Harmonic And Inter-harmonic Detection Of Power System Based On The Improved Synchrosqueezing Wavelet Transform

The increase of nonlinear load is prone to generate harmonics and inter-harmonics in the power system,which makes the power system unstable and leads to power quality problems.Accurate detection of harmonics and inter-harmonics in the power system is a foundation for solving power quality problems.Synchrosqueezing wavelet transform(SWT)has good modal decomposition and noise robustness,and it can decompose a signal into a series of physically meaningful intrinsic-mode-type(IMT)functions,which has been widely used in the signal detection of power system.However,the parameters of basic wavelet in this method affect the decomposition results.In order to improve the detection accuracy of harmonics and inter-harmonics,an improved SWT method is proposed in this paper.The improved method uses permutation entropy(PE)as the fitness function,and the optimal parameters of basic wavelet are determined by adaptively searching the PE value of the component IMT1 under different parameters based on whale optimization algorithm(WOA),where the minimum PE corresponds to the optimal parameters.Then,the number of components in the SWT method is determined by retaining the components with relatively high energy according to the energy change of components.In addition,signal waveform is predicted from the forward and reverse directions by long short-term memory(LSTM)network,respectively,and end effect is transferred to the predicted domain so that the end effect in current domain is improved by signal extension,and then the improved SWT method is obtained.Finally,Hilbert transform is performed on the IMT components extracted by this method,and the frequencies and amplitudes of components are measured to detect harmonics and inter-harmonics.The effectiveness of the improved SWT is verified by some different examples.The results show that compared with empirical modal decomposition(EMD),variational modal decomposition(VMD),time varying filtering based empirical mode decomposition(TVFEMD),short time Fourier transform(STFT),Kalman filter(KF)and estimation of signal parameters via rotational invariance techniques(ESPRIT),the improved SWT has better detection results on transient signals.The improved SWT not only retains good noise robustness of original SWT,but also improves modal mixing and end effect.It meets the accuracy requirements of the prevailing standard for harmonic and inter-harmonic detection,and the improved SWT has stronger modal separation ability and higher detection accuracy.