Adaptive Backstepping Control of the PMSG Based on the T2SFNN

In this paper, the permanent magnet synchronous generator (PMSG) is taken as a research object, which has nonlinear motions induced by chaotic oscillation and other factors. An adaptive control strategy combining the type-2 sequential fuzzy neural network (T2SFNN) is proposed to suppress chaos and ensure the smooth operation of the motor. Firstly, a mathematical model is established in the two-phase synchronous rotating coordinate system, and the chaotic motion phenomenon of the PMSG is revealed by constructing an integral operation circuit equivalent to the system state e quation. Then, based on the backstepping control, combined with cosine barrier function, nussbaum-type function and speed function, using a T2SFNN to approximate the unknown nonlinear term in the mathematical model of the PMSG, which can accelerate the convergence of tracking error, guarantee the boundedness of tracking error and output and make the nonlinear system have good transient and steady-state behaviors. Tracking differentiator is used to deal with the complex term explosion in the process of backstepping, so as to improve the accuracy of the control system. Furthermore, the Lyapunov function is used to predefined constraints are never violated and proved all closed-loop signals are ultimately bounded. Finally, the effectiveness and feasibility of the scheme a re verified by simulation.