考虑光伏机理与数据驱动结合的短期功率预测

精确的光伏功率预测对电网的可靠与稳定运行至关重要。现有研究大多数都是将天气条件直接作为数据驱动的输入，未深入分析天气条件与光伏输出功率直接耦合关系，因此本文将机理模型与数据驱动方法相结合，提出一种新型的光伏功率预测方法。首先，建立光伏系统物理模型，依据建立的物理模型得到不同的辐照度分量以及光伏电池温度。其次，将这些关键的天气特征重新构建数据驱动的输入，实现光伏机理与数据驱动结合的短期功率预测。最后，进行误差修正然后得到最终的光伏功率预测结果。根据光伏系统实测数据集进行仿真分析，结果表明因为从物理模型得到了关键天气特征，考虑了天气条件与天气因素的耦合关系，预测精度有了明显提升，验证了所提方法的有效性。

Consider short-term power prediction combining photovoltaic mechanism and data-driven

Accurate PV power forecasting is essential for the reliable and stable operation of the grid. Most of the existing studies take weather conditions directly as data-driven input, ignoring the direct coupling relationship between weather conditions and photovoltaic output power, so this paper combines the mechanism model with the data-driven method to propose a new photovoltaic power prediction method. First, establish a physical model of the photovoltaic system; According to the established physical model, different irradiance components and photovoltaic cell temperature are obtained, and secondly, these key weather characteristics are reconstructed data-driven inputs to realize short-term power prediction combining photovoltaic mechanism and data-driven, and finally, error correction is carried out to obtain the final photovoltaic power prediction results. According to the measured data set of photovoltaic system, the simulation analysis results show that because the key weather characteristics are obtained from the physical model and the coupling relationship between weather conditions and weather factors are considered, the prediction accuracy is significantly improved, and the effectiveness of the proposed method is verified.