| 引用本文: | 陆晓依,罗 建,姚志良,曾伟章.基于可观测量的同步发电机模型参数的频域辨识[J].电力自动化设备,2016,36(8): |
| LU Xiaoyi,LUO Jian,YAO Zhiliang,ZENG Weizhang.Frequency-domain parameter identification based on measurable variables for synchronous generator model[J].Electric Power Automation Equipment,2016,36(8): |
|
| 摘要: |
| 对含有不可观测量的同步发电机模型基本参数进行辨识，需要求解复杂的微分方程组，增加了辨识难度。提出一种由可观测量表示的同步发电机阻抗矩阵传递函数模型，简化了参数辨识方法，减小了辨识的计算量。利用阻抗实部和虚部分开表征的辨识算法进行模型的可辨识性分析，研究表明，结合稳态方程后，所提模型的基本参数是唯一可辨识的，避免了参数多值性问题，且辨识过程与参数经验值无关，能有效防止出现由参数经验值误差引起辨识精度降低的问题。通过自适应滤波获得信号的频域信息，结合粒子群优化算法辨识得到同步发电机基本参数。算例仿真结果验证了所提模型的正确性和辨识算法的有效性。 |
| 关键词: 同步发电机 参数辨识 自适应滤波 频域辨识 粒子群优化算法 |
| DOI： |
| 分类号: |
| 基金项目: |
|
| Frequency-domain parameter identification based on measurable variables for synchronous generator model |
|
LU Xiaoyi1, LUO Jian1, YAO Zhiliang2, ZENG Weizhang2
|
|
1.State Key Laboratory of Power Transmission Equipment & System Security and New Technology，Chongqing 400040，China;2.Guangdong Power Grid Dongguan Power Supply Bureau，Dongguan 523000，China
|
| Abstract: |
| The basic parameter identification of synchronous generator model with immeasurable variables has to solve complex differential equations，which increases the difficulty of identification. An impedance-matrix transfer function model of synchronous generator is proposed，which is only represented by the measurable variables to simplify the parameter identification and reduce the computational load. The identifiability of the model is analyzed by the identification algorithm represented separately by the real part and imaginary part of impedance. Research shows that，the basic parameters of the proposed model are uniquely identifiable according to the overdetermined equations，avoiding the multi-valuedness of parameter. Since the proposed parameter identification method does not need any empirical value of parameter，the decrease of identification accuracy due to the inaccurate empirical values can be effectively avoided. The frequency-domain information of signals is abstracted by adaptive filtering and the basic parameters of synchronous generator are identified by the particle swarm optimization algorithm. The correctness of the proposed model and the effectiveness of the identification algorithm are verified by case simulation. |
| Key words: synchronous generators parameter identification adaptive filtering frequency-domain identification particle swarm optimization algorithm |
