At present, around 1 billion people, mostly in rural areas,
lack access to electricity. In most of these areas, extending the
national electricity grid is truly expensive. Alternatively,
electrification systems based on renewable energy sources (wind and
solar) are a suitable option for providing electricity to these isolated
communities.
What is this website about?
This website offers a decision-support tool, which combines
combinatorial optimisation and multicriteria techniques, to help
electrification developers (governments, enterprises, NGOs) to design
efficient and socially-suited projects. The algorithm provided by the
tool faces two problems regarding rural electrification: sizing and
siting. The first one consists in defining the size of generators and
storage capacity to meet the demand according to the availability of
energy resources. The second problem focuses on siting generation and
storage equipment according to the variability of energy resources and
the location of demand over the target area. Therefore, questions such
as which the best point of the community is for the generation equipment
to be set, which type of generation source will be appropriate in each
case and which points should be connected to others in a microgrid, will
be analysed and solved. Thus, the final result shows an optimized
electrification system regarding the use of equipment and considering
both individual users and microgrids.
The proposed decision-making process allows an easy and intuitive
interaction with the user in order to progressively take decisions
according to their relevance and aiming to adequate the final solution
to the specific characteristics and needs of the target community.
Social constraints regarding the management of the systems can be introduced by the user. Further information about the solving model is available in the correspoding tab.
How does it work?
The decision-making process is organised into three
decision levels, ordered according to the importance of the decisions
taken. This is why the user has the chance to select which solutions
showed at one level better fit the target community needs. The number of
solutions showed at each level depends on the combinations proposed by
the user, as explained next. Finally, in order to rank the solutions
from a level, some criteria (to be weighted) are taken into account. The
purpose of each decision level and the criteria and subcriteria
considered are the following.
Required Data
First Level
Second Level
Initial parameters
Nominal voltage of the microgrid [V]
Maximum voltage drop allowed in the microgrid [%]
Solar panels
Peak sun-hour [h]: solar energy available in
an area during a typical day. In other words, the number of daily hours
at an equivalent radiation of 1 kW/m2.
Average temperature of the coldest month [ºC].
Maximum power of each panel type [W]
Cost of each panel type
Solar controllers
Maximum power of every type of solar controller (in W)
Cost of every solar controller
Wind turbines
Cost of each wind turbine type
Energy potential of each wind turbine type [Wh/day], this value can be specified for each point.
Batteries
Battery efficiency
Maximum discharge factor allowed
Equivalent capacity of each battery type [W]
Cost of each battery type
Power Inverters
Inverter efficiency
Maximum power of each inverter type [W]
Cost ofeach inverter type
Meters
Cost of each meter type
Lines
Cost of each line type (including support infrasructure)
Resistance of each line type [Ω/m]. If you are not using earthing (or grounding) system, introduce twice this value
Maximum allowed intensity of each line type [A]
Consumption points
Coordinates (x,y,z)
Demand of energy [Wh/day], Power [W] and autonomy [days]
Non-consumption points
Coordinates (x,y,z)
First level: Strategic decisions
At
the first level, several electrification solutions are generated
minimizing the cost for a set of different demand scenarios. Such
scenarios can be introduced by defining percentage increases in the
energy, power and autonomy demand, with regards to the basic demand
previously introduced. Thus, a solution is generated for the basic
values as well as for each combination of increases.
The solutions generated are ranked according to two
criteria (cost and demand) and three sub-criteria (energy, power and
autonomy). Default weights obtained through surveys to rural
electrification experts are used for a first ranking of the solutions.
However, you can adjust the weights according to the characteristics of
each studied community, so as the ranking better fits your preferences.
Once satisfied with the ranking shown, you can select some solutions for further analysis in the second level.
Second level: Tactical decisions
At
the second level, several electrification solutions are generated
minimizing the cost for a set of constraints regarding the size, number
and scope of microgrids in order to ease the system management.
Experiences from real projects in different countries show that a system
with a low number of large microgrids is easier to manage than many
small microgrids (please see the ‘Examples’ tab for further
information). Thus, such constraints can be introduced by defining
maximum values for both the number of microgrids and the number of
individual users and minimum values for the number of users per
microgrid. A solution is generated for each combination of such
constraints.
