基于神经网络的电力系统谐波测量方法研究

提出了一种新的基于三角基函数神经网络的电力系统谐波测量方法，给出了该神经网络算法的收敛定理，并采用加窗插值算法修正基波频率的准确度．该方法不需要同步采样和整周期截断，可一次性获得电力系统基波及各次谐波的频率、幅值和相位．计算机仿真结果表明，该方法计算精度高，计算量小，收敛速度快．

Study on Harmonic Analysis in Power Systems Based on Neural Network

A new harmonic analysis method for electric power systems based on triangle basis functions neural network was presented,the convergence theorem of the algorithm was proposed,and a window function and interpolation algorithm were employed to correct the frequency of fundamental waves.This approach doesn't require synchronized sampling and integer period truncation,and can obtain directly the frequencies,amplitudes and phases of fundamental waves and harmonics.The result of computer simulations has shown that the algorithm is an ideal analysis method with high precision,small amount of calculation and speedy convergence.