Customer loads connected to electricity supply systems may be broadly categorized as either linear or nonlinear. Nonlinear loads inject harmonics in a power distribution network. The interaction of the nonlinear load harmonics with the network impedances creates voltage distortions at the point of common coupling (PCC) which in turn affects other loads connected to the same PCC. When several nonlinear loads are connected to the PCC, it is difficult to predict mathematically how each nonlinear load is affecting the voltage distortion level at the PCC. Typically, customers with nonlinear loads apply harmonic filtering techniques to clean up their current and avoid penalties from the utility. When corrective action is taken by the customer, one important parameter of interest is the change in the voltage distortion level at the PCC due to the corrective action of the customer. This paper proposes a new method based on neural networks to predict the change in the distortion level of the voltage at the PCC if the customer were to draw only fundamental current and filter out its harmonics. The benefit of the proposed method is that it would indicate the impact of the customer's front end filters on the voltage distortion at the PCC without actually having to install the filters. This paper presents the results of the proposed method applied to actual industrial sites.

Meeting Name

2007 IEEE Industry Applications Conference, 2007. 42nd IAS Annual Meeting


Electrical and Computer Engineering


Georgia Power Company

Keywords and Phrases

Neural Networks; Power Quality; Power System Harmonics; Source Modeling; Harmonic analysis

Document Type

Article - Conference proceedings

Document Version

Final Version

File Type





© 2007 Institute of Electrical and Electronics Engineers (IEEE), All rights reserved.

Publication Date

01 Sep 2007