Quantifying the performance of the Kim AC/DC Hybrid Power Flow Model

Foster, Peter

Permalink

https://hdl.handle.net/2142/42254 Copy

Description

Title

Quantifying the performance of the Kim AC/DC Hybrid Power Flow Model

Author(s)

Foster, Peter

Issue Date

2013-02-03T19:29:18Z

Director of Research (if dissertation) or Advisor (if thesis)

Overbye, Thomas J.

Department of Study

Electrical & Computer Eng

Discipline

Electrical & Computer Engr

Degree Granting Institution

University of Illinois at Urbana-Champaign

Degree Name

M.S.

Degree Level

Thesis

Keyword(s)

• Power Flow Model
• Newton Method
• DC Power Flow Model

Abstract

The landscape of electric utilities is changing at an unprecedented rate. With the rise of nondispatchable renewable generation technologies utilities are in the process of aggressively modifying their control and modeling techniques to cope with increasingly shorter forecast horizons. This analysis examines one proposed modeling methodology, the Kim Hybrid Power Flow Model, looking to quantify its accuracy as compared to the two most common steady-state power flow modeling techniques. For the analysis, a 37-bus system was selected as the common modeling platform in order to facilitate like-for-like comparisons. With an emphasis on minimizing both the system topology and operating point’s impact on the comparison, numerous configurations were examined. For each condition, the system was modeled with each technique to allow for as objective a comparison as possible given the single system. A total of seven case studies were performed to isolate and examine the system’s sensitivity to certain physical parameters and operating conditions. These studies, combined with 24 selected hybrid topologies, were used to perform a comprehensive examination of the Hybrid model’s performance as it compares to the AC and DC models. This led to the assessment that the model was noticeably sensitive to the number of Boundary buses and the power flow across the boundary between the inner (AC region) and outer (DC region) areas. In order to make use of these observations, a number of Hybrid topology selection guidelines can be drawn: 1) Select regions such that the number of Boundary buses is minimized. This is achieved functionally by bisecting as few transmission lines as possible to isolate the regions of interest. 2) Select regions so as to minimize the power flow across the Boundary buses. This is more complicated in practice, especially if multiple inner and outer regions are being considered. The approach relies on selecting the topology such that the major loads and generation remain in the AC region. 3) Once the preliminary analysis is performed, repeat step (2) as necessary (i.e. as larger Boundary flows are identified) to reduce Boundary power flows. While this methodology is essentially identical to that originally assumed from a qualitative standpoint, the two selection criteria shape the manner in which the original method can be employed. Areas of concern will always be those of significant generation and loading, but this analysis points out that the transmission route between the two is just as important to the overall accuracy of the model. In general, the Kim hybrid model has many promising characteristics that should be examined in future research. The computational benefits and accuracy improvements need to be examined on larger and more complex systems. The one significant drawback identified is that the model (at least for this system) was no more stable or convergent than the AC model, and at times was more sensitive to the loading conditions than the AC model.

Graduation Semester

2012-12

Permalink

http://hdl.handle.net/2142/42254

Copyright and License Information

Copyright 2012 Peter Foster

Owning Collections