Reduced Modeling of Unbalanced Radial Distribution Grids in Load Area Framework

Giovanni Mercurio Casolino,Arturo Losi

doi:10.1109/access.2020.3027990

Giovanni Mercurio Casolino, Arturo Losi

Open Access

https://doi.org/10.1109/access.2020.3027990

Copy DOI

Abstract

A simplified representation of a distribution network offers relevant advantages in limiting the amount of information processed and reducing the computational burden necessary for its treatment, in particular in planning studies, real-time simulations, monitoring and control. The identification of the relevant information for obtaining a correct reduced modeling of a distribution network results from the Load Area (LA) approach. It is based on the concept of groups of nodes/prosumers whose power injection has a similar impact on the operating conditions of distribution grids; the groups are identified through overload and voltage indexes. The paper presents a method for obtaining the reduced equivalent modeling of unbalanced radial grids in the LA framework, once the LAs have been identified. It is based on the backward-forward graph navigation technique for radial grids and a generalization of the modeling of single-phase and two-phase branches. The application and discussion of the method for five different size test cases highlight the computational issues related to the representation and solution of the reduced networks. The results show the feasibility of the proposed method and highlight the computational gains deriving from its adoption.

Highlights

Distribution systems are profoundly involved in the change of the electricity industry towards higher flexibility, accessibility, reliability, affordability
To get a model of a distribution system that is accurate as needed without being unduly detailed, the concept of Load Area (LA) can be usefully adopted; a LA is a group of prosumers/nodes whose power injection has a similar impact on the operating conditions of the grid
The present work is focused on distribution systems within the LA approach, which allows identifying the homogeneous parts of the grid whose representation can be simplified by a reduced equivalent model

Summary

INTRODUCTION

Distribution systems are profoundly involved in the change of the electricity industry towards higher flexibility, accessibility, reliability, affordability. To get a model of a distribution system that is accurate as needed without being unduly detailed, the concept of Load Area (LA) can be usefully adopted; a LA is a group of prosumers/nodes whose power injection has a similar impact on the operating conditions of the grid. The present work is focused on distribution systems within the LA approach, which allows identifying the homogeneous parts of the grid whose representation can be simplified by a reduced equivalent model. Reference [11] proposes a general network reduction method that groups nodes based on congestions only, while the current study considers voltage issues and is specialized for unbalanced radial distribution systems. A method to get the equivalent reduced modeling of an unbalanced radial distribution network is presented, starting from the identification process of LA subgrids proposed in [14]. The relationships between active and reactive powers for different prosumers’ categories can be found in [14]

NODAL INJECTIONS

LOAD AREA NETWORK REPRESENTATION

BACKWARD STAGE

NUMERICAL RESULTS

CONCLUSION