ESTIMATING SOLAR ENERGY PRODUCTION …docs.fct.unesp.br/departamentos/cartografia/eventos/2009...ESTIMATING SOLAR ENERGY PRODUCTION POTENTIAL OF URBAN REGION USING LASER SCANNER DATA

ESTIMATING SOLAR ENERGY PRODUCTION POTENTIAL OF URBAN REGION USING LASER SCANNER DATA AND HIGH RESOLUTION IMAGES

G. V. Souza a*, J. A. S. Centeno b, A. M. L. Machado c

a, b, cDepartamento de Geomática, Universidade Federal de Paraná - UFPR.Caixa Postal 19001, CEP 81531-990, Curitiba, PR, Brasil.

a [email protected] [email protected]

c [email protected]

Commission VI, WG V/I

KEY WORDS: Photovoltaic energy, Laser Scanner, high resolution image.

ABSTRACT:

Currently, the use of alternative energies, aiming at the reduction of carbon emission, has become a necessity, mainly in urban regions, where the use of energy is high. For the production of electric energy from solar radiation, a clean energy source, it is necessary to know the photovoltaic potential and solar resource of a region. In urban environments, a great amount of flat surfaces is to capture solar energy in the form of roofs. In order to use the buildings to collect solar energy, it is necessary to know if the available roofs are viable, in economic terms, in comparison to other alternative sources. This research uses airborne laser scanner data together with a Quickbird panchromatic imagery to estimate the photovoltaic potential of the roofs within an urban region of Curitiba, in Brazil. The spatial limits and form of the roof is derived from the integrated dataset, combining laser scanner altitude data with the multispectral image. Both data sources are complementary and allow the estimate of the form, orientation, slope and size of the roof. The result is the used to compute the photovoltaic potential for different dates, considering the seasonal variations of the incident solar energy. The results show that the method can be used to estimate the potential of the use of roofs as solar collector, and then the size of the roofs is big enough in relation to the resolution of the scanner and the images. A data set with low density on the roof can lead to an inaccurate estimation of the slope, size and orientation of the planes that form the roof, causing errors in the estimation of the photovoltaic potential. Nevertheless, the method is valid for the choice of the best places to use of solar energy as complementary energy source in an urban environment.

1.INTRODUCTION

The percentage of domiciles supplied with electricity comes gradual increasing each year, making with that the demand grows in short term. This world-wide demand if has expanded quickly, had to world-wide the economic growth and the increase of the population. Ally to other causes, this generated the necessity of energy rationing. Advances in science and technology have provided some alternatives with production of energy to a sustainable level. One of the technologies you renewed more promising of energy generation is the photovoltaic generation that comes more being each time used by the industrialized countries. It has left of the beginning of being a quiet, static, non-polluting source and without any depletion of materials that directly converts the energy of the sun into electric energy. The photovoltaic solar energy has provided electric energy for any application and in any localization in the land and the space, having been that the urban way started if to detach as a great absorber of this ecological technology. As characteristic main, the photovoltaic generation shows an excellent capacity of work with other power plants. Its installations can be found working with nuclear central offices, hydroelectric plants and of all the types.

With the objective to get a bigger income of the solar panel, it is necessary to have a direction that goes following the trajectory of the sun during the day. However in the majority of the simple installations, the localization is fixed, and for this, its positioning must possess some characteristics in accordance with the orientation and inclination (Quadri, 1991). The excellent orientation of the solar panel is guided to the north, to use to advantage the biggest number of hours of the sun, independent of the station of the year and the latitude of the place. In the case of that it is not possible this positioning, the localization does not have deviation of the North more than 20º (Quadri, 1991). The inclination depends on two factors: • Latitude of the place; • Period of use during the year. For this work the inclinations had been calculated ideas for use only in the summer, only in the winter and for all the year. The extracted variable of the integration of data laser with the high resolution image that influence in the estimate of the energy generated in a residence in one day, they had been given of altitude of the lower point highest and of the residence in order to know the inclination of the water, and with the perpendicular the line of the top of the roof, has if the azimuth of the roof.

* Corresponding author.

6th International Symposium on Mobile Mapping Technology, Presidente Prudente, São Paulo, Brazil, July 21-24, 2009

This identifies the best part of the roof to locate the photovoltaic panel, and through the inclination it is calculated incidence of solar energy in one day. This technique allows evaluating the photovoltaic potential of some residences with an only survey. To know the potential photovoltaic of a residence, as well as its correct positioning, is decisive factor for use of this clean power plant.

2.STUDY AREA AND MATERIALS

2.1 Study area

The study area is located in the state of the Paraná, in the city of Curitiba, quarter Jardim das Americas.

