Research On Modeling And Control Method Of Permanent Magnet Synchronous Generator Coupling Syste

With the increasing proportion of wind power generation in the power grid,the power sector has put forward higher requirements for the stability,transmission quality and efficiency of wind power systems,and the control problems of wind power systems have become more complex.At the same time,the diversification of power generation methods,the randomness of wind resources,the uncertainty of loads,manufacturing errors,and sudden current and mechanical shocks bring great challenges to the grid-connected control,robust control and stability control of wind power generation systems.In this paper,modeling and control methods are studied for the stability control of permanent magnet synchronous generators(PMSGs)coupling system for grid connected power generation.The main contents are as follows:(1)Considering the impact of impulse current and harmonic of generator units on the whole wind farm and the real-time adjustment of output voltage of each generator unit by the voltage feedback signal on the transmission bus of the wind farm,the differential equation of the PMSGs coupling system is established based on Kirchhoff’s law and Lenz’s law,and the mathematical model of the PMSGs coupling system is obtained through dimensionless transformation.Then,the nonlinear dynamic characteristics affecting the stable and efficient operation of the coupled power generation system are revealed by nonlinear analysis tools such as phase diagrams,time history diagrams and Lyapunov exponent diagram.(2)An event triggered adaptive backstepping control scheme is proposed for the PMSGs coupling system.A distribution function is designed to simplify the interval type-2fuzzy neural network,which greatly reduces the amount of calculation of the neural network.Combining the fixed threshold event triggering mechanism and the relative threshold event triggering mechanism,a switching threshold event triggering method is designed to ensure the control accuracy when the control signal is small and limit tracking errors when the control signal is large.In the framework of backstepping technology,the whole controller integrates simplified interval type-2 fuzzy neural network,Nussbaum type function,improved saturation function reaching law,cosine obstacle function,event triggering strategy and second-order tracking differentiator.Then,the stability of this scheme is proved by the Lyapunov function.Finally,the simulation results verify the feasibility of the scheme.(3)An accelerated adaptive neural network backstepping control scheme is designed for the PMSGs coupling system.Considering the computational complexity of the bottom layer of fuzzy wavelet neural network,the pseudo exponential function is used as the fuzzy membership function to avoid complex exponential operation.The limits of working speed and transmission efficiency are transformed into state constraints through tangent barrier function to ensure the boundedness of state variables.The steady-state and transient performance of the system is constrained by the speed function.The second-order tracking differentiator is used to deal with the explosion of complex terms in the process of backstepping design.The whole control scheme combines tangent obstacle function,velocity function,fuzzy wavelet neural network and second-order TD based on backstepping technology.Then,the stability of the scheme is proved by the Lyapunov function.Finally,simulation experiments verify the robustness of the scheme.(4)The event triggered adaptive backstepping control scheme and accelerated adaptive neural network backstepping control scheme designed above are further expanded and deepened,and applied to the triaxial MEMS gyroscope system.A prescribed performance accelerated adaptive neural network control method for triaxial MEMS gyroscope is proposed.In order to restrict the transient and steady-state performance of tracking errors,a quadratic prescribed performance function is designed.The function does not involve exponential operation and the bottom calculation is simple.The pseudo exponential function is used to improve type 2 fuzzy wavelet neural network to reduce the computational complexity of neural network.In the framework of backstepping strategy,the whole controller integrates speed function,quadratic prescribed performance function,improved type 2 fuzzy wavelet neural network and second-order tracking differentiator.Then,the stability of the controller is proved by the Lyapunov function.Finally,simulation experiments verify the effectiveness of the controller.