Research On Adaptive Predictive Control For Grid Connected Front-End Speed Regulated Wind Turbines

The development of transportation energy Internet provides a new solution for the consumption of new energy such as wind power and photovoltaic.While accelerating the informatization,gridding and diversification of transportation,it promotes the multi-scenario application of new energy,and promotes the deep integration and collaborative development of transportation,electric power and energy industries.As a kind of grid-friendly wind turbine,the front-end speed regulated wind turbine adopts hydraulic torque converter to adjust the speed of the generator,and uses electrically excited synchronous generator to directly couple with the grid,which solves the limitation of traditional wind turbines relying on converter for grid-connected control and speed regulation in structure.In this paper,through in-depth analysis of the structural characteristics of the front-end speed regulated wind turbine and many factors affecting its operation control,starting from the turbine modeling,aiming at the problem of poor grid-connected control performance of the wind turbine,the following innovative research work is carried out.(1)The basic operation principle of front-end v speed regulated wind turbine is analyzed in detail.According to the demand of grid-connected adaptive predictive control of wind turbine and based on the energy conversion process of wind turbine,the output power model of wind turbine,the dynamic model of transmission chain of wind turbine,the fifth order model of electrically excited synchronous generator and excitation system models are established,which provide the basis for grid-connected adaptive predictive control of wind turbine.(2)Aiming at the problem that the constant speed control of hydraulic torque converter of front-end speed regulated wind turbine is difficult to realize,a fuzzy controller based on variable universe is designed,and the parameters are optimized by using multi-population genetic algorithm.The intelligent optimization of universe expansion factor is realized,and the accuracy and speed of hydraulic speed control are improved.When the front-end speed regulated wind turbine is affected by wind speed fluctuation,grid side interference and other factors,the output speed of the generator can be kept within a certain error range,which ensures the constant speed operation of pump wheel of the hydraulic torque converter and input shaft of the generator,and provides a guarantee for the stability of the output voltage frequency of the unit.(3)Aiming at the problem of unstable output power of the front-end speed regulated wind turbine,the idea of multi-model predictive control is introduced into the output power control of the generator.Through the fuzzy clustering modeling of the measured operation data,the fuzzy C-means clustering and multi-model predictive control method are combined to determine various operation scenarios of the unit,and the corresponding model switching predictive controller is designed,which can effectively solve the problem of randomness and uncertainty in power control of grid-connected units,and thus improves the stability of output power.(4)Aiming at the problem of voltage fluctuation in the process of grid-connected operation,based on the idea of predictive control,a generalized adaptive predictive controller for generator grid-connected voltage control is designed.The generalized predictive control algorithm is combined with back propagation neural network to track and control the output voltage of the generator,so as to reduce the fluctuation of grid-connected voltage.Furthermore,a predictive controller based on multi-objective genetic algorithm is designed to achieve coordinated control of all subsystems of the wind turbine and ensure the low voltage ride through characteristics of the generator.(5)In order to clarify the stability of grid connected voltage,the differential algebraic equation of a grid-connected wind farm composed of front-end speed regulated wind turbines is established.Based on the nonlinear dynamics theory,the whole process of operation voltage from stability to instability and collapse after grid-connected is studied by using bifurcation theory,and the influence of wind speed and load on front-end variable-speed wind turbines is clarified.It is found that with the increase of reactive power on load-side,the voltage of load node will gradually decrease.When reactive power exceeds a certain value,saddle node bifurcation point will appear on the manifold of system equilibrium solution,and the generator will reach the limit state of operation.When wind speed is less than rated wind speed,its variation has little effect on the voltage at load node.When the wind speed exceeds14.8m/s,the system voltage will gradually decrease and become unstable.