2.2 Materials

For this work they had been used a high resolution image gotten by the Quickbird sensor, dated of March of 2002, with 0,70m of space resolution (hybrid image), and spectral resolution of 4 bands. Together, given rude gotten for Laser sweepings Scanner, dated of April of 2005.

3.METHODOLOGY

3.1 Treatment of the rude data

3.1.1 Laser Scanner

The rude data of the laser to scanner had been used for the generation of a Digital Terrain Model (DTM), and a Digital Elevation Model (DEM), for this, used a regular grating with 20 cm of spaced for the generation of both models. Later, in the ENVI, these models had been deducted, giving origin to a Digital Model of Normalized Surface (DMNS).

Figure 1. Digital Model of Normalized Surface.

3.1.2 Quickbird image

Figure 2. Quickbird image composition 4,3,2.

The Quickbird image was corrected geometrically, and a fusing for main components was carried through for the attainment of the image with 4 bands and space resolution of 0,70m. Later, a clipping of this image was carried through to locate the same area gotten with the data laser to scanner.

3.2 Data integration

The Quickbird image and the DMNS had been overlapped in software ENVI, generating a hybrid image, with altitude data. The gotten data of the Quickbird image had allowed evaluating the azimuths of waters of the roofs, together with the distances between the top of the roof and the part lowest, for the calculation of the inclination. Already the happened data of the DMNS had allowed calculating the inclination of the roof.

3.3 Choosing edifications

For the analysis of the photovoltaic potential, residences had been chosen that possess waters in ideal conditions, that is, its water is in an azimuth of until + 20º in relation to the north. For verification, some samples had been gotten with distinct azimuths, in order to evaluate the differences of gotten potential.

3.4 Calculating inclination

For this work, 23 residences had been chosen, with distinct characteristics between itself, in order to evaluate its photovoltaic potential. Distinct the hybrid image was chore in two software, in order to get the data for the calculation of the inclination: • In the ENVI, the heights of the points highest and lower of the roof of the construction had been chores; • In the AutoCAD the horizontal distances between the top and the edge of the constructions had been chores. For the calculation of the inclination the formula was used (1):

1 1 2A AI tgd

− − = (1)

Where I = Roof inclination;A1 = height of the roof’s top;A2 = height of the roof1s bottom;D = distance between the top and the bottom of

the roof.

3.5 Azimuth

For the attainment of the azimuth of the water, in the AutoCAD a perpendicular to the line of the top of the house was traced, in the chosen direction, and later read its angle in relation to the north.

3.6 Calculating the duration of the day

For the city of Curitiba, the declination of the Sun uses it following declination of the sun: • Summer = -23º; • Winter = 23º. From this data, the hourly angle of the rising is calculated and to occult of the sun, being used formulas (2) and (3):

1

1

( )cos( )

( )cos( )

n

o

tgHtg

tgHtg

ϕδ

ϕδ

−

−

−=

=

(2)

(3)

Where Hn = clockwise angle of the sunrise;H0 = clockwise angle of the sunset;φ = latitude;δ = sun’s declination.

Finally, the duration of the day is given by (4):0| | | |nD H H= + (4)

Where D = duration of the day;

3.7 Calculating the insolation

The calculation of the insolation of the chosen water of the house is made in parts. First with the equation (5), angle auxiliary M is calculated:

( )( )

1 tg AzM tg

sen ϕ−

= (5)

Where Az = roof’s azimuth;φ = latitude

For the calculation of insolation, it will use the value of declination of the sun of 0,409173051 for winter and -0,409173051 for the summer, having remembered that this value is given for the city of Curitiba. The angle of touch of the sun for the winter and the summer is calculated then, using the above-mentioned values of declination, through the equation (6):

( )1 cos( )* ( )cos M tgHt

tgδ

ϕ−

= (6)

Where Ht = touch angle;M = auxiliary angle;δ = sun’s declination of the season;φ = latitude.

Now, it is gotten insolation of the water of the roof for the winter and summer, following the equation (7):

0 tIs H H= + (7)

Where Is = insolation;H0 = clockwise angle of the sunset;Ht = touch angle.

3.8 Calculating the photovoltaic potential

For the calculation of the photovoltaic potential, the value was used 5000Wh/m2, gotten of Atlases of Solar Irradiation of Brazil, for the South region. The value approached for the generation of energy for one day is gotten by the equation (8):

5000*fP Is= (8)

Where Pf Photovoltaic potential;Is = insolation.

The values are gotten in Wh/m2 per day, that is, for a solar panel of 1 m2 per one day.

