Document Type

Dissertation - Open Access

Award Date

2022

Degree Name

Doctor of Philosophy (PhD)

Department / School

Electrical Engineering and Computer Science

First Advisor

Siddharth Suryanarayanan

Keywords

Contingency analysis, extreme weather events, load-shed recovery, transmission switching

Abstract

The primary goal of this dissertation is to develop a suite of computationally improved transmission switching (TS) based heuristic algorithms for load shed recovery (LSR) after N-k contingencies in electric power grid. Change in power system topology may result in a reduction in required load shed because some topologies may favor a better generation redispatch than others. TS is one type of topology control and TS is a planned line outage to minimize the impact of an unplanned contingency in the electric power system. Despite many advantages, TS is not widely employed by the industry. One of the main reasons for the low (or slow) adoption of TS by the industry is the complexity (i.e., computational expense) of the algorithms proposed thus far to find a potential TS candidate for larger systems. In this dissertation, we proposed three computationally improved heuristic algorithms for transmission switching to reduce load shedding without modifying the optimal power flow (OPF) formulation. The aim is to develop computationally fast algorithms as compared to complete enumeration (CE) algorithm but at the same time, achieving same load shed recovery results as achieved by the CE algorithm. If there exist a best TS candidate, CE can find it. We used CE as a base case and compared our heuristic algorithms with the CE algorithm. LBTS is the first heuristic algorithm which is capable of finding the potential TS candidates using line flow thresholds. LBTS is computationally fast compared to CE but performs poorly for LSR post N-2 line contingencies. LSBmax is the second heuristic algorithm and is capable of finding the best TS candidates using proximity to the load shedding bus with maximum load shedding. LSBmax is computationally slower compared to LBTS but the LSR achieved by LSBmax is greater than LBTS. LiSTS is the third heuristic algorithm which is an extension of LBTS algorithm with linear sensitivities. LiSTS is computationally better than LSBmax and has LSR close to CE. High-performance computing helped with scalability of these heuristic algorithms by further improving the computational performance of the three heuristic algorithms. We also presented a case study to show the practical application of TS for the hurricane Harvey from 2017. The aim is to present a real-world practical example where TS can be useful tool to recover load shedding after an extreme weather event (EWE). The results showed that approximately 216 millions dollars could have been saved by switching a transmission line.

Number of Pages

143

Publisher

South Dakota State University

Available for download on Thursday, May 15, 2025

Share

COinS
 

Rights Statement

In Copyright