The solutions generated are ranked according to two
criteria (cost and system management) and three sub-criteria (number of
microgrids, number of individual users and number of users per
microgrid). Default weights obtained through surveys to rural
electrification experts are used for a first ranking of the solutions.
However, you can adjust the weights according to the characteristics of
each studied community, so as the ranking better fits your preferences.
Once satisfied with the ranking shown, you can download a
file with all the information of the solution(s) selected.
In the 'Microgrid Optimizer' tab you can find the different steps of the optimizer tool.
In the 'Examples' tab some real-case examples are provided
as well as a didactic one, which shows how the decision process and the
interaction between the decision maker and the tool works.
In the 'Contact' tab there is information about the team
behind this project as well as a contact form to write to us.
Illustration of the design methodology
In this section, the design methodology offered in this website
is illustrated through its application to a small community from the
Andean highlands. Please focus on the design process itself, rather than
the reasons behind the decisions taken at each moment. Through the
process, a variety of possibilities regarding electrifications systems
will be presented in order to obtain the solution that better suits the
community. A linear procedure consisting of four steps will be followed
in the Microgrid Optimizer tab to first introduce the input data
required to generate the electrification designs and then obtain such
designs within two interactive decision levels (Figure 1).
Figure 1. Scheme of the design methodology
Equipment data and Points data
As shown in Figure 1, data related to the electrical equipment
and the consumptions and non-consumptions points of the community is
needed. The example community consists of 6 consumption points, a
community center and 5 houses (points 2 to 6) distributed as it can be
seen in Figure 2. The energy and power demand for the community center
are 600 Wh/day and 500 W, while for the houses they are 300 Wh/day and
200 W, respectively. Considering the resources variability, an autonomy
of 2 days is required for the system. Finally, both solar radiation and
wind speed and direction are measured throughout a year to identify the
worst potential month, using specialized software. To complete the input
data, the main features of the equipment available are the following
(Table 1).
Equipment
Data
Wind Turbines
4 types are considered. Energy: 259 to 18007Wh/day. Cost: $1139 to $5645.
PV Panels
4 types are considered. Maximum Power: 50 to 150W. Cost: $451 to $1000.
PV Controllers
4 types are considered. Maximum Power: 50 to 200W. Cost: $67 to $125.
Batteries
4 types are considered. Capacity: 1500 to 3000Wh. Cost: $225 to $325. Efficiency: 0.85. Discharge factor: 0.60.
Inverters
4 types are considered. Maximum Power: 300 to 3000W. Cost: $377 to $2300. Efficiency: 0.85.
Meters
1 types is considered. Cost: $50.
Conductors
3
types are considered. Resistance: 2.6 to 0.16Ω/km. Maximum Intensity:
64 to 380A. Cost: $4.94 to $5.79/m. Nominal Voltage: 220V.
Maximum/Minimum Voltage: 230/210V
Table 1: Initial equipment data.
Figure 2: Coordinates of consumption points
Once the equipment data, the coordinates and the basic demand
of all consumption points have been introduced, it is time for the
decision maker (from now on DM) to select different demand and
management scenarios to generate the electrification alternatives in the
first and second level. In order to avoid misunderstandings, the word
“user” is now exclusively referred to potential beneficiaries of the
electricity supply, while DM is used to describe any person or
institution making the decisions to reach a final electric configuration
of the system.
First Level. Strategic decisions
In this example, two demand scenarios, with increases of 25% and
50% with regards to the basic demand, are studied for the energy and
power supply simultaneously, keeping autonomy at 2 days. Once these
demand scenarios are introduced, the algorithm is executed to generate
the corresponding electrification solutions. With such admissible
values, a total of 3 solutions are generated (the first solution
corresponds to the basic demand scenario and the other two to the
increases considered) and shown in Figure 3. Each solution presents its
electric distribution scheme, where green circles represent generation
points while yellow circles are points supplied by a microgrid. Also, it
informs about the cost needed to implement the system and the energy,
power and autonomy supplied in the worst case scenario. The worst case
scenario is defined by the consumption point which presents the lowest
difference between the demand supplied and the demand required, in
parenthesis. For example, in Solution 2, 378 Wh/day are supplied to one
consumption point, which represents a 26% increase in relation to the
base demand required. That means that all consumptions points of the
community receive at least this percentage of energy increase. A PDF is
offered for each solution to see the detail of the demand supplied to
each consumption point, as well as the whole equipment needed to
implement the system. Finally, the compromise programming technique is
used to obtain a final out-of-ten score for each solution indicating its
quality with regards to an ideal solution, which is a utopian solutions
optimal for all criteria. These scores are calculated with the default
weights given (Table 2, iteration 0), which have been obtained through
multiple surveys handed out to rural electrification experts.
