Degree Type


Date of Award


Degree Name

Doctor of Philosophy


Electrical and Computer Engineering

First Advisor

Venkataramana Ajjarapu


The continuous increase in electric energy demand and limitations in the reinforcement of generation and transmission systems, have progressively led to a greater utilization of power systems and transmission lines. As a result, system conditions may arise where voltage collapse phenomena have a high probability to occur, either due to the accidents in the system structure, or to load becoming particularly heavy. Recently, "Workshop on Residential Air Conditioner (A/C) Stalling" of Department of Energy (DOE) reported that fault-induced delayed voltage recovery (FIDVR) is now a national issue since residential A/C penetration across U.S. is at an all time high and growing rapidly. The unique characteristics of air conditioner load could cause short-term voltage instability, fast voltage collapse, and delayed voltage recovery. In order to study and mitigate FIDVR problem, a systematic load modeling methodology utilizing novel parameter identification technique and an online demand side control scheme based on load shedding strategy are developed in this dissertation.

As load characteristics change from traditional incandescent light bulbs to power electronics-based loads, and as the characteristics of motors change with the emergence of high-efficiency, low-inertia motor loads, it is critical to understand and model load responses to ensure stable operations of the power system during different contingencies. Developing "better" load models, therefore, has been an important issue for power system analysis and control. It is necessary to take advantage of the state-of-the-art techniques for load modeling and develop a systematic approach to establish accurate, aggregate load models for bulk power system stability studies. In this dissertation, a systematic methodology is provided to derive aggregate load models at the high voltage level (transmission system level) using measurement-based approach. A novel parameter identification technique via hybrid learning is also developed for deriving load model parameters accurately and efficiently.

According to NERC's definition, FIDVR is defined as the phenomenon whereby system voltage remains at significantly reduced levels for several seconds after a fault in transmission, subtransmission, or distribution has been cleared. Various studies have shown that FIDVR usually occurs in the areas dominated by induction motors with constant torque. These motors can stall in response to sustained low voltage and draw excessive reactive power from the power grid. Since no under voltage or stall protection is equipped with A/Cs, they can only be tripped by thermal protection which takes 3 to 20 seconds. Severe FIDVR event could lead to fast voltage collapse. In this dissertation, a novel online demand side control method utilizing motor kinetic energy is developed for disconnecting stalling motors at the transmission level to mitigate FIDVR and fast voltage collapse.


Copyright Owner

Hua Bai



Date Available


File Format


File Size

161 pages