Neuro Fuzzy based controller for Power Quality Improvement

Neuro Fuzzy based controller for Power Quality Improvement


Use of power electronic converters with nonlinear loads leads to power quality problems by producing harmonic currents and drawing reactive power. A shunt active power filter provides an elegant solution for reactive power compensation as well as harmonic mitigation leading to improvement in power quality. However, the shunt active power filter with PI type of controller is suitable only for a given load. If the load is varied, the proportional and integral gains are required to be fine tuned for each load setting. The present study deals with hybrid artificial intelligence controller, i.e. neuro fuzzy controller for shunt active power filter. The performance of neuro fuzzy controller over PI controller is examined and tabulated. The salvation of the problem is extensively verified with various loads and plotted the worst case out of them for the sustainability of the neuro fuzzy controller.



  1. Active Power Filter
  2. Neuro Fuzzy Controller
  3. Back Propagation Algorithm
  4. Soft Computing





Fig 1. Schematic Diagram of Shunt Active Power Filter




Fig 2. (a) Waveform of Load Current, Compensating Current, Source

Current and Source Voltage for Case V of Table1 (1kVA with α=60o) and

(b) Waveform of Source Voltage and in phase Source Current of Fig. (a) Reproduced



The application of hybrid artificial intelligence technique on shunt active power filter is proved to be an eminent solution for the mitigation of harmonics and the compensation of reactive power. The hybrid artificial intelligence used here is the neuro fuzzy controller. It takes the linguistic inputs as a fuzzy logic controller and it adapts any situation in between the running of the program as the neural network. The simulation results states that the active power filter controller with neuro fuzzy controllers have been seen to eminently minimize harmonics in the source current when the load demands non sinusoidal current, irrespective of whether the load is fixed or variable when compared to PI Controller. Simultaneously, the power factor at source also becomes the unity, if the load demands reactive power. The neuro fuzzy controller is far superior to the PI controller for all the loads. In the present work, a range of values of the load is considered to robustly test the controllers. It has been demonstrated that neuro fuzzy controller offers more acceptable results over the PI controller. The neuro fuzzy controller, therefore, significantly improves the performance of a shunt active power filter.



  • Laszlo Gyugyi, “Reactive Power Generation and Control by Thyristor Circuits”, IEEE Transactions on Industry Applications, vol. IA-15, no. 5, September/October 1979.
  • Akagi, Y. Kanazawa, and A. Nabae, “Instantaneous reactive power compensators comprising switching devices without energy storage components,” IEEE Transaction Industrial Applications, vol. IA-20, pp. 625-630, May/June 1984.
  • Z. Peng, H. Akagi, and A. Nabae, “A study of active power filters using quad series voltage source pwm converters for harmonic compensation,” IEEE Transactions on Power Electronics, vol. 5, no. 1, pp. 9–15, January 1990.
  • Conor A. Quinn, Ned Mohan, “Active Filtering of Harmonic Currents in Three-phase, Four-Wire Systems with Three-phase and Single-phase Non-Linear Loads”, IEEE-1992.
  • A. Morgan, J. W. Dixon, and R. R. Wallace, “A three-phase active power filter operating with fixed switching frequency for reactive power and current harmonic compensation,” IEEE Transactions on Industrial Electronics, vol. 42, no. 4, pp. 402–408, August 1995.

Leave a Reply

Your email address will not be published. Required fields are marked *