风能具有较强的波动性和随机性,风电功率也具有同样的特点。因此,本文提出了改进极限学习模型来提高风能的利用率。首先,本文提出了通过乌鸦搜索算法来优化参数,通过收敛性能的测试,乌鸦搜索算法在全局搜索能力以及局部的开发能力上具有较大的优势。其次,确定模型函数的输入输出变量,然后分别用多个评估性指标与其他模型进行比较。最后,对改进极限学习机模型的预测结果进行了评估和分析。预测结果表明改进极限学习模型的预测精度和稳定性均高于其他模型。

Wind energy has strong volatility and randomness, and wind power also has the same characteristics. Therefore, this study proposes an improved limit learning model to improve the utilization of wind energy. Firstly, this paper proposes to optimize the parameters by crow search algorithm, and through the test of convergence performance, crow search algorithm has a greater advantage in global search ability as well as local exploitation ability. Secondly, the input and output variables of the model function are determined and then compared with other models using several evaluative metrics respectively. Finally, the prediction results of the improved extreme learning machine model are evaluated and analyzed. The prediction results show that the prediction accuracy and stability of the Improved Extreme Learning Model are higher than other models.