Research On Adaptive Compensation Strategy Of Harmonic Current Under Complex Conditions Of Weak Power Grid

Under the guidance of the strategy of "carbon peak,carbon neutralization",the new energy-distributed generation system represented by wind power and photovoltaic has been expanded on a large scale,and the penetration rate of new energy is increasing.Long-distance power transmission and multi-level voltage transformation lead to the increase of grid line impedance,which makes the grid present weak power grid characteristics.With the strengthening of weak power grid characteristics,the degree of grid voltage distortion,three-phase grid voltage unbalance,and grid frequency fluctuation and deviation will be increased.The three-phase voltage unbalance,grid voltage distortion,and frequency deviation of the common connection point constitute the background of the harmonic current compensation method studied in this dissertation,which is the complex working condition of the weak power grid.It poses a serious challenge to conventional power quality detection and control methods under complex conditions of weak power grids.Hence,this dissertation focuses on the problem of frequency and phase accuracy degradation of the three-phase synchronous phase-locked loop under complex working conditions of the weak grid,the problem that resonant current tracking controller cannot cope with large-scale frequency deviation and discretization leads to resonant pole offset,thereby reducing the tracking performance of harmonic compensation reference current,and the problem that the digitization of deadbeat control leads to the degr adation of harmonic current tracking performance and ensures the robust stability of control algorithm.The specific research contents are as follows:Aiming at the problem that frequency deviation under complex working conditions of a weak power grid leads to the decrease of phase and frequency estimation accuracy,a synchronous phase-locked method based on simplified gradient descent algorithm and adaptive lattice notch is proposed.The phase-locked method can accurately lock the phase when the grid-side voltage is unbalanced,the voltage distortion and the frequency has a certain range of deviation.At the same time,the use of simplified gradient descent algorithm can ensure the response performance of the phase-locked loop while reducing the amount of calculation.Accelerate the calculation speed.The frequency adaptability of the proposed Adaptive Lattice Notch Filter Phase-Locked Loop(ALNF-PLL)is proved by simulation experiments,which provides theoretical support and technical support for the high-performance harmonic compensation of Active Power Filter(APF)under complex conditions of the weak power grid,and has certain engineering application value.Aiming at the problem that the tracking performance of harmonic reference current under complex working conditions of a weak grid is affected by frequency deviation,a current tracking control strategy of frequency adaptive Proportional-Integral(PI)+ Vector Proportional Integral(VPI)is proposed.This strategy can adaptively adjust the frequency change of the power grid and improve the steady-state performance and dynamic performance of harmonic compensation.Considering that the cosine term generated by the discretization of the transfer function of the VPI controller is not conducive to online calc ulation,a structure of the VPI controller based on the double integrator is adopted.The high-order Taylor series expansion of the cosine term is introduced to improve the VPI controller of the above double integrator structure,and the appropriate Taylor expansion order is selected through the quantitative calculation to improve the resonant pole offset.At the same time,this strategy introduces the phase and frequency estimation of the above ALNF-PLL into the PI + VPI current control in synchronous rotating coordinates in real time to improve the frequency adaptability of harmonic compensation.The experimental results show the harmonic compensation performance of the proposed control strategy under the conditions of unbalanced,distorted,and frequency deviation of three-phase grid voltage.Aiming at the problem that the deadbeat current tracking control under complex working conditions is a deadbeat control in the digital implementation process,a hybrid current predictive deadbeat tracking control meth od is proposed.The reference current prediction link comprehensively uses the adaptive forward linear prediction and the repetitive prediction algorithm to give the harmonic compensation reference current at two sampling moments in advance.The switching of the two current prediction algorithms is realized by setting the dynamic occurrence criterion of the load.The inverter output current prediction link uses both the inverter output current historical data and the harmonic compensation reference current data to achieve effective prediction of the inverter output current while taking into account the measurement noise suppression.Considering the strong mathematical model dependence and weak anti-high frequency interference ability of the deadbeat control method,the robust stability and measurement noise transmission of the control method is further analyzed under the uncertainty of the output filter inductance and its equivalent resistance parameters.The steady-state compensation accuracy and dynamic response speed of the proposed hybrid current predictive deadbeat control method under steady-state and load mutation conditions are verified by simulation and experiment.The harmonic current performance is verified under the condition of uncertain output filter inductance and equivalent resistance parameters,and the robust stability of the control system is verified.The experimental system of 60 k VA three-phase three-wire shunt APF based on a rapid control prototype is built.Based on the experimental syst em,the performance test of ALNF-PLL under complex conditions of a weak power grid is carried out.The experimental verification of the adaptive PI + VPI harmonic compensation control strategy and the experimental verification of harmonic compensation performance based on the hybrid prediction deadbeat control method prove the effectiveness and superiority of the proposed method or strategy.