基于RBF神经网络的谐波电流检测方法

Harmonic Current Detection Method Based on RBF Neural Network

静止无功发生器(SVG)是一种广泛应用的动态无功补偿装置,其一般采用传统的ipiq运算方式进行无功谐波电流检测。传统的ip-iq电流检测方法存在检测相位滞后和误差等问题,造成补偿效果不理想。在分析无功补偿理论的基础上,提出了一种基于RBF神经网络的谐波电流检测方法,克服了传统的ip-iq谐波电流检测方法存在的问题。并利用MATLAB对2种电流检测方法的补偿效果进行了数字仿真,验证了RBF神经网络的谐波电流检测方法的可行性、快速性和准确性。

Static var generator （SVG） is a kind of dynamic reactive power compensation device which is widely used, and it usually uses the traditional ip-iq operation mode to detect reactive harmonic current. The traditional ip-iq current detection method has the problems of detecting phase lag and error, and the compensation effect is not ideal. Based on the analysis of reactive power compensation theory, a method of harmonic current detection based on RBF neural network is proposed, which overcomes the problems of traditional ip-iq harmonic current detection method. The simulation results of two kinds of current detection methods using MATLAB are given, which verify the feasibility, rapidity and accuracy of the harmonic current detection method based on RBF neural network.