Power Flow Analysis of Networks with Loads of Constant Current - Constant Power Factor Type

Modelling loads as constant current loads is a common technique used for load flow analysis of distribution networks, however this technique is not able to represent the power factor of loads correctly. Typically, the angle of the current representing a load is set relative to the angle of slack bus...

Full description

Saved in:
Bibliographic Details
Main Authors: Hoger, Marek, Siranec, Marek, Altus, Juraj
Format: Conference Proceeding
Language:English
Subjects:
Admittance
constant current amplitude
constant power factor
Generators
Load flow
Load modeling
load type
Loading
low voltage network
Mathematical model
Reactive power
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Modelling loads as constant current loads is a common technique used for load flow analysis of distribution networks, however this technique is not able to represent the power factor of loads correctly. Typically, the angle of the current representing a load is set relative to the angle of slack bus voltage and not to the related nodal voltage, which is unknown prior the calculation. In networks with high R:x ratio and limited reactive power flow, the resulting error in power factor of individual loads is often relatively small and can be neglected. However, if the power factor of the loads has to be respected during calculation with high precision, modification of the standard calculation algorithms is necessary. This paper presents the modifications of standard load flow algorithms necessary to allow load flow analysis of networks with loads of constant current amplitude - constant power factor type. Algorithms presented in this paper were tested on a set of simple test networks with variable size to verify, how the presented modifications of standard implementations are affecting the convergence characteristics of the algorithms and the absolute accuracy of results. Experimental results have confirmed that all presented algorithms are able to provide accurate results and the impact on the speed of convergence is neglectable.
ISSN:2376-5631
DOI:10.1109/EPE.2019.8777975