dc.contributor.advisor | Butenko, Sergiy | |
dc.creator | Cisneros Saldana, Jorge Ignacio Domingo | |
dc.date.accessioned | 2018-02-05T21:14:00Z | |
dc.date.available | 2019-08-01T06:53:55Z | |
dc.date.created | 2017-08 | |
dc.date.issued | 2017-07-26 | |
dc.date.submitted | August 2017 | |
dc.identifier.uri | https://hdl.handle.net/1969.1/165853 | |
dc.description.abstract | This thesis develops network-based approaches to analysis and optimization of wind energy systems. The wind energy system is a complex system that produces a massive amount of wind speed data over time, characterized by high level of variability. We study this system using the powerful tools of graph theory and network analysis, which provide a valuable tool for extracting important information from systems generating large amounts of data.
The main contribution of this thesis is a network-based method for finding appropriate locations for wind farms that maximize the overall energy, while controlling the effects of wind speed variability. For this purpose, we constructed networks that model potential wind farm locations as vertices and represent the pairwise correlations of the corresponding wind speed measurements using edges. More specifically, two vertices are connected by an edge if the correlation of their wind speeds considered over the given time period is below zero. If the weights of vertices are given by the average wind speed at the corresponding locations, then the problem of finding appropriate locations for wind farms is formulated as the problem of finding a tightly knit cluster of vertices with high weights. More specifically, we model clusters using the graph-theoretic concept of a clique and its relaxations, 2-plex and 3-plex.
To test the proposed approaches, we used real data from Bolivian studies of wind velocities conducted over a 10-year period at 201 locations spanning the entire Bolivian territory. The solutions obtained using the proposed approaches provide sets of diverse locations with high possible wind energy outputs. In particular, using clique relaxations results in larger number of diversified locations compared to that given by the maximum clique solutions.
Another studied problem deals with determining a small number of locations that would be representative of the overall behavior of wind speeds in the whole system. This problem was addressed searching for small dominating sets in graphs where edges correspond to pairs of locations with positively correlated wind speeds.
Finally, we proposed a methodology for evaluating costs of setting up wind farms in certain locations in Bolivia. The cost of setting up wind farms involves many variables, wind speeds being an important factor in determining the profitability of the system. We observe that for sites with higher wind speeds the net present value (NPV) of setting up and operating wind farms is positive and the internal rate of return (IRR) is higher than the discount rate, which ensures some profit to the investor. More specifically, the study has shown that with wind speeds around 6.9 m/s, the 2MW and 3MW wind turbine installments yield IRR of 13% and 15%, respectively. On the other hand, we concluded that lower wind speeds would result in projects that would not be able to recover the investment in the first 25 years. However, these projects could be profitable if the government develops policies for some green credits, or carbon bonus as income for generating and selling clean energy produced from wind power plants. The results obtained in this study could help the governments and investors interested in developing wind energy farms in Bolivia and other countries with similar geographical characteristics | en |
dc.format.mimetype | application/pdf | |
dc.language.iso | en | |
dc.subject | optimization | en |
dc.subject | wind | en |
dc.subject | energy systems | en |
dc.subject | network analysis | en |
dc.subject | maximum clique | en |
dc.subject | maximum k-plex, dominating set | en |
dc.subject | integer programming | en |
dc.subject | cost analysis | en |
dc.subject | potential wind farm locations | en |
dc.title | Network-based Optimization Techniques for Wind Farm Location Decisions | en |
dc.type | Thesis | en |
thesis.degree.department | Industrial and Systems Engineering | en |
thesis.degree.discipline | Energy | en |
thesis.degree.grantor | Texas A & M University | en |
thesis.degree.name | Master of Science | en |
thesis.degree.level | Masters | en |
dc.contributor.committeeMember | Pistikopoulos, Efstratios | |
dc.contributor.committeeMember | Damjanovic, Ivan | |
dc.type.material | text | en |
dc.date.updated | 2018-02-05T21:14:02Z | |
local.embargo.terms | 2019-08-01 | |
local.etdauthor.orcid | 0000-0001-6010-4774 | |