关键词: 光伏阵列, 数学建模, 参数辨识, 特征提取, 健康评估, 故障诊断

Abstract: This paper addresses the health operation and maintenance needs of PV arrays and systematically describes the current research status of PV modelling, parameter identification, feature extraction and health status evaluation method. First, the modelling application scenarios of equivalent circuit models are analysed to compare the advantages and disadvantages of Simulink-based circuit simulation, numerical simulation and neural network black-box models. Then, the parameter identification method focuses on the complementarity between the analytical method and the meta-heuristic algorithm and proposes that the combined method optimizes the iterative process by analysing the initial value, which can consider the computational speed and accuracy. Next, the differentiated characterization capabilities of statistical features, signal decomposition and deep learning are compared to emphasize the necessity of I-V curve transformation to enhance feature robustness. Finally, the five-level health state classification system based on PV energy efficiency performance ratio and fuzzy judgment is proposed to clarify the logical relationship between health assessment and fault diagnosis. Through the systematic sorting of the above existing technical systems, we provide theoretical support for efficient and accurate PV array condition monitoring and intelligent operation and maintenance program development.

Key words: Photovoltaic array, Mathematical modeling, Parameter identification, Feature extraction, Health assessment, Fault diagnosis