Harmonic Current Detection Method Based On Adaptive Neural Network Research

Enhances the harmonic current compensation effect of the active power filter (APF), first must use the suitable method for harmonic current detection, and causes harmonic current detection circuit real time, the detection precision to be high. When the non-linearity random load current changes, the detection circuit has the good adaptive and the track detection.The creation of the electric power harmonic, endanger and management, are discussed and analyzed. Two harmonic and reactive current detection methods are introduced in this paper, the method based on the instantaneous reactive theory and method based on adapted noise cancellation neural network.In this paper, it is analysis the adaptive linear neural model and the harmonic current, has obtained the harmonic current detection method based on the adaptive neural network (ADNN), and simply this method. In the research detection time delay of APF, the paper carries on the comparison the harmonic current detection method based on the adaptive neural network with the harmonic current detection method based on the instantaneous reactive theory and based on the adapted noise cancellation neural network. Finally the harmonic current method based on the adaptive neural network has proven the good real time by the simulation and the experimentation.In order to confirm harmonic current detection method based on adaptive neural network real time and the detection precision, the paper carry on harmonic current detection method based on the instantaneous reactive theory and based on he adapted noise cancellation neural network, and has established the simulation model separately. Then carried on the harmonic current detection to the identical load current, has carried on the analysis to its real time and the detection precision. Indicated compared with the result, the harmonic current detection method based on the adaptive neural network has the good real time and the detection precision.