Investigation On Parameter Identification Technology Of Grid-connected Operation Control System Based On Doubly-fed Induction Generators

As the energy crisis,environmental pollution and climate issues become increasingly prominent,new energy power generation has received widespread attention.In recent years,a large number of new energy power stations represented by wind power have been connected to the grid,and the proportion of new energy power generation in the power system has increased significantly,and its impact on the power system cannot be ignored.Doubly-fed induction generators(DFIG)are the main models put into use in wind farms.In order to clarify the impact of DFIG on the grid,it is extremely important to accurately model their grid-connected operation control systems.However,in actual wind farms,due to the confidentiality of the manufacturer or changes in operating conditions,due to confidentiality of the converter manufacturer or changes in operating conditions,some control parameters and even control structures of DFIG are unknown,and the operation control system is characterized by gray-box or black-box.The traditional modeling method relies on the accurate parameters and structure of the system,and cannot meet the modeling requirements of gray-box or black-box systems.The parameter identification does not rely on the internal information of the system,and can determine the accurate value of the internal parameters of the control system based on its external information,such as the output power,and can solve the modeling dilemma of the gray-box or black-box system.In view of the current status of research on parameter identification of grid-connected operation control system for DFIG,the existing identification method have higher accuracy in identifying the outer-loop control parameters of the vector control(VC)strategy,but the accuracy of the inner-loop control parameter identification is significantly lower than that of the outer-loop.The high-precision identification of the control parameters of the inner and outer loops cannot be achieved at the same time.In addition,most of the existing parameter identification studies are aimed at the steady-state operating conditions of the unit,and lack of consideration of the unit’s control mode switching under the power grid failure and the influence on the control parameter identification.And it is limited to meet the parameter identification requirements of a gray-box system with a certain control strategy,and does not consider the situation where multiple control strategies may be applied to grid-connected operation control,and it is difficult to meet the parameter identification requirements of a black-box system with unknown control strategies.According to the steady-state operation and fault ride through conditions of DFIG,this paper considers two different control strategies to study the parameter identification of the gridconnected operation control system of the units.The specific research content and innovation results are as follows:1.Aiming at the identification of steady-state operation control parameters of DFIG under VC the identification accuracy of inner-loop parameters is much lower than that of outer-loop parameters,a parameter identification method based on improved damping least squares method is proposed.Firstly,a time domain mathematical model of the power step response of the gridside converter and the generator-side converter of a DFIG with a VC strategy is established.According to the mathematical model,the trajectory sensitivity of the control parameters of the inner-loop and the outer-loop with respect to the power response is analyzed.Then the traditional damping least squares method of damping factor adjustment method is improved,and based on the improved identification algorithm,combined with the sensitivity analysis results,the identification scheme of the VC inner-loop and outer-loop parameters under steady-state operation of the unit is studied.Two-stage identification reduces the identification error of the inner-loop control parameters.2.Aiming at the problem that the control mode switching of DFIG under VC during the low voltage ride through process makes the control parameter identification scheme no longer applicable under steady-state operating conditions,a parameter identification method based on particle swarm optimization is proposed.First,the control mode switching mechanism of the DFIG from the occurrence of grid voltage drop to the recovery process is analyzed.Furthermore,a mathematical model of the power response of the unit under the low voltage ride-through condition is established,and the nonlinear characteristics of the model due to the limiting link are analyzed.Then combined with the particle swarm optimization,a parameter identification scheme suitable for nonlinear models is studied.Through parameter trajectory sensitivity analysis,the identification accuracy of all control parameters is guaranteed.3.Aiming at the problem that VC and virtual synchronous generator(VSG)control strategies may be applied to DFIG grid-connected operation control,and when the control strategy is not clear,it is difficult to reflect the power response characteristics of the unit through control parameter identification.A method based on improved damping least squares is proposed.The power response model parameter identification scheme at the grid connection point of the unit.First,the VC and VSG control strategies are compared and analyzed,and the control structure and the function of the control parameters of the two and the setting principles are clarified.Then an equivalent model of the power response at the grid connection point of the unit is established.The built model can express the active power response at the grid-connected point under two different control strategies in a unified form,and meets the parameter identification requirements of the black-box system with unknown control strategies.Then combined with the improved damping factor adjustment method,the damping least square method,the identification scheme of the equivalent parameters of the built model is studied.Furthermore,the influence of generator parameter changes on the identification accuracy of equivalent parameters is analyzed,and the error evaluation of the identification results is realized based on the calculation of trajectory sensitivity.