This paper presents a single-phase shunt active 7 power filter (APF) for current harmonic compensation based on 8 neural filtering. The shunt active filter, realized by a current9 controlled inverter, has been used to compensate a nonlinear 10 current load by receiving its reference from a neural adaptive 11 notch filter. This is a recursive notch filter for the fundamental 12 grid frequency (50 Hz) and is based on the use of a linear adaptive 13 neuron (ADALINE). The filters parameters are made adaptive 14 with respect to the grid frequency fluctuations. A phase-locked 15 loop system is used to extract the fundamental component from the 16 coupling point voltage and to estimate the actual grid frequency. 17 The current control of the inverter has been performed by a 18 multiresonant controller. The estimated grid frequency is fed to 19 the neural adaptive filter and to the multiresonant controller. In 20 this way, the inverter creates a current equal in amplitude and 21 opposite in sign to the load harmonic current, thus producing an 22 almost sinusoidal grid current. An automatic tuning of the mul23 tiresonant controller is implemented, which recognizes the largest 24 three harmonics of the load current to be compensated by the APF. 25 The stability analysis of the proposed control system is shown. 26 The methodology has been applied in numerical simulations and 27 experimentally to a properly devised test setup, also in comparison 28 with the classic sinusoidal current control based on the P-Q theory
Current Harmonic Compensation by a Single-Phase Shunt Active Power Filter Controlled by Adaptive Neural Filtering
Marcello Pucci;Gianpaolo Vitale;
2009
Abstract
This paper presents a single-phase shunt active 7 power filter (APF) for current harmonic compensation based on 8 neural filtering. The shunt active filter, realized by a current9 controlled inverter, has been used to compensate a nonlinear 10 current load by receiving its reference from a neural adaptive 11 notch filter. This is a recursive notch filter for the fundamental 12 grid frequency (50 Hz) and is based on the use of a linear adaptive 13 neuron (ADALINE). The filters parameters are made adaptive 14 with respect to the grid frequency fluctuations. A phase-locked 15 loop system is used to extract the fundamental component from the 16 coupling point voltage and to estimate the actual grid frequency. 17 The current control of the inverter has been performed by a 18 multiresonant controller. The estimated grid frequency is fed to 19 the neural adaptive filter and to the multiresonant controller. In 20 this way, the inverter creates a current equal in amplitude and 21 opposite in sign to the load harmonic current, thus producing an 22 almost sinusoidal grid current. An automatic tuning of the mul23 tiresonant controller is implemented, which recognizes the largest 24 three harmonics of the load current to be compensated by the APF. 25 The stability analysis of the proposed control system is shown. 26 The methodology has been applied in numerical simulations and 27 experimentally to a properly devised test setup, also in comparison 28 with the classic sinusoidal current control based on the P-Q theoryI documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
