Transcript Poster - CEProfs - Texas A&M University

Texas A&M University
Department of Civil Engineering
Instructor: Dr. Franscisco Olivera
CVEN 689 Applications of GIS to Civil Engineering
Using GIS to Find Suitable Locations for Solar Power Plants
Submitted By: Scott Peterson
May 12, 2005
Conclusion:
Abstract:
The location for alternative energy generation stations is of
paramount importance. Attempts to simply quantify and
model the parameters involved in selecting the best area for
development are quite complex. Depending on how great the
influence is of certain factors the optimal location can vary
significantly. GIS tools offer opportunities to more fully model
in advance the factors associated with choosing a site.
Current methods of selecting sites could be benefited by work
on models accurately predicting the influence of resistance,
initial cost, and solar input. More complex models which take
into account multiple load centers are also desirable. In order
to make a useful tool for planners the current implementation
of the predictive model must be improved to more accurately
reflect the geographic constraints. Companies in the position
to have concrete data on resistance losses, and capital costs
will more fully be able to utilize GIS to pick likely sites.
Choosing a location for alternative energy technologies is of
high importance. Geographic information system (GIS) tools
are useful in choosing prime locations for hypothetical power
generation facilities. The initial cost of development can be
estimated by employing GIS to measure the lengths of new
transmission lines and connections. In this project a map
showing the average solar input for the state of Texas was
utilized as an example of how GIS can help choose locations.
Calculating the power loss due to resistance on a point to
point basis should also be feasible with GIS tools. Resistance
of transmission lines can also be estimated in GIS. In the end
it would be possible to make desirability maps based on the
time to repayment. In the near term an experienced GIS user
is necessary to accomplish these tasks; however, it is feasible
to automate the process so that a few simple inputs and some
data could automatically be processed to output the
desirability map.
Methodology:
Most efforts to place solar facilities begin with a map showing the solar resources
available in the area. GIS data for such a map can be found at the National Renewable
Energy Laboratory (NREL). The following figures show the modification of the data
available from NREL (Fig 1) so that it was useful in the project. The data was
rasterized and smoothed since it is apparent that the sun does not simply stop shining
once it crosses and arbitrary line (Fig 2). The rasterized data was then used to
mathematically compute the parameters necessary for choosing a site.
Fig 1
Finally with the Resistance losses, and estimates for infrastructure cost it was possible to
map out the areas with the highest ratio of power available to capital investment. When
simply dividing the values obtained previously a somewhat unusual result was obtained
(Fig 7). Subsequently the same analysis was carried out for San Antonio (Fig 8) and El
Paso (Fig 9). The results for these other cities suggested that there was a problem in
the reasoning behind the model. The two figures are markedly similar which sems to
point to the inadequacies of the current model.
Fig 2
Fig 7
Fig 8
NREL also had a map of high voltage transmission lines available on their site. This
information allowed estimates for resistance, new line, and interconnection cost. To
delineate the state into cost districts and allocation was made to current power lines
(Fig. 3). After the allocation values for new line cost including connection costs were
calculated as measured from Dallas (Fig 4). The model used here does not allow
multiple load centers to be calculated at once, so throughout the study unless
mentioned the measurements are relating to Dallas as the load center.
Fig 9
Fig 3
Fig 4
Estimates for the resistance of given lines were made based on the transmission line
allocation in Fig 3, this results in a raster showing the loss of power in kW for every
point in Texas (Fig 5). This estimation is based on the average loss for lines of given
voltages. The total power available to a load center is then calculated by simply
subtracting the resistance loss from the solar input available (Fig 6). The resistance loss
as predicted was so low that it did not impact the power available within the tenth of a
kW anywhere in the state.
The most obvious influencing factor in this model was the costs associated with new
transmission lines and interconnection to already available infrastructure. Due the the
negligible impact of resistance in the previous steps this was the only factor that
promised to have a large impact. With this in mind another effort was made this time
assuming that all new transmission lines would be the same type (as is somewhat
logical). The result in this case (Fig 10) showed a very different pattern of desirable
areas. This also seemed less than satisfactory in the results. Finally an attempt to
average the two methods by assuming that half of the cost was constant and half was
determined by the current infrastructure resulted in what looks like a reasonable
allocation (Fig 11).
Fig 10
Fig 5
Fig 6
Fig 11