Power System Harmonic Detection Based On FFT And Wavelet Packet Transform

With the increasing power grid network capacity and increasingly complex power system structure, more and more non-linear loads and electrical equipment used in power systems, resulting in a large number of harmonic currents, affecting the quality of power and safe operation of the user equipment, it can cause severe burning of the capacitor device.Therefore, how to effectively suppress power harmonic becomes many scholars’ research topics. The main reason of power grid pollution is the reactive power loss and harmonic interference, therefore, reducing or even eliminating the grid harmonic interference and compensating reactive power can achieve the purpose of effectively suppressing power harmonic.Eliminating the effects of harmonic on the power grid must analysis harmonic, and suppressing harmonic based on the analysis. Therefore, the analysis of power harmonic is important for the maintenance of grid safety and meaningful for the harmonic suppression.Firstly, the paper studied the Fourier transform harmonic detection algorithm, mainly studied the basic principle of this algorithm, it has the disadvantage of non-synchronous sampling, barrier effect, etc. This paper used the Fourier transform which is based on the Blackman window algorithm to detect the power harmonic. Simulation results show that Fourier transform with Blackman window harmonic detection algorithm can effectively solve the shortcomings, while compared with detection of no windowed Fourier harmonic, power system harmonic detectionaccuracy has been improved.Secondly, this paper studied the wavelet packet transform and wavelet transform basic principle, compared the differences between wavelet packet transform and wavelet transform, using the wavelet packet transform which has higher accuracy to analysis the harmonic because Fourier transform can’t analysis the transient harmonic, and use the Fourier transform with Blackman window combined with wavelet packet transform to analysis the harmonic. Simulation results show that the algorithm can analysis the transient signal with damping component and obtain the amplitude and frequency of the signal and the time of damping component occurred.Finally, this paper studied the noise of power system harmonic,according to advantages of the traditional soft threshold function and hard threshold function, combining their advantages to denoise the noisy signal.Simulation results show that compared with traditional soft threshold fuction or hard threshold function, this algorithm’s mean spuare error is smaller, noise ratio is higher.