风电变流器DC-Link电容器准在线状态监测方法

Quasi-online Condition Monitoring Method of DC-Link Capacitor of Wind Power Converter

DC-Link电容是风电变流器中最核心但又最易失效的元件之一,对其进行状态监测可有效提高系统运行的可靠性。该文提出一种风电变流器DC-Link电容器准在线状态监测方法。首先,利用预充电阶段的交流电流重构DC-Link电容器的直流充电电流;采用二阶指数函数对电容器在预充电阶段的电压和电流波形进行精确拟合;取拟合函数的参数作为特征值,将实测的电容器状态参数(C和RESR)作为特征值对应的输出,并通过实验获得的样本数据,构建特征值与其输出的映射关系;最后利用多输出最小二乘支持向量回归算法对数据集合进行学习训练,建立特征值与电容器状态参数的回归模型。该方法无需增加额外传感器和改变变流器控制算法。实验结果验证了本方法的有效性。

DC-Link capacitor is one of the most important but vulnerable devices of wind power converters,its condition monitoring can improve system reliability.This paper proposes a quasi-online condition monitoring method for DC-Link capacitor of wind power converter.Firstly,DC charging current of DC-Link capacitor is reconstructed with AC-side current during pre-charging process.The voltage and current waveforms of capacitor during pre-charging stage are accurately fitted with second-order exponential functions.The parameters of fitting functions are taken as eigenvalues,measured condition parameters(C and RESR)are used as corresponding output of eigenvalues.Then mapping relation between eigenvalues and the output is build with samples obtained by experiments.Finally,multi-output least-squares support vector regression machines(MLSSVR)algorithm is adopted to perform learning and training with the data set,and to establish regression model between eigenvalues and condition parameters of capacitor.The proposed method does not rely on extra hardware circuit and modifying existing control algorithm of converters.The experiment results could verify the validity of the method in this paper.