基于模型的并网逆变器早期故障参数辨识

Identification of Early Fault Parameters for Grid-Side Inverters Based on the Model

以三电平T型逆变器为研究对象,研究了并网逆变器系统早期故障下的退化参数辨识方法.首先,基于键合图理论(bond graph,BG)结合逆变器系统早期故障下的参数退化特征建立了系统的混合键合图模型(hybrid bond graph,HBG);然后采用遍历路径法推导出了符合能量守恒的全局解析冗余关系式(global analytical redundancy relationships,GARRs),由此建立起系统的数学模型,并基于该系统行为约束方程将故障参数的辨识问题转化为对目标函数的优化问题;最后,通过麻雀搜索算法(sparrow search algorithm,SSA)对所构造出的函数进行参数寻优,并在20-sim和Matlab的联合仿真下,实现了对已发生参数退化情况下的逆变器系统的故障参数辨识.通过对辨识结果的分析,验证了基于GARRs和SSA的方法在并网逆变器早期故障参数辨识方面的有效性和可行性.

The identification method of degraded parameters under the early failure for grid-side inverter system is studied.The three-level T-inverter is taken as the object and firstly the model of hybrid bonding graph(HBG)is constructed based on the bond graph(BG)theory combined with the parameter degradation characteristics of the early failure for the inverter system.Then,the global analytical redundancy relationship(GARRs)which is conforming to energy conservation is deduced by traversal path method.Thus,the mathematical model of the system is established,and the problem of fault parameters is transformed into the optimization problem of the target function based on the behavior constraint equation of the system.Finally,the constructed function is found by Sparrow Search Algorithm(SSA)and the identification of fault parameter for the inverter system in the case of parameter degradation is realized under the joint simulation of 20-sim and MATLAB.The results show the validity and feasibility of the GARRs and SSA methods in the identification of early fault parameters for the grid-side inverter.