一种光伏系统短期功率预测模型

Short-term power forecasting model of photovoltaic system

为提高光伏系统发电功率预测精度,优化系统的发电计划,减少电力系统运行成本,进而为系统调度和实时运行控制提供依据以有效减轻光伏发电系统接入对电网的影响,建立一种基于三层神经网络和功率波动特性的短期光伏出力预测模型。利用气象局已发布的日类型和温度信息挑选与预测日最相关的相似日,基于神经网络用相似日历史太阳辐照、温度、输出功率建立光伏系统出力初步预测模型;以预测日天气预报信息作为神经网络的输入获得预测日的功率预测值;基于由光伏系统相似日历史出力数据统计分析得到的波动量统计规律对初步预测结果加以修正,建立了具有较高精度的光伏系统出力预测模型。仿真结果表明该方法建立的预测模型具有较高精度,能够为调度运行人员提供决策辅助。

To improve the prediction accuracy of the photovoltaic power generation system, optimize the system＇s power generation plans and reduce the operating costs of power system, provide the basis for real-time scheduling and run-time control to effectively mitigate the impact on photovoltaic power generation system while it accesses the grid, a short-term forecasting model based on three -layer neural network and fluctuation characteristics of photovoltaic power was set up. Firstiy, the information of day type and temperature which was released by Bureau of Meteorology was used to pick the similar day which was most relevant to the prediction day, and then the similar days＇ previous solar irradiance, temperature, output power output of the PV system was used to establish a preliminary prediction model based on neural network. Secondly, the predicted day＇s weather forecast information as input of neural network was put to obtain the preliminary output power of the predicted day. Lastly, the fluctuation statistics law was got through counting and analyzing the similar day＇s historical output data, then the preliminary predictions was amended by the law, and a PV system output forecast model with higher precision was established. The simulation results show that the prediction model established by this method has higher accuracy, thus it can provide decision support for dispatchers.