基于果蝇算法与自适应性遗传算法组合优化神经网络的光伏电站短期出力预测

光伏电站的输出功率会随着很多因素发生波动，若能够提高光伏系统出力预测的准确性，则能有效地降低光伏电站并网后对电网造成的冲击，提高电力系统的稳定性。建立了果蝇算法与自适应遗传算法组合优化的BP神经网络的预测模型。从预测结果可以发现，采用组合优化算法的BP神经网络模型能够有效避免地BP神经网络易陷入局部极小值点的缺陷，相比于仅优化权值和阈值的BP神经网络模型提高了预测精度，具有一定的应用价值。

Short-Term Output Prediction of Photovoltaic Power Stations Based on Combined Optimized Neural Network of FOA Algorithm and AGA Algorithm

The output power of the photovoltaic power station fluctuates due to many factors. If the accuracy of the output prediction of photovoltaic system can be improved, impacts of the grid-connected photovoltaic power station on power grids can be effectively reduced and stability of power system be improved. In this paper, the prediction model of the BP neural network based on the combination of fly (FOA) algorithm and adaptive genetic algorithm (AGA) is established. From the predicted results, it is found that the BP neural network model with combined optimization algorithm can effectively avoid the fault that the BP neural network is prone to fall into the local minimum point. Compared with the BP neural network model which only optimizes weights and thresholds, it is found that the new model improves the prediction accuracy and has better prediction performance and certain application value.