Double closed-loop control method for hydraulic loading systems in Stewart-platform-based wind turbine drivetrain test benches
Jianjun Yan, Zeyu Yu, Yunfei Zhang, Yufan Lu, Yanxing Huang, Yongming ZhangPurpose
As wind turbine capacities increase, wind turbine drivetrain test benches face challenges due to loading inaccuracies caused by coupling effects in parallel loading devices and assembly errors. This paper proposes a new control method for a non-torque hydraulic loading device based on the Stewart platform to enhance loading accuracy and response speed in multi-megawatt wind turbine testing.
Design/methodology/approach
A double closed-loop control strategy with a fuzzy PID controller is implemented to address these issues. A co-simulation model combining MATLAB, AMESim, and ADAMS software is developed to simulate wind loading tests. Unidirectional and multi-degree-of-freedom loading experiments are conducted to evaluate the effectiveness of the proposed method.
Findings
In unidirectional loading experiments, the settling time ranged from 0.06s to 0.14s, with a maximum overshoot of 1.9%. Significant steady-state errors were observed for F_z and M_y, reaching 11.6 and 8.2%, respectively. Multi-degree-of-freedom loading tests revealed longer settling times and higher errors due to coupling effects. The proposed control method effectively suppressed these coupling effects and reduced the steady-state error significantly compared with the conventional PID controller, thereby ensuring accurate load reproduction and validating the effectiveness of the fuzzy PID controller and the double closed-loop control approach.
Originality/value
This paper proposes a novel control method for a non-torque hydraulic loading device based on the Stewart platform, addressing challenges in multi-megawatt wind turbine drivetrain testing. The method integrates a double closed-loop control strategy with a fuzzy PID controller to improve load accuracy, response speed, and mitigate coupling effects. This approach enhances hydraulic loading system performance, contributing to wind turbine testing and advancing control technologies for complex multi-degree-of-freedom loading scenarios.