Modal Identification Of Power System Low Frequency Oscillation And Coordinated Optimization Design Of Stabilizer

With the continuous development and interconnection of contemporary power networks,the security and stability of interconnected systems are greatly challenged,especially due to weak damping or under-damping of power networks,which leads to the lack of sufficient damping torque for the occurrence of oscillation accidents and the inability to effectively and quickly suppress low-frequency oscillation.For interconnected systems,if the oscillation continues or diverges,it may lead to system cracking.Therefore,it is of great practical significance for maintaining the safe and reliable operation of the power grid to accurately find out the modal parameters that play a key role in the oscillation of the system,and it is also the necessary basis for the subsequent optimization of PSS and the construction of optimization objective function.Therefore,aiming at the current security and stability of power grid,an oscillation parameter identification method based on the combination of Hilbert transform(HT)and empirical mode decomposition of energy function(EFEMD)is proposed.Firstly,the initial signal with noise interference is decomposed by modal decomposition technology(EMD)to obtain the decomposed small signals.Then,the energy of each natural modal function is calculated and weighted by EFEMD-HT energy method,and the dominant oscillation mode of the system is selected.Finally,the key parameters are extracted by HT transform.At the same time,a joint identification method of key modal parameters is proposed,which combines adjustable Q-factor wavelet transform(TQWT)and sparse time domain method(STD).Firstly,TQWT is used to preprocess the noise of the signal containing noise,and then STD method is used to extract modal parameters.Finally,the validity and feasibility of the EFEMD-HT energy method and TQWT-STD method proposed in this paper are verified by a variety of test signals.The results show that the joint algorithm has better noise removal ability,and the time required in the identification process is shorter and the identified parameters are more accurate.In order to prevent further expansion of the fault range and to avoid serious system divergence oscillation and even collapse,we usually adopt the method of installing the economical and efficient power system stabilizer(PSS),however,the PSS parameters selected by traditional methods usually can not exert the effect of suppressing lowfrequency oscillation to the greatest extent,which leads to long suppression time and continuous occurrence of oscillation.Therefore,when installing PSS,we must optimize the parameters of the stabilizer.By using the EFEMD-HT energy method or TQWTSTD method proposed above to obtain the key parameters of the oscillation mode of the system,the objective function is constructed to maximize the minimum damping ratio of the system,in order to find the optimal control parameter combination to restrain the low frequency oscillation of the system,the parallel system model and the parameter optimization algorithm are proposed.In order to find the optimal parameters,this paper introduces the social learning mechanism into the Particle swarm optimization to realize the optimal design of PSS parameters,the existing optimization effect is poor,can not find the optimal value and so on.Through the optimization of the parameters,the PSS optimization effect,the oscillation suppression speed and the stability in the suppression process have been greatly improved.