=Paper=
{{Paper
|id=Vol-1136/paper8
|storemode=property
|title=Integrated Spatial Optimization Model for Re-newable Energy Planning
|pdfUrl=https://ceur-ws.org/Vol-1136/paper8.pdf
|volume=Vol-1136
|dblpUrl=https://dblp.org/rec/conf/agile/Grassi13
}}
==Integrated Spatial Optimization Model for Re-newable Energy Planning==
https://ceur-ws.org/Vol-1136/paper8.pdf
Integrated spatial optimization model for re-
newable energy planning
Stefano Grassi
Institute of cartography and geoinformation, ETH Zurich, Switzerland
(sgrassi@ethz.ch)
Abstract. Renewable Energy is becoming more and more fundamental for the
supply of the increasing global energy demand but, it poses some issues con-
cerning the identification of the buildable land, the estimate of its exploitable
potential, the integration into the power grid and the setting of efficient energy
policies. Addressing these aspects requires interdisciplinary planning that can
be integrated in GIS. Nevertheless, the lack of dedicated functions in GIS soft-
ware in the field of wind and solar energy is a critical aspect when it comes to
estimating the economic and technical potential energy production and to quan-
tifying the uncertainties necessary to carry out sensitivity analysis. The aim of
this research is the development of a GIS process that optimizes and assesses
the technical and economic integration of solar and wind power plants in re-
spect to the energy production and to the impact on the power grid.
Keywords: wind and solar energy, optimization, technical and economic ex-
ploitable potential, GIS, energy production.
1 Introduction
The integration of renewable energy in a defined region implies two main issues
during the planning phase: the definition of buildable lands where certain constraints
have to be taken into account and the integration into the power grid. As renewable
energy sources are intermittent and characterized by an irregular spatial distribution,
their integration into the power grid represents an issue in terms of absorption and
transmission capacity of the generated electricity over the day time and the year.
Therefore the planning of renewable energy (RE) energy power plants implies also
the forecast of future scenarios in order to optimize both the location of RE power
plants in respect to the actual capacity of the power grid and to define further devel-
opment of the new transmission lines.
Among all RE technologies, solar and wind resource play an important role in the
future energy supply. Nevertheless the study of these two resources with GIS presents
different issues. Often coarse resolution data are used to find buildable land; buildable
sites are identified by applying setbacks from main natural and anthropological ele-
ments but neglecting property land. The estimate of the electricity generation of each
buildable land is based on basic assumptions such as the mean wind speed value and
simplified power curves of wind turbines (WTs). Moreover, the actual economic
boundary conditions are neglected and therefore generating additional uncertainties in
estimating the economic exploitable potential. This is a fundamental step, as the eco-
nomic assessment (integrated with the policy measures) of each buildable land repre-
sents the actual added value in demonstrating the level of attractiveness of an area.
The commercial GIS software suites have predefined functions developed for spatial
analysis and lack of specific and dedicated functions for RE analysis. This issue in-
creases the uncertainties when estimating the annual energy production.
In respect to the power grid planning, potential RE sites are located where the
transmission lines are often missing or insufficient where the RE resource is available.
Moreover, the outline and the quantification of paths and of the costs of new trans-
mission lines that have to link potential sites to the existing grid is often neglected or
estimated assuming rough hypotheses such as considering only the land within a
20km distance from main power grid at a fixed cost (1M CHF/km) independently
from the topography and the characteristics of the land cover (settlement, water bod-
ies, etc.). The visual impact is often neglected when routing new paths.
2 Previous work
The integration of any RE power plant into a given area is generally more complex
than it appears and requires great expertise and experience in order to tackle the di-
verse problems [1, 2]: finding buildable land for optimal RE resources (i.e., wind
resource greater than class 3 in NREL classification1), verifying the required condi-
tion for the grid connection and the environmental and anthropological constraints
(settlements, water bodies, protected areas, infrastructures, etc.), or collecting land-
owner information. Early research (since the 1990s) focused on estimating the ex-
ploitable solar and wind resource in a given region using GIS software [3, 4] with
some economic analysis at a large scale [5, 6] that highlighted the need for interdisci-
plinary approaches. In the first applications, large-scale maps [7] (i.e., regional or
larger) and low resolution were used because of the lack of accurate GIS data. There-
fore, estimates and scenarios were lacking robustness, reliability, and specifics. In
very few cases, GIS was applied for wind farm siting [8, 9] but always lacking in
accuracy and information.
