基于数字孪生的光伏发电功率超短期预测

Ultra-short-term Prediction of Photovoltaic Power Generation Based on Digital Twins

光伏发电功率超短期预测对减小光伏并网对电网冲击及维持电网安全运行具有重要意义。提出一种基于数字孪生的光伏发电功率超短期预测机制,通过构建数字孪生体进行实时、高精度的光伏功率预测。首先根据GA-BP神经网络(genetic algorithm-back propagation neural network)构建光伏发电功率预测虚拟模型,并通过多维度的传感器采集光伏电池以及周围环境的各项孪生数据,同时更新历史数据库。然后以采集到的孪生数据为基础进行功率预测并得到初步预测结果。最后通过相似气象搜索,得到相似情况下的实际功率值和当时的预测功率,进而修正初步预测结果,得到最终预测功率。仿真算例结果表明,所提方法能有效提高光伏发电输出功率超短期预测精度。

Ultra-short-term prediction of photovoltaic power generation is of great significance to reduce the impact of grid-connected PV System on the power grid and maintain the safe operation of the power grid. In this paper, an ultra-short-term prediction mechanism of photovoltaic power generation based on digital twin is proposed. By constructing digital twins, the real-time and high-precision photovoltaic power prediction is carried out. Firstly, a virtual model of photovoltaic power prediction is established based on the GA-BP neural network, and the twin data of the photovoltaic cells and the surrounding environment are collected with the multi-dimensional sensors, and the historical database is updated. Then, based on the collected twin data, the preliminary prediction results are obtained. Finally, through the similar weather condition searching, the actual power value and the predicted power values at those times under the similar conditions are found, and then the preliminary prediction results are modified to obtain the final predicted power. Experimental results show that the proposed mechanism can effectively improve the ultra-short-term prediction accuracy of the photovoltaic power output.