研究目的
To solve the problem of distinguishing two groups of similar faults in the open-circuit faults of power switching devices in cascaded H-bridge multilevel inverters for grid-connected photovoltaic systems.
研究成果
The proposed secondary classification fault diagnosis strategy based on PCA-SVM effectively distinguishes similar faults in cascaded PV grid-connected inverters, achieving higher accuracy compared to other methods. It shows good performance under different external conditions, offering a new approach to handling similar faults.
研究不足
The study focuses on open-circuit faults of IGBTs, excluding short-circuit faults due to their short duration and difficulty in real-time detection. The strategy's effectiveness under varying external conditions is demonstrated, but further optimization may be needed for other fault types or conditions.
1：Experimental Design and Method Selection:
A secondary classification fault diagnosis strategy based on PCA-SVM is proposed. The first classification makes a preliminary fault diagnosis between all types of faults, and the second classification distinguishes between two groups of similar faults.
2：Sample Selection and Data Sources:
The output voltage of the inverter is used as fault signature, collected under different external conditions.
3：List of Experimental Equipment and Materials:
MATLAB/SIMULINK tool for modeling the system, with specific parameters for PV array output voltage, filter inductance, resistance, and public grid voltage frequency.
4：Experimental Procedures and Operational Workflow:
Data acquisition involves setting the switching frequency of the inverter, sampling frequency, and collecting data after the system is stable.
5：Data Analysis Methods:
PCA for feature extraction and SVM for fault classification, with comparisons to PCA-ELM and PCA-BP methods.
独家科研数据包，助您复现前沿成果，加速创新突破
获取完整内容