With the intensification of RE deployment, the improvement of techniques, en-
hanced software and data, more detailed analyses have been developed leading to
more accurate estimates [10, 11] to optimize the siting of wind farms [12]. Some ap-
proaches involved the use of multi-criteria analysis (MCA) [13] to support the energy
or environmental planning [14] and the local communities in understanding the im-
pact and benefits of RE projects [15]. Moreover, despite the search of site selection
and the optimized location being one of the main features and capabilities of GIS
[16], it has not been extensively applied in the RE field yet. New studies were carried
out at large scale to quantify more accurately the RE exploitable potential and resulted
1 http://rredc.nrel.gov/wind/pubs/atlas/tables/1-1T.html
in one of the first nationwide GIS tools (called Geospatial Toolkit) of the last decade.
It visualizes the main anthropological and natural constraints and uses the solar and
wind resources represented at a low resolution (10 to 40 kilometers) with value inter-
vals instead of mean values. Another NREL project called ReEDS2 estimateted at
large scale over multiple time frames the capacity expansion in the electric sector of
the U.S using a linear programming model. The United States are divided into 365
regions, corresponding to the areas for actual and future potential power generation.
Resolution is low (some 10s to 100s kilometers) and does not take local constraints
into account. Moreover, the GIS data used to identify the buildable land are coarse
and limited [17]. In Europe, the main projects address the estimate of spatial distribu-
tion of wind resource at different heights above ground level (e.g., Wind Atlas of
Germany3, Spain4) or for regional assessment of wind energy potential such as the one
developed in Austria. GIS tools for the technical and economic analysis and scenarios
including nation-wide sensitivity assessment are not present yet except for a few pro-
jects at local levels (mainly county-wise).
In 2004, a first project - called “Konzept Windenergie Schweiz” [18] – delivered
the first assessment of the wind energy potential in Switzerland. A list of potential
sites is provided but with little information about WTs. No additional technical infor-
mation is provided. One of the limitations of this analysis relates to the limited selec-
tion of the WT model to estimate the likely annual power generation:. In general, such
analysis lacks the integration of current WT models, the estimate of the costs and the
real technical-economic assessment of the RE energy resource at a small scale. In
addition, no specific information of the uncertainties of the wind data used for the
analysis is explained. Wind sites have to be ranked in function of estimated economic
benefits in order to highlight those that are attractive as recently demonstrated [19].
In conclusion, most attempts aimed at resource evaluation and wind farm siting
have been performed at a large scale without estimating the real potential and con-
straints at a small scale (such as parcels).
The assessment of new transmission line paths represents a significant issue in
term of spatial planning. Multi Criteria Analysis (MCA) and Least-Cost Path (LCP)
will be adopted to identify the suitable routes for new transmission lines as they have
been widely applied to support decision makers in assessing the optimal routing of
linear infra-structures such as TLs. Research work described the theoretical basis of
the MCA to implement efficient methods for decision-aid [20]. LCP is typically used
to identify the shortest path between two points and different approaches have been
developed in the last decades [21, 22].
The first GIS applications have been proposed in order to offer support in power
line routing using a weighting approach [23, 24] to define preferred routings. Other
work addressed the routing problems using a “funnel” approach in order to identify
the likely corridors with an improved level of accuracy and information [25]. All
2 http://www.nrel.gov/analysis/reeds/
3 http://www.dwd.de/bvbw/
4 http://www.globalwindmap.com/VisorCENER/mapviewer.jsf?width=660&height=587
these studies proposed a solution for the routing, but neglected the constraints of elec-
trical interconnection of RE plants to the existing grid.