Figure 3. Solutions obtained in the first level
As shown, the most compromised solution is Solution 1
(highest score) followed by Solution 3 and 2. However, the DM wants to
adjust the weights of the criteria and sub-criteria to adjust them
according to his preferences (Table 2, iteration 1). Then, Microgrid
Optimizer recalculates the score of each solution. With this
recalculation, the scores for each solution are 5, 5.1 and 5,
respectively. Since the second one is the best solution according to the
preferences of the DM, he decides to select it for further study in the
second level.
Criteria
Weigths
Sub-criteria
Weigths
Iteration 0
Iteration 1
Iteration 0
Iteration 1
Cost
0.48
0.50
Demand
0.52
0.50
Energy
0.40
0.10
Power
0.32
0.90
Autonomy
0.28
0.00
Table 2. Weights of the criteria and sub-criteria of first level
Second level. Tactical decisions
The DM wants now to consider the following scenarios for the
management of the system: two scenarios regarding the maximum number of
microgrids: 1 and 2; two scenarios of the minimum number of users per
microgrid: 2 and 3; and one non-limiting scenario for the maximum number
of individual users: 6. With these starting values, 4 alternatives are
generated combining all scenarios specified and considering also the
base scenario, without any limiting values for the management of the
system. As shown in Figure 4, only two alternatives are actually
different, and are ranked according to the default weights for the
criteria and sub-criteria of the second level.
Figure 4. Solutions obtained in the second level (iteration 0)
Not completely satisfied with the number of alternatives
obtained, the DM wants to explore new configuration designs. Therefore,
he goes into an iterative procedure and proposes a new bunch of
scenarios: the same two scenarios regarding the maximum number of
microgrids: 1 and 2; the same two scenarios of the minimum number of
users per microgrid: 2 and 3; and two scenarios for the maximum number
of individual users: 6 and 0 in order to obtain fully microgrid-based
systems. Such as the previous iteration, a total of 8 solutions are
calculated combining the scenarios suggested and the base scenario.
Since the DM thinks the default weights given for the criteria and
sub-criteria at this second level fit his requirements, they are not
modified. Figure 5 presents the new solutions and their scores
considering the default weights. As it can be seen, 4 of the 8 solutions
are actually different.
Figure 5. Solutions obtained in the second level (iteration 1)
With this information, the DM is ready to make a decision. He
discards the lowest scored solution (2-2) and also solution 2-1 due to
the high amount of individual users. The other solutions are considered
feasible and are therefore presented to all stakeholders in the
community in order to discuss their pros and cons. Although the solution
2-3 presents the highest overall score, the community prefers to have
all consumption points connected to a microgrid in order to take
advantage of a more flexible supply. The DM considers that the increase
in the cost is affordable and accepts their petition. Consequently, the
solution 2-6/2-8 is implemented as final electrification design.
Without using this website, only 2 or 3 electrification designs
could have been manually evaluated with great effort. Alternatively,
throughout this procedure, a total of 6 different electrification
designs have been generated in a short time and within a structured
procedure aimed to guide the decision-making process. Demand scenarios
are first addressed in order to obtain preliminary electrification
designs. Then, different options of system configurations are considered
to adjust the final design to the preferences of the community.
Please, do not hesitate to contact us for any doubt you might have.
Ferrer-Martí, Laia Full professor
Pastor Moreno, Rafael Full professor
García Villoria, Alberto Senior Lecturer
Domenech Lega, Bruno Lecturer
Juanpera Gallel, Marc Researcher
Murillo Vilella, Andreu Researcher
Equipment Data
Points Data
First Level
Second Level
Equipment Data
Two options are
available to introduce the data of equipment. The user can either fill
the next form manually or download and fill in a template, and upload it
with the Select file button. If the first option is chosen, a file can
be generated with the 'Save' button to save the information.
