Research On Model Predictive Control Algorithm For Interline Power Flow Controller

With the increase of power load,combined with the upscaling of China’s power system,the architecture of the power system has grown more complex than ever before,and the stability of the power system has faced a huge challenged.The power flow control device represented by the UPFC(Unified Power Flow Controller)has already been putted into operation,which greatly improves the stability of the power system.IPFC can control the power flow of multiple lines at the same time and IPFC can continue to control the power flow of the remaining non-fault lines by switching the control mode after the controlled line break down.Compared with UPFC,IPFC has a higher reliability than UPFC.At the same time,because model predictive control has the advantages of fast dynamic response and strong ability to deal with non-linear systems,this thesis carried out research on the control strategy of IPFC based on model predictive control.In this thesis,we first analysis the working principle of the IPFC,developing mathematical models.Because of the high model accuracy required by the model prediction and the model of IPFC we predicted has some nondeterminacy,it will have a great influence on it’s control effect,for this purpose,this thesis adopts the parameter identification method based on FFRLS to identify the parameters of the IPFC.The mathematical model of IPFC is modified by FFRLS in real time,which improves the precision of power control and reduces the current total harmonic distortion rate.Simulation and experimental results verified the proposed method is correct.At present,most of the IPFC topology in the literature is connected to the grid through a single L filter.Compared to the L filter,the LCL filter has the advantages of small size,light weight,and strong high-frequency harmonic suppression.Therefore,this paper use LCL filter to suppress harmonics instead of the L filter.In order to solve the inherent resonance problem of the LCL filter,the traditional model predictive current control strategy with active damping to equate the converter and the converter side inductors as an ideal current source,ignoring the influence of the grid-side inductance change and current control delay on the selection of the damping coefficient.Therefore,by constructing an optimization function including the inductance,delay link and damping coefficient of the grid side,a second-order system optimization damping coefficient control strategy is proposed,which takes into account the current control delay and the inductance variation of the grid side.It still have a better dynamic response and damping effect when the grid-side inductance changes.Further considering the influence of the switching action of the converter,the converter and converter-side inductance are not ideal controlled current sources in actual situations,and the converter-side inductance will also affect the selection of the damping coefficient.Therefore,the converter-side inductance variable is introduced into the optimization function to improve the optimization function.For this reason,a thirdorder optimal damping coefficient control strategy is proposed,which takes into account the current control delay and the inductance change on both sides.The damping effect and dynamic response are still good when the inductance changes on both sides.Simulation and experimental results present the proposed method is correct.