Fault Diagnosis And Parameter Identification For Power Electronicconverters

Nowadays,power electronic converters have been widely used in various fields,and its reliability and safety issues have attracted the attention of experts and scholars.As the basis of electrical energy transformation,power electronic devices work in a high power state for a long time,so the probability of failure is large.In the event of a power electronic device fault,it can cause unpredictable effects,leading to catastrophic results.Therefore,timely detection of power electronic converter components faults,to ensure its reliability and safety has far-reaching significance.The main research content of this paper includes three aspects.Firstly,for the single-phase transformerless inverter,analyze the switching characteristics of the inverter link,analyze the relationship of voltage and current,and construct the system switched model.Exploring the effects of component faults on inverters and constructing the fault model for the system.Based on the fault model,a Luenberger observer with an unknown parameter is designed.Robust analysis is performed on the obtained error system,and a linear matrix inequality is obtained by theoretical derivation.The observer gain matrix is obtained by solving it.When a fault is detected,the fault identification is activated,the faulty component is identified,and the parameters of the component are estimated.Secondly,the state space modeling method is used to establish the switched system model of the Boost converter in boost link according to Kirchhoff's voltage-current law,and the fault model and stability of the system are analyzed.An observer-based method is proposed to achieve fault detection.Once the fault is detected,the Luenberger fault identification observers are activated.Using the parameter identification technology,the parameters of the established switched system model are estimated.By comparing the estimated parameters and the ideal parameters,the specific location of the fault is confirmed.Finally,the switching characteristics of power electronic devices in the process of electrical energy transformation are analyzed.Based on the topology of the circuit,a general modeling method that can fully characterize the switching characteristics in power electronic converters is constructed.When the system switching characteristics and faults are combined,they all affect the converter output.Therefore,based on different fault types and switching characteristics,a model-based fault detection method is established,and a robust sliding mode observer is designed to realize the fault detection.The adaptive method is used to identify the fault,the fault identification observers with sliding surface and the appropriate adaptive law are designed to accurately identify the faulty component and identify the component parameters after the fault.Finally,the solar photovoltaic Boost converter is taken as the specific research object,and the theoretical derivation is carried out.The feasibility of the proposed method is verified by simulation results.