For solar energy, a few examples of urban roof-top exploitation have been demon-
strated. A first example of estimating solar roof-top potential has been carried out for
some counties in the cantons of Saint Gallen and Inner and Outer Rhodes Appenzell 5
and in Germany 6 but no clear selection of the PV technology is possible for the user
in respect to the broad range of current PV technologies and their corresponding per-
formances and costs. Similar analyses have been done [26] but neglecting the pitch
orientation of the roofs, the lands that can be assigned for potential new PV plants and
real costs. For the solar potential analysis of buildable land, some studies have been
carried out using a DEM of only 1km resolution. The technical and economical com-
parison of different sites over a large region with regard to the variation of boundary
conditions is another issue, which has not been integrated within the actual tools and
its development was suggested in previous work [27].
3 Research questions
This research aims to develop an interactive GIS rule-based tool in order to:
define, at high resolution, different scenarios with respect to the distribution of
the buildable surfaces for wind and solar PV power plants
estimate the optimal technical and economic performance of each potential land
that can host a RE power plant
identify and optimize the suitable policy measures in order to achieve regional or
national energy targets
assess the impact of the integration of RE into the power grid and thus to identify
the optimal implementation of the grid itself.
This novel approach will be applied in the western part of Switzerland (canton
Neuchatel and Fribourg) as case study, but designed to fit any other region: the selec-
tion of this specific region is motivated by the availability of the geodatabase of the
power grid characteristics provided by the local grid operator. The integrated tool will
be able to support decisions makers, project developers and grid operators in order to
define precisely the exploitable potential and to set energy strategies, to find sites for
investments and to identify the impact on the power grid and its weaknesses in order
to optimize future development.
4 Methodology
GIS data describing the land features (anthropological and natural constraints) will
be used to define the buildable land while a customized code, coupled with computa-
tional software and RE resources (solar and wind resource) in GIS format, will be
5 http://www.geoportal.ch/
6 http:/ratingen.publicsolar.de/
developed to estimate the corresponding RE potential for each buildable land (Figure
1) and rooftop surfaces (Figure 3). The technical specifications of the wind and PV
technology are collected from the websites of the manufacturers and converted into a
compatible GIS format. Two separated algorithms for wind and solar PV power plants
will be developed. The algorithm for wind energy applications will use the spatial
distribution of wind data and the technical specifications of the WTs (wind turbines)
as input in order to estimate the wind mean annual power generation [19]. It takes the
interaction among WTs into account by integrating a wake effect (Figure 2) derived
from physical models [28].
Figure 1: excerpt of the estimated parameters of suitable land for wind farms
The raster analysis of the digital terrain model (DTM), digital surface model
(DSM), and the land cover will result in a map of the terrain features. These parame-
ters define the aerodynamic roughness factor that directly affects the wind profile and
thus the electricity generation. Wind energy generation will be estimated by coupling
the Weibull distribution and the power curve of WTs to estimate more in detail the
annual energy production of WTs. An optimization process will be developed in order
to determine for each project site the optimal layout of WTs that maximize the eco-
nomic performance assessed using parameters such as the internal rate of return (IRR)
or the net present value (NPV).
The algorithm for solar PV applications will use the spatial distribution of solar ra-
diation data characteristics and the technical specifications of the PV panels in table
format as input for the solar mean annual electricity generation.
Figure 2: representation of the wake overlapping a downstream WT (left). Estimate of the
annual energy production of each WT: the size of the spot is proportional to energy production.
A physical model is applied to take into account the wake effect.
The raster analysis will allow to identify and to sort the rooftop in respect to the orien-
tation, shadow over the year, inclination (Figure 3). These factors are fundamental for
defining which surfaces are suitable to host solar PV plants.
Figure 3: example of extraction of rooftop characteristics such as aspect and slope
Parameters affecting the technical performances of solar and wind technology (ar-
ray effect, shadow, albedo, temperature, icing, etc.) will be considered and integrated.
In this regard, roof tops will be identified through raster analysis of the DTM men-
tioned before (≤ 2m resolution), the DOM and the orthophotos.
Optimization processes will be implemented in order to maximize parameters such
as electricity generation or economic performance; therefore different options for
different users have to be taken into account to draw future scenarios.