4.RESULTS

4.1 Inclination and Azimuth

House Az. Inclination House Az. Inclination1 25 21,03751103 14 24 10,548776592 25 25,78480669 15 22 26,622461763 26 26,91299698 16 24 18,755359094 24 25,7307056 16 24 21,801409495 25 14,34613348 16 24 22,326858696 66 22,12194197 17 23 14,241358757 24 26,45091631 17 66 14,488203238 22 28,19989916 18 48 13,869686449 24 20,46227152 19 25 38,43730149

10 23 14,93141718 20 24 21,8409920611 22 9,104600865 21 25 30,7627195312 22 26,28141102 22 25 13,4233052513 24 18,20848447 23 26 13,80414378

Table 1: Azimuth and inclination

4.2 Insolation

The gotten values of the insolation had been for the winter and summer (extreme situations). The gotten values meet in table 2 to follow:

house W/day winter W/day summer1 25534,26 45642,662 25534,26 45642,663 25547,17 45852,714 25520,45 45425,865 25534,26 45642,666 11929,29 37054,477 25520,45 45425,868 25489,66 44971,39 25520,45 45425,86

10 25505,62 45202,1311 25489,66 44971,312 25489,66 44971,313 25520,45 45425,8614 25520,45 45425,8615 25489,66 44971,316 25520,45 45425,8616 25520,45 45425,8616 25520,45 45425,8617 25505,62 45202,1317 11929,29 37054,4718 13638,7 37121,2619 25534,26 45642,6620 25520,45 45425,8621 25534,26 45642,6622 25534,26 45642,6623 25547,17 45852,71

Table 2 - Photovoltaic potential in wh/m2 per day.

5.DISCUSSING

5.1 Roof’s inclination

For this work, the data had been extracted of a high resolution image Quickbird pan-sharpened, together with laser sweepings to scanner. It did not have much variation in the inclinations of the roofs. This fact occurs for the difficulty of the laser to always find the top and the edge of the construction, finding an approach value. To weigh of the regular grating to have been generated with 20 cm, the amount of points for square meter in the sweepings is basic for the extraction of this type of information. At the same time one verifies that this lack of variation does not influence in the results of the photovoltaic potential, therefore the constructions possess an average inclination of 25°, being the ones that present discrepancy in relation to these inclinations, to occur due to the fact of that with the data laser it is not obtained to detect the edges of the construction. When reading the height in the edge of these constructions, many times appeared value zero, (therefore a DMNS was used), having in the distance to find the value of the height of the edge of the construction in a more internal point in the polygon roof, diminishing between the top and the base, then increasing the inclination, case of construction 19.

To weigh of this type of problem, the value of the photovoltaic potential did not suffer a significant alteration.

5.2 Photovoltaic potential

As we saw previously, for the locality, it is ideal that the photovoltaic panel has its directed face northward. In case that this is not possible, that it has a variation of + 20°. This factor is verified clearly, when we use constructions 6, 17 and 18, that 66° (6 and a water of the 17) and 48° possess azimuths (18). The photovoltaic potential falls of 45 kWh/m2 for 37 kWh/m2. All the other constructions possess values seemed, due to the order square of them in the place.

6.CONCLUSIONS

This work carried through a estimate of the photovoltaic potential for residences through data gotten through high resolution image and laser sweepings to scanner. The methodology showed that, it is possible to extract this information in fast way and with a significant speed due to possibility to verify some constructions in an only survey. One also revealed that, for constructions that possess the inclination of its roofs with bigger azimuths that 20°, has a significant reduction in the photovoltaic potential. The attainment of data through laser to scanner is efficient for this type of estimate, therefore, possible problems in the calculation of the inclination of the roofs do not influence of significant form in the attainment of the estimate of the photovoltaic potential. This method proved to be practical, simple and very efficient.

7.BIBLIOGRAPHY

Colle, S. Pereira, E. B. Atlas de irradiação solar no Brasil. http://www.lepten.ufsc.br/pesquisa/solar/atlas_de_irradiacao.pdf (accessed March 18, 2009).

Nadal, C. A. Insolação de Paredes Verticais. DAEC. Curitiba, 1997. 46p.

Quadri, N. P. Energía Solar. Librería y Editorial Alsina. Buenos Aires, 1991. 150p.

Ribot, M. J. Curso de Energía Solar. Imprimeix S. Coop. Ltda. Barcelona, 1995. 8 volumes.

Salamoni, I. T. Metodologia para cálculo de geração fotovoltaica em áreas urbanas aplicadas a Florianópolis e Belo Horizonte. Dissertação de Mestrado. UFSC, 2004.