Department of Economics and Management, North China Electric Power University, 689 Huadian Road, Baoding, 071000, China.
Department of Computer, North China Electric Power University, 689 Huadian Road, Baoding, 071000, China.
Environ Sci Pollut Res Int. 2023 Jan;30(5):11689-11705. doi: 10.1007/s11356-022-22959-0. Epub 2022 Sep 13.
The randomness and instability of wind power bring challenges to power grid dispatching. Accurate prediction of wind power is significant to ensure the stable development of power grid. In this paper, a new ultra-short-term wind power forecasting model based on complete ensemble empirical mode decomposition with adaptive noise (CEEMDAN) and long-short-term memory (LSTM) network optimized by improved whale optimization algorithm (IWOA) is proposed. Firstly, CEEMDAN is applied to decompose the power history data into several intrinsic mode functions (IMFs) and a residual (RS) to reduce the complexity and unsteadiness of the original data. Then the partial autocorrelation method is used to analyze and select the input variables of each IMF and the residual. Finally, the IWOA-LSTM prediction model is established, and the parameters of LSTM are optimized by using the improved whale optimization algorithm. Each IMF and the residual are predicted respectively. The prediction results are superimposed to obtain the final wind power prediction value. The hybrid model is applied to the ultra-short-term wind power prediction of a wind farm in northern China. The prediction results of comparison experiments with other 11 models prove the effectiveness of the proposed model.
风力的随机性和不稳定性给电网调度带来了挑战。准确预测风力对于确保电网的稳定发展至关重要。本文提出了一种基于完全集合经验模态分解自适应噪声（CEEMDAN）和长短期记忆（LSTM）网络优化的改进鲸鱼优化算法（IWOA）的超短期风力预测新模型。首先，CEEMDAN 被应用于将功率历史数据分解为几个固有模态函数（IMF）和一个残差（RS），以降低原始数据的复杂性和不稳定性。然后，使用偏自相关方法来分析和选择每个 IMF 和残差的输入变量。最后，建立了 IWOA-LSTM 预测模型，并使用改进的鲸鱼优化算法对 LSTM 的参数进行优化。分别对每个 IMF 和残差进行预测，将预测结果叠加得到最终的风力预测值。该混合模型应用于中国北方一个风电场的超短期风力预测中，与其他 11 个模型的对比实验预测结果证明了所提出模型的有效性。
