Engineering-Economic Methods for Power Transmission Planning Under Uncertainty and Renewable Resource Policies
Embargo until
Date
2014-02-03
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Johns Hopkins University
Abstract
Power transmission networks are some of the world's largest machines. Investments in these assets have historically been driven by projections of load growth, interconnection of baseload conventional power plants, reliability standards, and the economic exchange of electricity. However, today the transmission system is also seen as a key enabler to meet the new public policy goals that seek to incorporate large amounts of generation from renewable resources into the grid. In this dissertation I analyze three different engineering-economic challenges of power transmission planning that arise from the large scale integration of renewable energy technologies.
In the first essay I study the effects of transmission approximations on the design and performance of Renewable Portfolio Standards. In particular, I analyze how disregarding the indivisibility of transmission investments (i.e., lumpiness) or Kirchhoff's Voltage Law yield distorted estimates of the type and location of infrastructure, as well as inaccurate estimates of the cost of complying with renewable goals. I also utilize multi-stage investment models to study the potential benefits of coordinating the timing of transmission investments and the design of multi-year renewable energy policies.
In the second essay I propose a stochastic programming-based tool for adaptive transmission planning under market and regulatory uncertainties. The model considers investments in two stages, generators' response, and Kirchhoff's Voltage Law enforced through disjunctive constraints. I use a 240-bus representation of the Western Electricity Coordinating Council to illustrate its application, calculate the Expected Value of Perfect Information and the Expected Cost of Ignoring Uncertainty, and compare its performance to heuristic investment decision rules based on scenario planning.
The third essay describes a new, two-phase bounding and decomposition method to solve large-scale transmission and generation investment planning problems under various environmental constraints designed to incentivize high amounts of intermittent generation in electric power systems. The first phase exploits Jensen's inequality which I extend to stochastic problems with expected-value constraints. The second phase is an enhancement of Benders decomposition that I utilize to reduce the residual solution gap from the first phase. Numerical results show that only the bounding phase is necessary if loose optimality tolerances are acceptable. Attaining tight solution tolerances, however, requires utilization of the decomposition phase, which performs much better in terms of convergence speed than attempting to solve the problem using either algorithm, the bounding method or Benders decomposition, separately.
The main contributions of this dissertation include a better understanding of the interaction between different renewable energy policy designs and transmission investments, a new method to conduct transmission planning studies under gross economic and public policy uncertainties, and new algorithms to solve large-scale transmission and generation planning problems.
Description
Keywords
Transmission Planning, Uncertainty, Renewable Portfolio Standards, Benders Decomposition, Bounding Methods, Parallel Computing