Reinforcement-Based Robust Variable Pitch Control of Wind Turbines

Abstract

Due to the influence of wind speed disturbance, there are some uncertain phenomena in the parameters of the nonlinear wind turbine model with time in an actual working environment. In order to mitigate the side effects of uncertainties in speed models of wind turbines, researchers have designed a variety of controllers in recent years. However, traditional control methods require more knowledge of dynamics. Therefore, based on reinforcement learning and system state data, a robust wind turbine controller that adopts adaptive dynamic programming (ADP) is proposed. The ADP algorithm is a combination of Temporal-Difference (TD) algorithm and actor-critic structure, which can guarantee the rotor speed is stable around the rated value to indirectly adjust the wind energy utilization coefficient by changing the pitch angle in the area of high wind speed and achieve online learning in real-time. In addition, the variation of the pitch angle command of the proposed controller is relatively gradual, which can reduce the energy consumption of the variable pitch actuator, and extend the service life of the equipment. Finally, the wind speed model is simulated by combined wind speed based on Weibull distribution, the comprehensive simulation results show that the proposed controller has better control effect than some existing ones. INDEX TERMS Neural dynamic programming, reinforcement learning, robust control, wind turbine system.

An Effective Dense Co-Attention Networks for Visual Question Answering

“…Compared with the existing method MCAN, DCAN can make use of the complex correlation between multimodal features in a more effective way and extract more discriminative features for images and questions. This exploration of modeling dense intra- and inter-modality interactions has been applied to intelligent transportation [ 42 ], intelligent robot [ 43 ], and other fields [ 44 , 45 , 46 ]. Applying it to a wider range of scenarios will be an inevitable trend in the future.…”

Exploring Reward Strategies for Wind Turbine Pitch Control by Reinforcement Learning

“…In Hosseini et al, passive RL solved by particle swarm optimization policy (PSO-P) is used to handle an adaptive neuro-fuzzy inference system type-2 structure with unsupervised clustering for controlling the pitch angle of a real wind turbine [33]. Chen et al also propose a robust wind turbine controller that adopts adaptive dynamic programming based on RL and system state data [34]. In a related problem, deep reinforcement learning with knowledge assisted learning are applied to deal with the wake effect in a cooperative wind farm control [35].…”

A Control Scheme with the Variable-Speed Pitch System for Wind Turbines during a Zero-Voltage Ride Through

“…The process of comparing these proposed controllers with a well-tuned proportional integral (PI) controller is interesting, but the practicability and feasibility of such a replacement for operating WTs still need to be proven. Chen presented a robust controller that adopts adaptive dynamic programming based on reinforcement learning and system state data in [22]. The pitch variation commands from this controller are relatively gradual, which reduces the energy consumption of the pitch actuator.…”

“…Tang proposed an active power control strategy that integrated rotor speed and pitch angle regulation; this approach avoids the need for frequent pitch actuator actions while sustaining dispatched active power [23]. However, in practice, the reason given in [22,23] for reducing pitch actuator consumption may not be recommended. Timely pitching is the main requirement for WTs because that aspect is a basic safety rule; in contrast, the consumption of the pitch system is not a pressing issue for WT designers.…”