Graduation Semester and Year
Summer 2026
Language
English
Document Type
Thesis
Degree Name
Master of Science in Electrical Engineering
Department
Electrical Engineering
First Advisor
Dr. Liwei Zhou
Second Advisor
Dr. Yichen Zhang
Third Advisor
Dr. Tianqiao Zhao
Abstract
AI data centers can produce rapid changes in electrical demand that may influence current transformer performance during faults. This study evaluates the effect of an AI data center transient on CT saturation during single line-to-ground faults using a 400 V, 60 Hz grid connected inverter model in MATLAB/Simulink. The normal condition transient produced a maximum RMS current rate of approximately 211 A/ms, which was used along with the maximum power condition to define fault inception cases. A MATLAB time-domain CT model then swept the fault current DC offset coefficient to determine the minimum offset required for CT saturation. The calculated saturation thresholds were approximately −0.86 pu at the beginning of the transient, −0.67 pu near the peak RMS current-rate condition, and −0.68 pu near the maximum-power condition. Although the modeled system’s maximum realistic DC offset remained below these thresholds, the results indicate that fault timing during the data center transient can reduce CT saturation margin and increase CT sensitivity to saturation.
Keywords
Power Systems, Power Systems Analysis, AI Data Center, Dynamic Load Analysis, Transient Analysis, Current Transformer, CT Saturation
Disciplines
Power and Energy
License
This work is licensed under a Creative Commons Attribution 4.0 International License.
Recommended Citation
Hernandez, Sergio A., "AI DATA CENTER DYNAMIC LOAD EFFECTS ON CURRENT TRANSFORMER SATURATION" (2026). Electrical Engineering Theses. 4.
https://mavmatrix.uta.edu/electricaleng_theses2/4