Genetic algorithm based reference current control extraction based shunt active power filter.

Research output: Contribution to journal/Conference contribution in journal/Contribution to newspaperJournal articleResearchpeer-review


  • Elango Sundaram
  • ,
  • Manavaalan Gunasekaran
  • ,
  • Ramakrishnan Krishnan
  • ,
  • Sanjeevikumar Padmanaban, Aalborg University
  • ,
  • Sharmeela Chenniappan
  • ,
  • Ahmet H. Ertas
Traditional approaches towards proportional‐integral (PI) controller tuning often fail to provide optimum gain values in situations where shunt active power filter (SAPF) is connected to systems containing complex, dynamic, and nonlinear loads. Optimum gain values are, however, crucial in the generation of compensating currents with less transient and steady‐state error, that would nullify the harmonic currents in a short time. This work proposes two soft computing techniques, genetic algorithm (GA) and Queen Bee assisted GA (QBGA) for better controller tuning to obtain optimum gain values to switch SAPF. These algorithms are used in local search technique mode to arrive at the optimal solutions based on the desired characteristics. The PI controller controls the voltage of the DC capacitor to generate the required compensating current. The proposed algorithms are practical since reliable solutions are obtained with a limited number of iterations. Implementation of the suggested algorithm reduces the THD of supply current to less than 5%, in compliance with IEEE‐519 standards. The system performance is evaluated through MATLAB simulation tool. Suitable hardware model is also developed and tested for validating the simulation results. The hardware results are found in close agreement with simulation results. The highlight of this work is the introduction of QBGA algorithm as a novel technique for tuning of PI controller for SAPF.
Original languageEnglish
Article numbere12623
JournalInternational Transactions on Electrical Energy Systems
Number of pages22
Publication statusPublished - 1 Jan 2021
Externally publishedYes

    Research areas

  • DESIGN, HARMONICS, genetic algorithm, pulse width modulation, queen beegenetic algorithm, shunt active power filter, total harmonic distortion

See relations at Aarhus University Citationformats

ID: 218928428