Loading Force Control of Non Torque Loading Device on Wind Turbine Drivetrain Test Bench

The development of wind turbines towards largescale and deep-sea directions has put forward higher requirements for wind turbine testing technology. The drivetrain test bench has a six-degree-of-freedom load simulation capability and is an important development direction for wind turbine testing technology. The valve-controlled hydraulic cylinder system used for non-torque loading on the drivetrain has parameter uncertainty and nonlinear characteristics. To improve the tracking accuracy of the loading force on the experimental platform, a loading force control method is proposed. Firstly, based on the non-torque loading scheme and hydraulic cylinder distribution scheme proposed in this article, the non-torque loading equation and hydraulic cylinder mathematical model of the test platform were derived. Secondly, a sliding mode control strategy for the loading force control of the test platform was derived. In response to the modeling uncertainty of the dynamic model, an Radial Basis Function (RBF) neural network was used to approximate the unknown terms in the model. The weights in the neural network were adjusted online using an adaptive method, thus establishing an RBF neural network adaptive control for the loading force of the test platform. Finally, establish a simulation model for loading force control on the test bench for verification. The results indicate that this method has good control effect and can achieve loading force tracking in the presence of external interference.