With the high penetration of inverter-based distributed energy resources (DERs), network situational awareness is a challenge for both transmission and distribution (T&D) system operators and planners. It is becoming increasingly important to equip transmission planners with the visibility of DER dynamic performance in distribution systems. Not having visibility of the disconnection of DERs with the occurrence of transmission events, could result in an erroneous view of the stability of bulk power system. The aggregated distributed energy resource (DER_A) model is able to model the aggregated dynamic response of inverter-based DERs in distribution circuits. The parameterization of DER_A model requires extensive Monte Carlo dynamic simulations of a large range of highly DER penetrated distribution circuits. These dynamic simulations provided with dynamic voltage responses and parameters of the partial voltage trip characteristic. In this dissertation, the parameterization process of DER_A model is first streamlined to effectively represent the behavior of the aggregated DERs' response, with their impact on the bulk power system thoroughly studied. Then, machine learning (ML) models are developed to harness the extensive data generated during the parameterization of the DER_A model. These data are carefully examined to identify parameters that have a significant impact on the dynamic response of distribution systems due to a fault in bulk power systems. A convolutional neural network (CNN) model is designed and trained to accurately predict the dynamic response and the tripping curve of aggregated DERs. Furthermore, probabilistic distributions are applied by modeling a Bayesian CNN in order to provide the user with information about the prediction. The model's output provides the prediction, its level of confidence, as well as the distribution of predicted response. This research work not only provides a data-driven and learning-enabled tool to support users' analysis of integrated T&D systems with high DER penetration, but also increases the industry's trust in ML applications for power systems through probabilistic modeling.


If this is your thesis or dissertation, and want to learn how to access it or for more information about readership statistics, contact us at STARS@ucf.edu.

Graduation Date





Sun, Wei


Doctor of Philosophy (Ph.D.)


College of Engineering and Computer Science


Electrical and Computer Engineering

Degree Program

Electrical Engineering




CFE0008797; DP0026076





Release Date


Length of Campus-only Access

1 year

Access Status

Doctoral Dissertation (Open Access)