sharif A. The impact of informational feedback on energy
consumptionda survey of the experimental evidence. Energy 2010;35(4).
[6] Spees K, Lave LB. Impacts of responsive load in PJM: load shifting and real time
pricing. Energy J 2008;29(2).
[7] Lund H, Andersen AN, Østergaard PA, Mathiesen BV, Connolly D. From elec-
tricity smart grids to smart energy systems e a market operation based
approach and understanding. Energy June 2012;42(1):96e102.
[8] ANEEL. Base de dados Setor Elétrico. Site, www.aneel.gov.br; 2013.
[9] Redeinteligente.com. Available at:http://www.redeinteligente.com/2012/06/
30/smart-grid-evolucao-do-mercado-no-brasil/, ; 2012 [accessed June 2012].
[10] Cemig.com. Available at:http://www.cemig.com.br/pt-br/A_Cemig_e_o_
Futuro/sustentabilidade/nossos_programas/Redes_inteligentes/Paginas/as_
redes_inteligentes-na_cemig.aspx, ; 2012 [accessed April 2012].
[11] EDP Bandeirante.com. Available at:www.edpbandeirante.com.br/energia/
empresa/inovcity/inovcity.aspx, ; 2012 [accessed March 2012].
[12] Smartgridnews.com.br. Available at:http://smartgridnews.com.br/inovcity-
lancamento-de-projetos-de-inovacao-social-e-mobilidade-eletrica/, ; 2012
[accessed May 2012].
[13] Redeinteligente.com. Available at: http://www.redeinteligente.com/2013/01/
12/buzios-a-primeira-cidade-inteligente-da-america-latina; 2013 [accessed
January 2013].
[14] Ampla.com. Available at:www.conscienciaampla.com.br/tag/cidade-
inteligente-buzios/, ; 2013 [accessed February 2013].
[15] Amazonasenergia.gov.br. Available at:http://www.amazonasenergia.gov.br/
cms/eletrobras-lanca-projeto-parintins, ; 2011 [accessed June 2011].
[16] ANEEL e PRODIST. Procedimentos de Distribuição. Site, www.aneel.gov.br;
2012.
[17] European Commission. European technology platform smart grids. Vision and
strategy for Europe’s electricity networks of the future. Available at, http://ec.
europa.eu/research/energy/pdf/smartgrids_en.pdf; 2006 [accessed January
2011].
[18] Gorgette F, Devaux O, Fraisse JL, CIRED. Possible roadmaps for new re-
quirements for French distribution control and automation; 2007.
[19] MME e Ministério de Minas e Energia, Portaria n 440, Disponível em:http://
www.mme.gov.br/mme/galerias/arquivos/noticias/2010/Port_440_Rede_
Inteligente. pdf, [acesso em: Junho 2010].
[20] Albright RE. A unifying architecture for roadmaps frames a value scorecard.
New Jersey, USA: Albright Strategy Group; 2004.
[21] Bose A. Smart transmission grid applications and their supporting infra-
structure Available at: http://certs.lbl.gov/pdf/smart-transmis sion-grid.pdf.
[22] Heidt GT, et al. Professional resources to implement the smart grid. In: Pro-
ceedings of the 2009 North American power symposium (NAPS), Starkville,
Mississippi, USA, October 4e6; 2009.
[23] IEA (International Energy Agency). Technology roadmap: smart grid. Paris:
OECD/IEA; 2011.
[24] IEC, Smart grid standardization roadmap, Edition 1.02010, Smart Grid Stra-
tegic Group (SG3).
611
[25] IEC. Intelligrid methodology for developing requirements for energy systems;
2007. Publicly available specification (PAS).
[26] NEMA. Smart grid standards publication SG-AMI 1-2009 e requirements for
smart meter upgradeability. Available at, http://www.nema.org/stds/sg-ami1.
cfm; 2009.
[27] NIST. Framework and roadmap for smart grid interoperability standards.
