Harmony search algorithm-based controller parameters optimization for a distributed-generation system

Mir Nahidul Ambia, Hany M. Hasanien, Ahmed Al-Durra, S. M. Muyeen

Research output: Contribution to journalArticlepeer-review

102 Scopus citations


This paper presents the utilization of a harmony search algorithm (HSA) to optimally design the proportional-integral (PI) controllers of a grid-side voltage source cascaded converter with two additional loops for smooth transition of islanding and resynchronization operations in a distributed-generation (DG) system. The first loop is the frequency-control loop which is superimposed on the real power setpoint of the cascaded controller of the voltage-source converter to minimize the frequency variation during the transition from the grid mode to islanding mode. The second loop is the resynchronization loop which reduces the phase shift of the ac voltages of the DG with the utility grid ac voltages during islanding operation leading to a successful grid reconnection event. The response surface methodology (RSM) is used to build the mathematical model of the system dynamic responses in terms of PI controllers' parameters. The effectiveness of the proposed PI control scheme optimized by the HSA is then compared with that optimized by both genetic algorithm and conventional generalized reduced gradient techniques. The HSA code is built using the MATLAB software program. The validity of the proposed system is verified by the simulation results which are performed using PSCAD/EMTDC.

Original languageBritish English
Article number6908020
Pages (from-to)246-255
Number of pages10
JournalIEEE Transactions on Power Delivery
Issue number1
StatePublished - 1 Feb 2015


  • Distributed generation (DG)
  • Genetic algorithm (GA)
  • Grid-side converter
  • Harmony search algorithm (HSA)
  • Proportional-integral (PI) controller


Dive into the research topics of 'Harmony search algorithm-based controller parameters optimization for a distributed-generation system'. Together they form a unique fingerprint.

Cite this