Abstract
Microgrids are the cornerstone for a new model of electrical generation based on renewable resources. Commonly microgrids are controlled with a centralised hierarchical structure, which is inherited from power systems. However, a time-scale separation between traditional fast frequency restoration and slow economic dispatch may be counterproductive in the long run because the slow long-term economic dispatch increases the prediction uncertainty. In an effort to improve the economical operation of microgrids, this work proposes a distributed model predictive control strategy for the operation of isolated microgrids based on a consensus strategy that tackles both the economic dispatch and frequency restoration over the same time-scale. The proposed controller can operate without knowledge of the microgrid's topology: instead, typical local measurements and other information from neighboring generation units are required. Experimental results demonstrate that the controller is robust to load variations and communication issues, but the plug-and-play nature of the system is preserved.
Funding Information
  • Fondo Nacional de Desarrollo Científico y Tecnológico (1170683)
  • Solar Energy Research Center SERCChile (ANID/FONDAP/15110019)
  • Instituto Sistemas Complejos de Ingeniería (ANID PIA/BASAL AFB180003)
  • ANID BECAS/DOCTORADO NACIONAL (2019-21190961)
  • Secretaría de Educación Superior, Ciencia, Tecnología e Innovación de Ecuador (SENESCYT/ARSEQBEC-0058482018)

This publication has 30 references indexed in Scilit: