Asynchronous self-triggered sliding mode control for wind turbine based on Markov jump model

Abstract

This work is concerned with the sliding mode control of a wind turbine driven by randomly-switching wind speeds, with the aim of adjusting the generator speed to acquire rated power while reducing the fatigue load of the wind turbine. Due to the stochastic nature of wind speed, the operating point of the wind turbine changes frequently. The stochastic characteristics of wind speed are described by a Markov process, so that the traditional operating point of the wind turbine is divided into separate modes accordingly, in which model parameters and control gains for each mode can be determined. Considering that the wind turbine status data is transmitted to the controller via a wireless communication network with limited bandwidth, a self-triggered mechanism is introduced to enhance channel resource utilization and reduce bandwidth occupancy, in which the triggering instants can be calculated by utilizing the current and past triggered state information. Meanwhile, a mode estimator is employed to estimate the unobtainable system mode. Then, an asynchronous self-triggered sliding mode controller is constructed, and the sufficient conditions are derived to achieve the stochastic stability of the system with a specified $H_{\infty }$ performance level. Finally, the simulation results of a 2MW wind turbine verify the feasibility and effectiveness of the present control strategy.