Release 1.0, NIST Special Publication 1108, January 2010. Available at: http://
www.nist.gov/public_affairs/releases/smartgrid_interoperability_final.pdf;
2010.
[28] NYS. Smart grid consortion, smart grid roadmap for the state of New York;
2010. PEA smart grids team, 2010. Framework and directions of PEA smart
grids development, internal document of PEA [in Thai].
[29] U.S. Department of Energy. Smart grid system report. Available at: http://
www.oe.energy.gov/DocumentsandMedia/SGSRMain_090707_lowres.pdf;
July 2009.
[30] Southern California Edison. General rate case smart grid strategy & roadmap
www.sce.com/smartgrid; 2010.
[31] Finn P, Fitzpatrick C, Connolly D, Leahy M, Relihan L. Facilitation of renewable
electricity using price based appliance control in Ireland’s electricity market.
Energy 2011;36(5).
[32] Podmore R, Robinson MR. The role of simulators for smart grid development.
IEEE Trans Smart Grid 2010;1(2).
[33] Myrda P, Donahoe K. The true vision of automation. IEEE Power Energy Mag
May/June 2007;5(3):32e44.
[34] Ayres RU. Limits and possibilities of large-scale long-range societal models.
Technol Forecast Soc Change 1984;25:297e308.
[35] Coates JF, Mahaffie JB, Hines A. Technological forecasting: 1970e1993. Tech-
nol Forecast Soc Change 1994;47:23e33.
[36] Martino JP. Technological forecasting for decision making. 3rd ed. New York:
McGraw-Hill Inc.; 1993.
[37] Chowdhury SP, Crossley P, Chowdhury S. Microgrids and active distribution
networks. London: IET; 2009.
[38] Masiello Ralph, Nispen Hugo van, Wilhite Robert. Utility of the future: di-
rections for enhancing sustainability, reliability and profitabilityvol. 1;
September 2008, ISBN 978-0-6152-7035-7.
[39] Hledik R. How green is the smart grid. Electr J 2009;22(3).
[40] King D, Morgan MG. Customer-focused assessment of electric power micro-
grids. J Energy Eng 2007;133(3).
[41] Walawalkar R, Fernands S, Thakur N, Chevva K. Evolution and current status
of demand response (DR) in electricity markets: Insights from PJM and NYISO.
Energy 2010;35(4).
[42] Faruqui A, Hajos A, Hledik R, Newell S. Fostering economic demand response
in the Midwest ISO. Energy 2010;35(4).
[43] Lund H, Mathiesen BV. Energy system analysis of 100% renewable energy
systems e the case of Denmark in years 2030 and 2050. Energy 2009;34(5):
524e31.
[44] Maarten W. The research agenda on social acceptance of distributed gener-
ation in smart grids: renewable as common pool resources. Renew Sustain
Energy Rev 2012;16(1):822e35.
[45] Sanseverino ER, Di Silvestre ML, Ippolito MG, De Paola A, Lo Re G. An
execution, monitoring and replanning approach for optimal energy manage-
ment in microgrids. Energy 2011;36(5):3429e36.
[46] Guindani S. Indicadores de Sustentabilidade em Agroecossistemas: Uma
Análise [Dissertação Mestrado]. Florianopolis, SC: UFSC; 1999.
[47] Mitchell G. Problems and fundamentals of sustainable development in-
dicators. Disponível on-line, http://www.lec.leeds.ac.uk/peope/Gordon.html;
1997 [acesso em; abril 2011].
[48] Giovinazzo RA, Fischmann AA. Delphi Eletrônico e Uma Experiência de Uti-
lização da Metodologia de Pesquisa e seu Potencial de Abrangência Regional.
Trabalho apresentadado no XIV Congreso Latinoamericano de Estrategia, 17e
19 de maio de 2001, Buenos Aires, Argentina.
[49] Estes GM, Kuespert D. Delphi in industrial forecasting. Chem Eng News; Aug.
1976:40e7.
[50] Turoff Murray, Linstone Harold A. The Delphi method. New York: Addison
Wesley Publishing Company Inc.; 1975.