Tuning of a PI-MR Controller Based on Differential Evolution Meta heuristic Applied to the Current Control Loop of a Shunt-APF Simulation Projects



This paper aims to present an alternative methodology to improve the tuning of proportional integral multi-resonant (PI-MR) controller applied to the current control loop of a shunt active power filter (SAPF) by using differential evolution (DE) metaheuristic method. For computing the PI-MR controller gains, the methodology based on Naslin polynomial (NP) can be employed, although its performance is limited. On the other hand, tuning procedures accomplished by trial-and error methods have been largely used. In order to fill the lack of methodological procedures to determine the PI-MR controller gains, the use of DE metaheuristic optimization algorithm is proposed to perform an optimal adjustment. The DE algorithm operates minimizing an appropriate cost function, taking into account the total harmonic distortion of the source current, the error of the compensation\ current control loop, and the saturation limit of the control action. The effectiveness of the proposed methodology is compared to the NP-based method, applied to a single phase SAPF. The proposed methodology is validated by means of both simulation and experimental results. The evaluation of the static and dynamic performances is obtained from experimental tests performed based on digital signal processor.


  1. Differential evolution
  2. Multi-resonant controller
  3. Optimization
  4. Shunt active power filter.



Fig. 1. Single-phase SAPF scheme.


Fig. 2. Simulation results of SAPF operating with Load 1 (utility voltage vs(200V/div), utility current is and compensation current ic(20A/div), load current iL(30A/div)): (a) PI controller; (b) PI-MRNP controller; (c) PI-MRDE controller.

Fig. 3.Simulation results of SAPF operating with Load 2 (utility voltage vs(200V/div), utility current is and compensation current ic(20A/div), load current iL(30A/div)): (a) PI controller; (b) PI-MRNP controller; (c) PI-MRDE controller.

Fig. 4. Harmonic spectra and THD of the source current obtained from simulation results with PI, PI-MRNP and PI-MRDE controllers: (a) Load 1; (b) Load 2.


This paper proposes an alternative methodology using DE metaheuristic optimization algorithm for obtaining systematic procedures to achieve the PI-MR controller gains employed in the current control loop of a SAPF. A comparative analysis involving the classical PI controller and two other PI-MR controllers was performed. The classical PI controller was tuned using the well-known frequency response method, while the MR controllers were tuned considering the NP-based approach, as well as DE metaheuristic optimization algorithm.Extensive simulation and experimental results proved that superior performance is achieved when the SAPF operates with the DE-based PI-MR controller. Therefore, it can be concluded that the DE metaheuristic method represents a promising approach for tuning PI-MR controller for active power filtering applications.


  • C. Dugan, M. F. McGranaghan, S. Santoso, H. W. Beaty, Electrical Power Systems Quality, 3rd. ed., McGraw-Hill Education, USA, 2012.
  • A. Modesto, S. A. O. Silva, and A. A. Oliveira Jr., “Versatile unified power quality conditioner applied to three-phase four-wire distribution systems using a dual control strategy,” IEEE Trans. Power Electron., vol. 31, no. 8, pp. 5503–5514, Aug. 2016.
  • A. Angélico, L. B. G. Campanhol, and S. A. O. Silva, “Proportional integral/ proportional-integral-derivative tuning procedure of a single phase shunt active power filter using bode diagram,” IET Power Electron., vol. 7, no. 10, pp. 2647–2659, Aug. 2014.
  • Singh, K. Al-Haddad, and A. Chandra, “A review of active filters for power quality improvement,”IEEE Trans.Ind. Electron., vol.46, no.5, pp.960–971, Oct. 1999.
  • Vodyakho, T. Kim, S. Kwak, and C. Edrington, “Comparison of the space vector current controls for shunt active power filters,” IET Power Electron., vol. 2, no. 6, pp. 653–664, Nov. 2009.


Leave a Reply

Your email address will not be published.