A validation process is required in order to quantify the uncertainties and to show
the reliability of the tool. The validation will be carried out comparing the estimated
wind and solar PV electricity generation to the actual electricity generation data of
wind and solar PV plants of Switzerland that is publicly accessible.
The integration of the GIS data of the existing power grid will be used to identify
the likely path of the grid interconnection between the buildable sites and the existing
grid and the corresponding cost that represents one of the main bottlenecks to diffu-
sion of distributed generation.
Optimizing processes will be developed to assess:
the optimal location of power plants in respect to the characteristics of the power
grid which represents the bottleneck to the development of RE projects.
the optimal effective energy policy to be implemented in order to achieve specific
targets.
5 Conclusion and outlook
The proposed research is aimed at developing an interactive analysis tool to esti-
mate, at high resolution and in short time, the actual technical and economic exploita-
ble potential of solar and wind sources that can be integrated into the grid and to set
energy strategies. This tool will fill the gap between the current GIS models and the
need for a detailed extensive analysis for RE planning and prospecting fulfilling the
need for different users: planners can easily identify new buildable sites and will re-
duce uncertainties, policy makers can outline different strategies and targets and grid
operators can optimize the interconnection of RE projects and the planning of future
transmission lines. As one of the main issues of distributed generation is the optimal
location and balance between the power generation and the electricity demand, the
additional added value is the capability to perform optimization analysis in respect to
boundary conditions such as setbacks from anthropological and natural constraints,
constraints of the power grid and policy framework.
The integration of the power grid will help to assess how and where the actual
structure of the grid has to be enhanced, the impact on the cost and load distribution
of the electricity. This will enable us to identify new corridors for further develop-
ment of new transmission lines in order to link potential exploitable sites for RE to the
grid. The following step will be the integration—into a single model—of all power
generation systems of Switzerland in order to study the global spatial impact of tradi-
tional and RE plants on the electricity (generation and final use), and transmission
costs and power grid reinforcement (domestic grid and interconnections).
The results will be represented both through thematic maps and through plots in
order to show their spatial distribution and the corresponding characteristics.
The coupling of computational tools and customized function with a GIS tool will
enable a more complex and accurate modeling, simulation and optimization of future
scenarios including the theory of solar and wind energy analysis that are normally
carried out with dedicated tools.
The potential application of the outcome this research is a multifunctional tool for
planning and analysis that addresses the needs of different types of users within one
single package: energy planners, project developers and grid operators in order to
reduce time and uncertainties in planning and to optimize the estimate and the fore-
casts of the integration of RE projects over a large region. At the moment the com-
mercial tools address a micro-scale analysis (10kmx10km) and enable users to study a
single project within a limited region, but there is no tool that estimates at high resolu-
tion the potential over a large region in order to identify in detail for each suitable
surface the technical and economic characteristics.
6 References
1. EU: European best practise guidelines for wind energy development. Energy
Commission; http://ec.europa.eu/energy/renewables/studies/wind_energy_en.htm
(2002)
2. Giovanello, A., Kaplan, C.S.: Wind Energy siting handbook. AWEA -
American Wind Energy Association (2008)
3. Clarke, J.A., Grant, A.D.: Planning support tools for the integration of
renewable energy at the regional level. Renewable Energy Volume 9, Issues 1-4:
1090-1093 (1996)
4. Rylatt, M., Gadsden, S., Lomas, K.: GIS-based decision support for solar
energy planning in urban environments. Computers,Environment and Urban Systems
Volume 25, Issue 26, Pages 579-603 (2001)
5. Marnay, C., Richey, R.C., Mahler, S.A., et al.: Estimating the environmental
and economics effects of widespread residential PV adoption using GIS and NEMS.
