Optimized FTR portfolio construction: the speculator's problem

Apostolopoulou, Dimitra

Permalink

https://hdl.handle.net/2142/29799 Copy

Description

Title

Optimized FTR portfolio construction: the speculator's problem

Author(s)

Apostolopoulou, Dimitra

Issue Date

2012-02-06T20:17:47Z

Director of Research (if dissertation) or Advisor (if thesis)

Gross, George

Department of Study

Electrical & Computer Eng

Discipline

Electrical & Computer Engr

Degree Granting Institution

University of Illinois at Urbana-Champaign

Degree Name

M.S.

Degree Level

Thesis

Keyword(s)

• Financial Transmission Rights
• Speculator's problem
• Portfolio construction
• FTR auctions
• FTR revenues
• Power transfer distribution factor

Abstract

In this thesis, we propose a systematic methodology to construct an optimized financial transmission rights (FTR) portfolio for the speculator, who purchases FTR holdings in order to have returns that are as good as possible. The conventional approach of selecting the FTR in a portfolio requires the exhaustive evaluation of all the possible FTR combinations, which in a large-scale network is computationally too demanding a task, particularly when the wide variations in the behavior of the locational marginal price (LMP) differences of nodes over the many hours of the holding period are taken into account. In order to make the speculator's problem more manageable, we recast the problem into a form that allows us to exploit the salient characteristics of power systems, the topological nature of the underlying network and the historical data, so as to gain mathematical insights that we apply to develop the proposed scheme. The speculator returns are collected from the hourly day ahead markets (DAMs) only for those hours that the grid is congested, i.e., the flows on one or more lines are at their maximum limits. Each MW flowing through those lines incurs a transmission usage cost. Unlike a physical transaction from a source node to a sink node that holds FTR in the amount of the flow and receives reimbursement for the transmission usage charges from the independent grid operator (IGO), the speculator who holds FTR for the same node pair simply receives those revenues, because of lack of physical flows. Thus, the identification of congested lines is a key step in the construction methodology. So, rather than focusing on the LMP differences of node pairs to choose FTR, we select node pairs such that the selected congested lines are on their paths from the source nodes to the sink nodes. Conceptually, we specify FTR such that transactions with same node pairs and amounts induce real power flows on the selected congested lines. The strategy of the speculator is to select each congested line and his level of participation on the congested flows on the line. In practice, the speculator cannot do this for all the congested lines, because that would imply the acquisition of too large a number of FTR, whose premiums add to his costs and, thus, lower his returns. Under the assumption that the past behavior continues in the future, he judiciously chooses a subset of lines whose transmission usage costs exceeded the speculator's specified price and time fraction thresholds historically. This subset forms the basis of the optimized FTR portfolio construction. In our proposed scheme, we construct the FTR portfolio with minimum number of node pairs; i.e., we find the minimum number of transactions that induce the desired real power flows on the subset of selected lines. To demonstrate the computational efficiency of the construction algorithm, we select a subset of nodes to specify the FTR node pairs in the portfolio. The manageability of the problem is further aided by focusing on a small number of node pairs. Fewer node pairs improve the manageability. The recasting of the problem in terms of congested lines, rather than LMP differences of node pairs, results in a simplified solution methodology that is amenable to practical implementation. We have extensively tested the proposed methodology on multiple test systems and we discuss representative case study results. The results on three test systems, including the PJM ISO network, illustrate the effectiveness of the proposed approach and provide insights into the nature of the problem.

Graduation Semester

2011-12

Permalink

http://hdl.handle.net/2142/29799

Copyright and License Information

Copyright 2011 Dimitra Apostolopoulou

Owning Collections