Abstract The drive train and generator parameters of doubly-fed induction generator (DFIG) may vary with the operation situation, which is significant for the stability of the DFIG. Previously used identification methods that have more measurements and experience difficulty in operation can easily fall into local optimum, and furthermore, there have been no studies on variable parameters. Therefore, in this study, we propose a novel identification method based on the improved particle swarm optimization (PSO) algorithm and trajectory sensitivity analysis. Initially, the equivalent model of DFIG was established. Selecting active power as a measurement based on trajectory sensitivity analysis can improve the accuracy of the identification method. Thereafter, an improved PSO algorithm compared with genetic algorithm (GA), PSO, and GA-PSO are verified by the classic test functions. Finally, a novel identification method is presented, which only requires one measurement and with no data processing of measurements. This method is suited for both constant and variable parameters. The results of parameter identification demonstrate that the proposed method can improve the accuracy and adaptability of the variable parameters. Besides, the proposed method can be used not only in doubly-fed wind turbines but also in other studies of new energy power generation.

中文翻译：

---

基于轨迹灵敏度分析的双馈感应发电机改进辨识方法

摘要 双馈感应发电机(DFIG)的传动系统和发电机参数可能随运行情况而变化，这对双馈感应发电机的稳定性具有重要意义。以前使用的识别方法测量量大，操作难度大，容易陷入局部最优，而且也没有对可变参数的研究。因此，在本研究中，我们提出了一种基于改进粒子群优化（PSO）算法和轨迹敏感性分析的新识别方法。初步建立了DFIG的等效模型。选择有功功率作为基于轨迹灵敏度分析的测量可以提高识别方法的准确性。此后，一种改进的 PSO 算法与遗传算法（GA）相比，PSO，和 GA-PSO 通过经典测试函数验证。最后，提出了一种新的识别方法，该方法只需要一次测量，无需对测量进行数据处理。此方法适用于常量和可变参数。参数辨识结果表明，该方法能够提高变参数的准确性和适应性。此外，所提出的方法不仅可以用于双馈风力涡轮机，还可以用于其他新能源发电的研究。参数辨识结果表明，该方法能够提高变参数的准确性和适应性。此外，所提出的方法不仅可以用于双馈风力涡轮机，还可以用于其他新能源发电的研究。参数辨识结果表明，该方法能够提高变参数的准确性和适应性。此外，所提出的方法不仅可以用于双馈风力涡轮机，还可以用于其他新能源发电的研究。