Research on Power Regulation Characteristics and Optimisation Strategies for Photovoltaic Power Stations under Low Irradiance Conditions Based on Operational Data Mining

Abstract

Addressing issues such as inadequate power regulation capability and reduced generation efficiency in photovoltaic power stations under low irradiance conditions, this study proposes a performance optimisation method based on operational data mining. Utilising actual operational data from a 50kW photovoltaic power station, an improved variational modal decomposition and fuzzy C-means clustering method were employed to analyse the system's power regulation characteristics under low irradiance conditions. Integrating NASA meteorological data, a system performance evaluation model accounting for multiple environmental factors was established, alongside an adaptive optimisation control strategy. Research indicates: the system exhibits pronounced power regulation lag under low irradiance, with response times extended by 42% compared to normal conditions. The proposed optimisation control strategy enhances power generation efficiency by 5.8% and reduces power fluctuations by 37.2% under low irradiance. [Conclusion] The proposed method effectively improves the operational performance of photovoltaic systems under low irradiance, providing a novel technical pathway to enhance power stations' all-weather power generation capabilities.