Ernest Lawrence Berkeley National Laboratory: University of California, EEUU
Report, (1997)
6. Voivontas, D., Assimacopoulos, D., Mourelatos, A.: Evaluation of renewable
energy potential using a GIS decision support system. Renewable Energy 13(13):333-
344 (1998)
7. Kirby, M., Dahle, D., D., H., et al.: Assessing the Potential for Renewable
Energy on Public Land. U.S. Bureau of Land Management (2003)
8. Baban, S.M.J., Parry, T.: Developing and applying a GIS-assisted approach
to locating wind farms in the UK. Renewable Energy Volume 24, Issue 21: 59-71
(2001)
9. Jangamshetti SH, Rau VG: Optimum Siting Of Wind Turbine Generators.
IEEE Transaction on energy conversion 16, 8-13 (2001)
10. Janke, J.R.: Multicriteria GIS modeling of wind and solar farms in Colorado.
Renewable Energy 35, 2228-2234 (2010)
11. Nguyen, H.T., Pearce, J.M.: Estimating potential photovoltaic yield with
r.sun and the open source Geographical Resources Analysis Support System. Solar
Energy Vol.84, Issue 85: 831-843 (2010)
12. Sliz-Szkliniarz, B., Vogt., J.: GIS-based approach for the evaluation of wind
energy potential: A case study for the Kujawsko–Pomorskie Voivodeship. Renewable
and Sustainable Energy Reviews 15, 1696-1707 (2011)
13. Tegou, L.I., Polatidis, H., Haralambopoulos, D.A.: Environmental
management framework for wind farm siting: Methodology and case study. Journal of
Environmental Management Volume 91, Issue 11, Pages 2134-2147 (2010)
14. Aydin, N.Y., Kentel, E., Duzgun, S.: GIS-based environmental assessment of
wind energy systems for spatial planning: A case study from Western Turkey.
Renewable and Sustainable Energy Reviews Volume 14, Issue 11: 364-373 (2010)
15. Van Hoesen, J., Letendre, S.: Evaluating potential renewable energy
resources in Poultney, Vermont: A GIS-based approach to supporting rural
community energy planning. Renewable Energy Vol.35, Issue 39: 2114-2122 (2010)
16. Church, R.L., Murray, A.T.: Business site selection, location analysis, and
GIS. John Wiley & Sons, Inc., Hoboken, New Jersey (2009)
17. NREL: Regional Energy Deployment System. NREL - National Renewable
Energy Laboratory, (2009)
18. Kunz, S., Dällenbach, F., Schaffner, B., et al.: Konzept Windenergie Schweiz
- Grundlangen für die Standortwahl von Windparks. Bern: Bundesamnt für Energie
(BFE), Bund. für Umwelt Wald und Landschaft (BUWAL), Bund. für
Raumentwicklung (ARE) (2004)
19. Grassi, S., Chokani, N., Abhari, R.: Large scale technical and economic
assessment of wind energy potential with a GIS tool: case study Iowa. Energy Policy
45, 58-73 (2012)
20. Malczewski, J.: GIS and multicriteria decision analysis New York. (1999)
21. Dijkstra, E.W.: A Note on Two Problems in Connexion with Graphs.
Numerische Mathematik 1, 269-271 (1959)
22. Eppstein, D.: Finding the k Shortest Paths. SIAM Journal on Computing 28,
652-673 (1998)
23. Jankowski, P., Richard, L.: Integration of GIS-based suitability analysis and
multicriteria evaluation in a spatial decision support system for route selection.
Environment and Planning B: Planning and Design 21, 323 – 340 (1994)
24. Towers, G.: GIS versus the community-Siting power in southern West
Virginia. Applied Geography 17, 111-125 (1997)
25. Glasgow, J., French, S., Zwick, P., et al.: A Consensus Method Finds
Preferred Routing. GeoWorld 19, 22-25 (2004)
26. Wiginton, L.K., Nguyen, H.T., Pearce, J.M.: Quantifying rooftop solar
photovoltaic potential for regional renewable energy policy. Computers, Environment
and Urban Systems Volume 34, Issue 34: 345-357 (2010)
27. Dominguez, J., Amador, J.: Geographical information systems applied in the
field of renewable energy sources. Computers & Industrial Engineering 52, 322–326
(2007)
28. Frandsen S, Barthelmie R, Pryor S, et al.: Analytical modeling of wind speed
deficit in large wind offshore windfarms. WindEnergy 9, 39-53 (2006)