Rotational speed sensor is one of the most important components used in wind energy conversion system control strategies. The latter is very expensive and may be faulty due to the harsh environment working, causing by that a low efficiency operation of the system. The use of a speed estimator or an observer may be a good alternative solution for the use either in sensorless control or detecting the sensor failure or degradation (FDI and FTC), provided that the observer is robust and ensures high rotational speed estimation accuracy. This paper presents a comprehensive study to solve the optimization problem of a model reference adaptive speed (MRAS) observer in a typical wind energy generation system based on permanent magnet synchronous generator using five adaptation mechanisms: The first one is the classical MRAS observer; it is based on proportional–integral (PI) controller. The second one uses a fuzzy logic controller (FLC) with two inputs. The third one uses the single-input fuzzy logic controller which represents the simplification of the conventional FLC. The fourth one uses the sliding mode controller, and the last one uses the super-twisting algorithm. A detailed comparison between the five adaptation mechanisms is carried out. The obtained results show a good estimation stability as well as a fast speed estimation at dynamic regime for the improved adaptation mechanisms compared with the conventional PI controller.

中文翻译：

---

改进风能转换系统模型参考自适应速度观测器的自适应机制技术

转速传感器是风能转换系统控制策略中最重要的部件之一。后者非常昂贵，并且可能由于恶劣的工作环境而出现故障，从而导致系统的低效率运行。速度估计器或观测器的使用可能是用于无传感器控制或检测传感器故障或退化（FDI 和 FTC）的良好替代解决方案，前提是观测器是稳健的并确保高转速估计精度。本文综合研究解决基于永磁同步发电机的典型风力发电系统中模型参考自适应速度 (MRAS) 观测器的优化问题，使用五种自适应机制：第一个是经典 MRAS 观测器；它基于比例积分 (PI) 控制器。第二个使用具有两个输入的模糊逻辑控制器 (FLC)。第三种使用单输入模糊逻辑控制器，它代表了传统 FLC 的简化。第四个使用滑模控制器，最后一个使用超扭曲算法。对五种适应机制进行了详细比较。获得的结果表明，与传统的 PI 控制器相比，改进的自适应机制在动态状态下具有良好的估计稳定性和快速估计。第四个使用滑模控制器，最后一个使用超扭曲算法。对五种适应机制进行了详细比较。获得的结果表明，与传统的 PI 控制器相比，改进的自适应机制在动态状态下具有良好的估计稳定性和快速估计。第四个使用滑模控制器，最后一个使用超扭曲算法。对五种适应机制进行了详细比较。获得的结果表明，与传统的 PI 控制器相比，改进的自适应机制在动态状态下具有良好的估计稳定性和快速估计。