Optimal sizing and location on energy storage systems

Official URL: https://risc01.sabanciuniv.edu/record=b2711501_(Table of contest)

Energy storage systems (ESSs) play an important role in the rate of renewable energy adoption. Because, in addition to their higher costs, renewable energy sources are also disadvantaged due to their highly variable and intermittent nature. Therefore, it is almost impossible to adopt renewable energy at a meaningful level without a well placed ESS infrastructure. Unfortunately, ESSs too may be quite expensive, which makes these infrastructure decisions even more important. Due to its popular nature, one might mistakenly believe that these ESSs are nothing but battery energy storage systems (BESS) placed at the sites of the renewable generation units. However, the most economical ESSs are those of pumped hydro storage (PHS) units that are dispersed across the geographies that are served by large electricity networks. The subject of this dissertation is to study siting and capacity decisions of ESSs in electricity transmission networks. We consider systems where electricity generation units from renewable and conventional sources are already established. For such systems our purpose is to find the location and capacities of ESSs and transmission line capacities to minimize the total system costs. Perhaps the greatest challenge in finding optimal solutions of these set of structural decisions is the incorporation of operational decisions which influence as well as are influenced by these structural decisions. In fact, optimal resolution of operational decisions alone is a daunting task even in the smallest networks because these decisions must be dynamically made while considering the uncertainties in electricity demand and supply. Therefore, we are compelled to adopt a sample average approximation (SAA) approach in the incorporation of those operational decisions. This dissertation has four main chapters. The introduction, apart from the overall motivation for this study, also presents a review of current technological and economical properties of wide variety of ESS alternatives. This section allows us to identify appropriate ESS alternatives that can be used in our models and their realistic cost estimates. The second chapter presents our problem in an island electricity system, which is commonly studied in energy literature for their simple transmission networks and small size. In our island system, there is only one generation unit, which is a wind farm, one demand node, and two alternative storage systems; one PHS with a known location and one BESS to be located at the site of the wind farm. Whenever there is a shortage, the demand is satisfied from diesel at the demand node. Hence, the structural decisions to be found are the capacities of ESSs and the transmission lines to minimize the total cost of investments, operations and maintenance, and diesel costs. We must also remark that capacity of an ESS is a pair of variables; one for the maximum energy storage and one for the maximum energy flow, which are commonly referred as the energy rate and power rate, respectively. By deploying two different storage types at different places, we investigate the circumstances where installation decisions change. Stochastic renewable energy generation and demand are taken into account by scenarios that are reproduced based on real data. The third chapter extends the mathematical model to a small grid system with 13 nodes that consist of various generation units and demand centers. Despite being far smaller than many realistic grid systems, the problem has shown to be far beyond resolution with the existing computational resources. Therefore, we have developed a two-stage algorithm, which determines the investment decisions in the first stage, followed by the second stage operational decisions. Finally, the fourth chapter revisits the island system with solar power instead of the wind power, to investigate how differences in the intermittent feature of the two most common renewable sources affect the optimal structural decisions.