Research On State Estimation Of Electrical System Based On Orthogonal Signal Generator

The intelligent realization of modern power industry is mainly based on the flexible and rapid regulation of the system,which requires accurate estimation of the real-time state of the electrical system.Power grid and motor are important parts in the transmission and distribution of electrical system.This paper studies their state estimation methods from these two angles respectively.The equipment composition of power grid system is complex and there is a lot of interference,which requires the estimation method to have strong anti-interference.In the motor control system,the speed of the motor varies widely,so the estimation accuracy of frequency conversion signal is required.Phase-locked/frequency-locked loop based on quadrature signal generator(QSG)is a kind of signal estimation method,but its estimation performance under disturbance and frequency conversion is insufficient.Therefore,this paper studies the state estimation method of electrical system in wide frequency domain based on QSG.The main work and innovations are as follows:Firstly,the classic SOGI-QSG PLL is reviewed,its advantages and disadvantages are analyzed,and the dynamic performance of several improved SOGI-QSG under common disturbances is compared.The discrete modeling scheme is given,and the estimation effects of several improved schemes are compared,and the estimation performance of various methods under frequency hopping is emphatically compared,which lays the foundation for the subsequent research on frequency conversion state estimation.Secondly,through data analysis,it is found that in the traditional SOGI-QSG method,the frequency adaptive feedback loop will lead to the increase of system complexity.According to the idea of fixed frequency SOGI-PLL structure,a new FFCD-SOGI-PLL method is proposed.In this scheme,the QSG structure is redesigned to enhance its filtering ability,making it simple and stable.The experimental results show that this method can accurately estimate the state information of power grid under power grid disturbance,and has stronger DC suppression and harmonic suppression capabilities than traditional methods.Thirdly,the error generation principle of the traditional phase-locked loop estimation method for frequency conversion power grid in frequency slope estimation is derived.Based on this,three enhanced state estimation methods(SRF-EPLL,SOGI-EFLL,EROGI-FLL)are proposed.In the EROGI-FLL scheme,a SOGI-OSG parameter adjustment strategy based on unit delay function is proposed.The experimental results show that this method can stably estimate the frequency slope produced by motor operation and has better filtering performance.Finally,the simulation models of power grid state estimation and motor state estimation established in this paper are verified by experiments.The experimental results show that under various disturbances,the power grid state estimation scheme FFCD-SOGI-PLL proposed in this paper can accurately estimate the power grid state and has good interference suppression ability.The rotor state estimation scheme EROGI-FLL proposed in this paper can accurately estimate the speed slope and has strong anti-interference ability.The research results of this paper provide a new idea for the state estimation method of broadband signals in electrical system,and are of great significance to the intelligent power grid and the simplified and efficient development of motor control system.