基于相似日和回声状态网络的光伏发电功率预测

光伏发电功率预测对提高光伏电站控制、调度性能以及保证电网的安全稳定运行具有重要意义。提出一种基于相似日和回声状态网络（ESN）的光伏发电功率预测模型。首先运用相关性分析法对光伏发电功率的影响因素进行了深入分析，并筛选出其主要影响因素；再利用主要影响因素的历史气象信息建立气象特征向量，通过计算灰色关联度（GRA）寻找合适的相似日；最后运用ESN创建预测模型，利用相似日历史数据训练ESN，而后对预测日的输出功率进行逐时预测。算例表明，对比传统模型，GRA-ESN模型具有更高的预测精度和更好的可行性。

Photovoltaic Power Prediction Based on Similar Day and Echo State Networks

Photovoltaic power generation forecasting is of great significance for improving the control and dispatching performance of photovoltaic power plants and ensuring the safe and stable operation of power grid. A photovoltaic power generation prediction model is proposed based on Echo State Networks（ESN）. Firstly，the correlation analysis method is used to analyze the influencing factors of photovoltaic power generation， and the main influencing factors are selected. The meteorological feature vector is established by using the historical meteorological information of the main influencing factors， and the appropriate similar day is found by calculating the gray correlation degree. Finally，ESN is used to create a predictive model， and the similar day historical data is used to train the echo state network，and then the output power of the forecast day is predicted on a time-by-time basis. The empirical analysis shows that the GRA-ESN model has higher prediction accuracy and better feasibility than the traditional prediction model.