Load margin assessment of systems with distributed generation with the help of a neuro-fuzzy method

• Published in: IET renewable power generation 2015-05, Vol.9 (4), p.331-339
• Main Authors: Zambroni de Souza, Matheus Ferreira, Reis, Yuri, Almeida, Adriano Batista, Lima, Isaias, Zambroni de Souza, Antonio Carlos
• Format: Article
• Language: English
• Subjects: Artificial neural networks; Assessments; Buses (vehicles); Computer simulation; CRIC method; distributed generation; distributed power generation; Electric potential; Fuzzy logic; fuzzy neural nets; fuzzy reasoning; fuzzy set theory; IEEE 34‐bus system; load margin assessment; Load margins; modified Brazilian real system; Monte Carlo methods; Monte Carlo simulation; neuro‐fuzzy methodology; power engineering computing; power system security; voltage security problems
• ISSN: 1752-1416; 1752-1424; 1752-1424
• DOI: 10.1049/iet-rpg.2014.0090

Voltage security problems became a matter of concern over the last three decades because of the systems complexity. This is about to worsen as the penetration of renewable sources grows. Different operating scenarios must be addressed because of the intermittent nature of some sources. This study discusses this problem by proposing a neuro-fuzzy methodology to determine the load margin when the intermittency of the sources is taken into account. Load margin is obtained by the continuation method enhanced by the Constrained Reactive Implicit Coupling (CRIC) method, so its computational effort is reduced. Monte Carlo simulation is employed to generate the bunch of data considered by the neuro-fuzzy and the results are obtained in two ways. First, a sample IEEE 34-bus system is employed, so the results may be reproduced. Then, a modified Brazilian real system with 115 buses is considered.